From gf at unixsol.org  Thu Dec  1 02:32:44 2011
From: gf at unixsol.org (Georgi Chorbadzhiyski)
Date: Thu, 01 Dec 2011 03:32:44 +0200
Subject: [vlc-devel] [RFC,
 PATCH] configure: Do not stop configure if component is not found.
Message-ID: <4ED6D93C.5010008@unixsol.org>

I'm compiling vlc on a server with only minimal needed components
(libav, decklink, a52, x264) and without adding bunch of --disable-xxx
options configure stops with error on several occasions. IMHO if
component is not requested via --enable-xxx the configure should
mark it as not found and go on. The attached patch does this for
the components that I do not have but are generating errors.

If the approach is acceptable I'll try look for more of these
in configure.ac

Note: This is my first crack at autoconf so I may have made a mess
out of it.

-- 
Georgi Chorbadzhiyski
http://georgi.unixsol.org/
-------------- next part --------------
From 882688db5826446626f32a839424835e71babae2 Mon Sep 17 00:00:00 2001
From: Georgi Chorbadzhiyski <gf at unixsol.org>
Date: Thu, 1 Dec 2011 03:22:06 +0200
Subject: [PATCH] configure: Do not stop configure if component is not found.

Configure should not exit with error when dbus-control, lua,
a52, alsa and xcb are not found. It should just disable the
component.
---
 configure.ac |   39 +++++++++++++++++++++++++++++----------
 1 files changed, 29 insertions(+), 10 deletions(-)

diff --git a/configure.ac b/configure.ac
index 09766ac..d831e55 100644
--- a/configure.ac
+++ b/configure.ac
@@ -778,8 +778,8 @@ then
       VLC_ADD_CFLAGS([libvlccore],[$DBUS_CFLAGS])
       dnl Check for dbus control interface
         AC_ARG_ENABLE(dbus-control,
-          [AS_HELP_STRING([--disable-dbus-control],
-            [D-Bus control interface (default enabled)])])
+          [AS_HELP_STRING([--enable-dbus-control],
+            [D-Bus control interface (default auto)])])
         if test "${enable_dbus_control}" != "no"
         then
           VLC_ADD_PLUGIN([dbus])
@@ -799,7 +799,9 @@ then
         VLC_ADD_LIBS([inhibit],[$DBUS_LIBS])
         VLC_ADD_CFLAGS([inhibit],[$DBUS_CFLAGS])
     ],
+    [AS_IF([test "${enable_dbus_control}" = "yes"],
     [AC_MSG_ERROR([${DBUS_PKG_ERRORS}.])]
+    )]
   )
 fi
 ;;
@@ -1541,8 +1543,8 @@ AM_CONDITIONAL(ENABLE_SOUT, [test "${enable_sout}" != "no"])
 
 dnl Lua modules
 AC_ARG_ENABLE(lua,
-  [AS_HELP_STRING([--disable-lua],
-    [disable LUA scripting support (default enabled)])])
+  [AS_HELP_STRING([--enable-lua],
+    [enable LUA scripting support (default auto)])])
 if test "${enable_lua}" != "no"
 then
   PKG_CHECK_MODULES(LUA, lua5.1,
@@ -1573,12 +1575,16 @@ then
      VLC_ADD_LIBS([lua],[$LUA_LIBS])
      VLC_ADD_CFLAGS([lua],[$LUA_CFLAGS])
   else
+      AS_IF([test "${enable_lua}" = "yes"],[
       AC_MSG_ERROR([Could not find lua. Lua is needed for some interfaces (rc, telnet, http) as well as many other custom scripts. Use --disable-lua to ignore this error.])
+      ])
   fi
   AC_ARG_VAR([LUAC], [LUA byte compiler])
   AC_PATH_PROGS(LUAC,[${LUAC} luac], [false])
   AS_IF([test "${LUAC}" = "false"], [
+    AS_IF([test "${enable_lua}" = "yes"],[
     AC_MSG_ERROR([Could not find the LUA byte compiler.])
+    ])
   ])
 fi
 AM_CONDITIONAL(BUILD_LUA, [test "${have_lua}" = "yes"])
@@ -2639,7 +2645,7 @@ dnl
 dnl A52/AC3 decoder plugin
 dnl
 AC_ARG_ENABLE(a52,
-  [  --enable-a52            A/52 support with liba52 (default enabled)])
+  [  --enable-a52            A/52 support with liba52 (default auto)])
 if test "${enable_a52}" != "no"
 then
   AC_ARG_WITH(a52,
@@ -2698,10 +2704,14 @@ then
         VLC_ADD_LIBS([a52tofloat32],[${LDFLAGS_test} -la52])
         VLC_ADD_CPPFLAGS([a52tofloat32],[${CPPFLAGS_test}])
       ],[
+        AS_IF([test "${enable_a52}" = "yes"],[
         AC_MSG_ERROR([Could not find liba52 on your system: you may get it from http://liba52.sf.net/. Alternatively you can use --disable-a52 to disable the a52 plugin.])
+        ])
         ], [$LIBM])
     ],[
+      AS_IF([test "${enable_a52}" = "yes"],[
       AC_MSG_ERROR([Could not find liba52 on your system: you may get it from http://liba52.sf.net/. Alternatively you can use --disable-a52 to disable the a52 plugin.])
+      ])
     ])
     VLC_RESTORE_FLAGS
   fi
@@ -2985,7 +2995,7 @@ dnl
 dnl  X C Bindings modules
 dnl
 AC_ARG_ENABLE(xcb,
-  [  --enable-xcb            X11 support with XCB (default enabled)],, [
+  [  --enable-xcb            X11 support with XCB (default auto)],, [
   AS_IF([test "${SYS}" != "mingw32" -a "${SYS}" != "mingwce" -a "${SYS}" != "darwin" -a "${SYS}" != "symbian"], [
     enable_xcb="yes"
   ], [
@@ -3004,8 +3014,15 @@ AC_ARG_ENABLE(glx,
 have_xcb="no"
 AS_IF([test "${enable_xcb}" != "no"], [
   dnl libxcb
-  PKG_CHECK_MODULES(XCB, [xcb >= 1.6])
-  have_xcb="yes"
+
+  PKG_CHECK_MODULES(XCB, [xcb >= 1.6], [
+     have_xcb="yes"
+  ], [
+      AC_MSG_WARN([$XCB_PKG_ERRORS])
+  ])
+
+  AS_IF([test "${have_xcb}" != "no"], [
+
   PKG_CHECK_MODULES(XCB_SHM, [xcb-shm])
   PKG_CHECK_MODULES(XCB_COMPOSITE, [xcb-composite])
 
@@ -3044,6 +3061,8 @@ AS_IF([test "${enable_xcb}" != "no"], [
       AC_MSG_ERROR([${XLIB_XCB_PKG_ERRORS}. Pass --disable-glx if you do not need OpenGL X11 support.])
     ])
   ])
+
+  ])
 ])
 AM_CONDITIONAL([HAVE_XCB], [test "${have_xcb}" = "yes"])
 
@@ -3432,8 +3451,8 @@ AS_IF([test "${enable_alsa}" != "no"], [
   PKG_CHECK_MODULES([ALSA], [alsa >= 1.0.16], [
     have_alsa="yes"
   ], [
-    AS_IF([test "x${enable_alsa}" != "x"], [
-      AC_MSG_ERROR([$ALSA_PKG_ERRORS. alsa-lib 1.0.16 or later required.])
+    AS_IF([test "${enable_alsa}" = "yes"],[
+      AC_MSG_WARN([$ALSA_PKG_ERRORS. alsa-lib 1.0.16 or is not found.])
     ])
   ])
 ])
-- 
1.7.5.1


From rsatom at gmail.com  Thu Dec  1 03:13:06 2011
From: rsatom at gmail.com (Sergey Radionov)
Date: Thu, 1 Dec 2011 09:13:06 +0700
Subject: [vlc-devel] [PATCH] npapi: zero NPClass before using.
Message-ID: <CADOtZwiKxZW1f+Frg=MYQ9U1K1qjh5T_GaA2vtJkt1Sw=cyxeg@mail.gmail.com>


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111201/cda12053/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-npapi-zero-NPClass-before-using.patch
Type: application/octet-stream
Size: 802 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111201/cda12053/attachment.obj>

From rsatom at gmail.com  Thu Dec  1 04:25:27 2011
From: rsatom at gmail.com (Sergey Radionov)
Date: Thu, 1 Dec 2011 10:25:27 +0700
Subject: [vlc-devel] [PATCH] ActiveX,
	npapi win32: setup parent hwnd on every "play".
Message-ID: <CADOtZwiKmfQ+KWyWSL5Y17=gbitHM5H0N-3byXtLhzJyrSZ4ZA@mail.gmail.com>

But I don't shure, may be it is a bug in libvlc, when after "stop" it
forget about previously seted parent hwnd (via
libvlc_media_player_set_hwnd)?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111201/3d5352c9/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-ActiveX-npapi-win32-setup-parent-hwnd-on-every-play.patch
Type: application/octet-stream
Size: 1060 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111201/3d5352c9/attachment.obj>

From funman at videolan.org  Thu Dec  1 04:53:28 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Wed, 30 Nov 2011 22:53:28 -0500
Subject: [vlc-devel] [PATCH] npapi: zero NPClass before using.
In-Reply-To: <CADOtZwiKxZW1f+Frg=MYQ9U1K1qjh5T_GaA2vtJkt1Sw=cyxeg@mail.gmail.com>
References: <CADOtZwiKxZW1f+Frg=MYQ9U1K1qjh5T_GaA2vtJkt1Sw=cyxeg@mail.gmail.com>
Message-ID: <20111130225328.51f3e806@tok>

Le Thu, 1 Dec 2011 09:13:06 +0700,
Sergey Radionov <rsatom at gmail.com> a ?crit :

> From cab0163f1f342909a74cd7d82afd237b14e3e95f Mon Sep 17 00:00:00 2001
> From: Sergey Radionov <RSATom at gmail.com>
> Date: Thu, 1 Dec 2011 09:08:23 +0700
> Subject: [PATCH] npapi: zero NPClass before using.
> 
> fix #3880
> ---
>  npapi/control/nporuntime.h |    1 +
>  1 files changed, 1 insertions(+), 0 deletions(-)
> 
> diff --git a/npapi/control/nporuntime.h b/npapi/control/nporuntime.h
> index 799111a..117c584 100644
> --- a/npapi/control/nporuntime.h
> +++ b/npapi/control/nporuntime.h
> @@ -323,6 +323,7 @@ RuntimeNPClass<T>::RuntimeNPClass()
>                  T::methodCount, methodIdentifiers);
>      }
>  
> +    memset(static_cast<NPClass*>(this), 0, sizeof(NPClass));

Why is that needed, doesn't C++ have default constructors?

>      // fill in NPClass structure
>      structVersion  = NP_CLASS_STRUCT_VERSION;
>      allocate       = &RuntimeNPClassAllocate<T>;



-- 
Rafa?l Carr?


From rsatom at gmail.com  Thu Dec  1 07:20:25 2011
From: rsatom at gmail.com (Sergey Radionov)
Date: Thu, 01 Dec 2011 13:20:25 +0700
Subject: [vlc-devel] [PATCH] npapi: zero NPClass before using.
In-Reply-To: <20111130225328.51f3e806@tok>
References: <CADOtZwiKxZW1f+Frg=MYQ9U1K1qjh5T_GaA2vtJkt1Sw=cyxeg@mail.gmail.com>
	<20111130225328.51f3e806@tok>
Message-ID: <4ED71CA9.9060701@gmail.com>



01.12.2011 10:53, Rafa?l Carr? ?????:
> Le Thu, 1 Dec 2011 09:13:06 +0700,
> Sergey Radionov<rsatom at gmail.com>  a ?crit :
>
>>  From cab0163f1f342909a74cd7d82afd237b14e3e95f Mon Sep 17 00:00:00 2001
>> From: Sergey Radionov<RSATom at gmail.com>
>> Date: Thu, 1 Dec 2011 09:08:23 +0700
>> Subject: [PATCH] npapi: zero NPClass before using.
>>
>> fix #3880
>> ---
>>   npapi/control/nporuntime.h |    1 +
>>   1 files changed, 1 insertions(+), 0 deletions(-)
>>
>> diff --git a/npapi/control/nporuntime.h b/npapi/control/nporuntime.h
>> index 799111a..117c584 100644
>> --- a/npapi/control/nporuntime.h
>> +++ b/npapi/control/nporuntime.h
>> @@ -323,6 +323,7 @@ RuntimeNPClass<T>::RuntimeNPClass()
>>                   T::methodCount, methodIdentifiers);
>>       }
>>
>> +    memset(static_cast<NPClass*>(this), 0, sizeof(NPClass));
>
> Why is that needed, doesn't C++ have default constructors?
ISO/IEC 14882:2003
12.6.2 Initializing bases and members
?4
<..skipped..>
After the call to a constructor for class X has completed, if a member of X is neither specified in the
constructor?s mem-initializers, nor default-initialized, nor value-initialized, nor given a value 
during execution
of the body of the constructor, the member *has indeterminate value*.

>
>>       // fill in NPClass structure
>>       structVersion  = NP_CLASS_STRUCT_VERSION;
>>       allocate       =&RuntimeNPClassAllocate<T>;
>
>
>

-- 
With best wishes,
Sergey Radionov


From fatbull at web.de  Thu Dec  1 07:34:08 2011
From: fatbull at web.de (=?ISO-8859-15?Q?Tobias_G=FCntner?=)
Date: Thu, 01 Dec 2011 07:34:08 +0100
Subject: [vlc-devel] [Patches] Dash: mostly cosmetics
In-Reply-To: <CAK-C75SiVvUN_tnh+pvo+H5xCCNN+_Afi3qG+ZzsSMthAf8QaQ@mail.gmail.com>
References: <CAK-C75SN=k0WL==GiO89Fm1VmiqapCCZ7k1uOefyg9brc=Aafw@mail.gmail.com>
	<CAK-C75TGwHkm5iTZpydXUe3JGeW5CASrxrsu7xe3CYc5veb9Eg@mail.gmail.com>
	<CAK-C75S4oXDKecXAxwZyP6qfFRh0Q+85iPgYtpt23Yx0SWF7mA@mail.gmail.com>
	<201111241914.49264.remi@remlab.net>
	<CAK-C75TcveJQ3LeKUxA4JzupLiN5JAyUMi5=tGC+QXiTrqFhPQ@mail.gmail.com>
	<CAK-C75QRRLSqe6EKSsNEfnJGwOnjA8b80_t-aa245=bq167Rrg@mail.gmail.com>
	<4ECFD86A.3050305@web.de>
	<CAK-C75QujbKk9EP2M4aDSw26ttq8+H2ALU=RhUiraUB-BzAu4g@mail.gmail.com>
	<4ED47565.1090602@web.de>
	<CAK-C75QcXCDoJF7cpjZQRDHf30soO4bK5-90K57r98ycNb11Kg@mail.gmail.com>
	<4ED4CC73.6010008@web.de>
	<CAK-C75SiVvUN_tnh+pvo+H5xCCNN+_Afi3qG+ZzsSMthAf8QaQ@mail.gmail.com>
Message-ID: <4ED71FE0.5070003@web.de>

Am 30.11.2011 16:55, schrieb Hugo Beauz?e-Luyssen:
> Well the argument stating that "you don't want to do this" is wrong to
> me. Otherwise we may want to rewrite Qt to catch those exceptions, for
> instance.

Sure, but we have to resolve VLC ticket #35 first. ;) Let's focus on 
"our" code for now.

> However, yes, we should check for std::bac_alloc everywhere,
> or replacing new by new(nothrow) and check for NULL pointers.

Ah, I forgot about nothrow. ;) But it's not that simple, unfortunately. 
new(nothrow) only applies to the allocation itself; it does not prevent 
the constructor from throwing an exception (including bad_alloc).

> My point about exceptions here was : they are not required, and can be
> replaced by a (IMHO) cleaner solution, IE a proper return value.

My point is: STL containers rely on exceptions, thus replacing 
exceptions with something else is not (easily) possible.

For example, could you rewrite DOMHelper::getElementByTagName without 
exceptions? And would it really be "cleaner" in some way?
(Specifically, I am referring to the allocation of, insertions into, and 
return of the elements vector. All those operations can throw. The calls 
to at(i) should not be a problem, however.)

I think we agree that errors must be handled, but we disagree regarding 
the error *reporting* mechanism. I look at it this way: Banning 
exceptions is like shooting the proverbial messenger, just so that you 
can deliver the bad news yourself. ;)

What we "gain" by this is that the compiler no longer takes care of 
error reporting for us; we need to do it ourselves. I find that 
extremely tedious and annoying. And since we cannot change the STL, we 
*still* have to deal with its exceptions. IMHO this is a lot of trouble 
for very little benefit, if any.

Regards,
Tobias



From remi at remlab.net  Thu Dec  1 08:04:46 2011
From: remi at remlab.net (=?UTF-8?Q?R=C3=A9mi_Denis-Courmont?=)
Date: Thu, 01 Dec 2011 08:04:46 +0100
Subject: [vlc-devel] [RFC,
 PATCH] configure: Do not stop configure if component is not found.
In-Reply-To: <4ED6D93C.5010008@unixsol.org>
References: <4ED6D93C.5010008@unixsol.org>
Message-ID: <819cd0191398e013825a23e1c98bdefd@chewa.net>

    Hello,

On Thu, 01 Dec 2011 03:32:44 +0200, Georgi Chorbadzhiyski <gf at unixsol.org>
wrote:
> I'm compiling vlc on a server with only minimal needed components
> (libav, decklink, a52, x264) and without adding bunch of --disable-xxx
> options configure stops with error on several occasions. IMHO if
> component is not requested via --enable-xxx the configure should
> mark it as not found and go on. The attached patch does this for
> the components that I do not have but are generating errors.

It is far too easy to build a completely useless VLC with that approach.
For instance, it is completely unreasonable NOT to fail by default if XCB
is missing on Linux.

So I am very much against your proposal.

-- 
R?mi Denis-Courmont
http://www.remlab.net/


From remi at remlab.net  Thu Dec  1 08:06:03 2011
From: remi at remlab.net (=?UTF-8?Q?R=C3=A9mi_Denis-Courmont?=)
Date: Thu, 01 Dec 2011 08:06:03 +0100
Subject: [vlc-devel] [PATCH] add a way to enable/disable video effects
 via libvlc
In-Reply-To: <1322690764.1554.3.camel@aldrin>
References: <1322429874.1554.13.camel@aldrin>
	<201111282223.13713.remi@remlab.net>
	<1322587662.1588.11.camel@aldrin> <1322690764.1554.3.camel@aldrin>
Message-ID: <e966df51def904cccca19876dcf46aef@chewa.net>

Please use a correct function prefix. libvlc_set_video_filter() sounds
like it applies to a whole Libvlc instance.

-- 
R?mi Denis-Courmont
http://www.remlab.net/


From remi at remlab.net  Thu Dec  1 08:07:08 2011
From: remi at remlab.net (=?UTF-8?Q?R=C3=A9mi_Denis-Courmont?=)
Date: Thu, 01 Dec 2011 08:07:08 +0100
Subject: [vlc-devel]
 =?utf-8?q?=5BPATCH=5D_Added_libvlc=5Fvideo=5Ftake=5Fs?=
 =?utf-8?q?napshot=5Faddr_to_libvlc_and_related_functions_to_libvlccore?=
In-Reply-To: <1322689414.1554.0.camel@aldrin>
References: <1322508588.1586.0.camel@aldrin>
	<20111128203428.GA31114@elivagar.org>
	<1322513160.1586.1.camel@aldrin> <1322515374.1586.2.camel@aldrin>
	<20111128213008.GA1747@elivagar.org>
	<1322586972.1588.3.camel@aldrin>
	<4ED5164B.9010103@alexeysokolov.co.cc>
	<1322689414.1554.0.camel@aldrin>
Message-ID: <a8a9b166865d290d4beb39c0e809cadc@chewa.net>

On Wed, 30 Nov 2011 22:43:34 +0100, PaulK <gci at paulk.fr> wrote:
> diff --git a/include/vlc/libvlc_media_player.h
> b/include/vlc/libvlc_media_player.h
> index a2b5748..b2a5c54 100644
> --- a/include/vlc/libvlc_media_player.h
> +++ b/include/vlc/libvlc_media_player.h
> @@ -35,6 +35,7 @@
>  extern "C" {
>  # else
>  #  include <stdbool.h>
> +#  include <stdlib.h>
>  # endif
>  
> 
/*****************************************************************************
> @@ -1151,6 +1152,25 @@ int
> libvlc_video_take_snapshot( libvlc_media_player_t *p_mi, unsigned num,
>                                  unsigned int i_height );
>  
>  /**
> + * Take a snapshot of the current video window and return the snapshot
> address.
> + *
> + * If i_width AND i_height is 0, original size is used.
> + * If i_width XOR i_height is 0, original aspect-ratio is preserved.
> + *
> + * \param p_mi media player instance
> + * \param num number of video output (typically 0 for the first/only
> one)
> + * \param psz_addr place to return the snapshot address (you MUST free
> it after use)
> + * \param i_width the snapshot's width
> + * \param i_height the snapshot's height
> + * \return the snapshot memory area size on success, -1 if there was a
> problem
> + */
> +LIBVLC_API
> +ssize_t
> +libvlc_video_snapshot_get_data( libvlc_media_player_t *p_mi, unsigned
> num,
> +                                 void **psz_addr, unsigned int
> i_width, 
> +                                 unsigned int i_height );
> +
> +/**
>   * Enable or disable deinterlace filter
>   *
>   * \param p_mi libvlc media player
> diff --git a/lib/libvlc.sym b/lib/libvlc.sym
> index 6582a96..099cbb8 100644
> --- a/lib/libvlc.sym
> +++ b/lib/libvlc.sym
> @@ -215,6 +215,7 @@ libvlc_video_set_spu
>  libvlc_video_set_subtitle_file
>  libvlc_video_set_teletext
>  libvlc_video_set_track
> +libvlc_video_snapshot_get_data
>  libvlc_video_take_snapshot
>  libvlc_vlm_add_broadcast
>  libvlc_vlm_add_vod
> diff --git a/lib/video.c b/lib/video.c
> index 6e1ea06..3071646 100644
> --- a/lib/video.c
> +++ b/lib/video.c
> @@ -34,6 +34,7 @@
>  #include <vlc/libvlc_media_player.h>
>  
>  #include <vlc_common.h>
> +#include <vlc_block.h>
>  #include <vlc_input.h>
>  #include <vlc_vout.h>
>  
> @@ -158,6 +159,63 @@ libvlc_video_take_snapshot( libvlc_media_player_t
> *p_mi, unsigned num,
>      return 0;
>  }
>  
> +ssize_t
> +libvlc_video_snapshot_get_data( libvlc_media_player_t *p_mi, unsigned
> num,
> +                                 void **psz_addr, unsigned int
> i_width, 
> +                                 unsigned int i_height )
> +{
> +    block_t *block_buffer;
> +    size_t buffer_len = 0;
> +    char *buffer = NULL;
> +    ssize_t ret_len;
> +    video_format_t fmt;
> +
> +    vout_thread_t *p_vout = GetVout( p_mi, num );
> +    if (p_vout == NULL)
> +    {
> +        libvlc_printerr ("No usable vout");
> +        return -1;
> +    }
> +
> +    var_SetInteger( p_vout, "snapshot-width", i_width);
> +    var_SetInteger( p_vout, "snapshot-height", i_height );

Not thread-safe.

> +
> +    if( vout_GetSnapshot( p_vout, &block_buffer, NULL, &fmt, "png",
> 500*1000 ) )
> +    {
> +        libvlc_printerr ("Can't get the snapshot");
> +        ret_len = -1;
> +        goto exit;
> +    }
> +
> +
> +    buffer_len = block_buffer->i_buffer;
> +    if( buffer_len <= 0 )
> +    {
> +        libvlc_printerr ("Snapshot buffer length is negative or zero");
> +        ret_len = -1;
> +        goto exit;
> +    }
> +
> +    buffer = malloc( buffer_len );
> +    if( buffer == NULL )
> +    {
> +        libvlc_printerr ("buffer memory allocation failed");
> +        ret_len = -1;
> +        goto exit;
> +    }
> +
> +    psz_addr[0] = buffer;
> +    memcpy( buffer, block_buffer->p_buffer, buffer_len );

For the second time, please don't do this. We have buffer management.

> +
> +    ret_len = buffer_len;
> +
> +exit:
> +    if( block_buffer != NULL )
> +        block_Release( block_buffer );
> +
> +    return ret_len;
> +}
> +
>  int libvlc_video_get_size( libvlc_media_player_t *p_mi, unsigned num,
>                             unsigned *restrict px, unsigned *restrict py
> )
>  {
> 
> 
> _______________________________________________
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
> http://mailman.videolan.org/listinfo/vlc-devel

-- 
R?mi Denis-Courmont
http://www.remlab.net/


From venkatesh.rk1 at gmail.com  Thu Dec  1 08:53:43 2011
From: venkatesh.rk1 at gmail.com (venkatesh k)
Date: Thu, 1 Dec 2011 13:23:43 +0530
Subject: [vlc-devel] VLC API for UPNP service discovery
In-Reply-To: <CAA+KajE1_oRPg9esBHVnM0dVuE73iyUOrWCqcm3RNmb3nfyOmw@mail.gmail.com>
References: <CAA+KajE1_oRPg9esBHVnM0dVuE73iyUOrWCqcm3RNmb3nfyOmw@mail.gmail.com>
Message-ID: <CAA+KajG-HAipVj0-Tcxv_SHzULAFasG=QUzdHRFj6yUivSJr+A@mail.gmail.com>

Hello All,

  I want to know the VLC API which searches all upnp servers connected in a
network, so that I can add some more new servers manually.
I am having VLC1.1.12 package.

Thanks and regards,
Venkatesh K
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111201/75d14d1d/attachment.html>

From jb at videolan.org  Thu Dec  1 09:50:08 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Thu, 1 Dec 2011 09:50:08 +0100
Subject: [vlc-devel] [PATCH] demux/mp4: add parsing of mfra atom
In-Reply-To: <CAKHNsC3r18w3y48j6+Wta=HYJhNezQcHs=wzndWaoKqKu1J-vg@mail.gmail.com>
References: <4ed50504.8872cd0a.7557.ffff95cf@mx.google.com>
	<20111130000352.GD27521@videolan.org>
	<CAKHNsC26tSz_pinXu+0HJE5svyVBf+z5_xzgC8VLPZdFsnCLAQ@mail.gmail.com>
	<CAKHNsC3r18w3y48j6+Wta=HYJhNezQcHs=wzndWaoKqKu1J-vg@mail.gmail.com>
Message-ID: <20111201085008.GA15774@videolan.org>

On Wed, Nov 30, 2011 at 05:48:54PM +0100, Frederic YHUEL wrote :
> > I'll send a new patch.
> >
> 
> Here it is.

Applied.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From thresh at videolan.org  Thu Dec  1 09:55:05 2011
From: thresh at videolan.org (Konstantin Pavlov)
Date: Thu, 1 Dec 2011 12:55:05 +0400
Subject: [vlc-devel] VLC API for UPNP service discovery
In-Reply-To: <CAA+KajG-HAipVj0-Tcxv_SHzULAFasG=QUzdHRFj6yUivSJr+A@mail.gmail.com>
References: <CAA+KajE1_oRPg9esBHVnM0dVuE73iyUOrWCqcm3RNmb3nfyOmw@mail.gmail.com>
	<CAA+KajG-HAipVj0-Tcxv_SHzULAFasG=QUzdHRFj6yUivSJr+A@mail.gmail.com>
Message-ID: <20111201085505.GI13414@snowwhite.immo>

On Thu, Dec 01, 2011 at 01:23:43PM +0530, venkatesh k wrote:
> Hello All,
> 
>   I want to know the VLC API which searches all upnp servers connected in a
> network, so that I can add some more new servers manually.
> I am having VLC1.1.12 package.

I am not sure what you mean about "add some more new servers manually",
but VLC uses libupnp (http://pupnp.sourceforge.net/) to talk to UPNP
servers.

You might also want to upgrade to VLC 1.2, VLC UPNP code changed a bit
there (see modules/services_discovery/upnp.* ).

-- 
Konstantin Pavlov
VideoLAN team


From jb at videolan.org  Thu Dec  1 09:57:28 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Thu, 1 Dec 2011 09:57:28 +0100
Subject: [vlc-devel] [PATCH] npapi: zero NPClass before using.
In-Reply-To: <20111130225328.51f3e806@tok>
References: <CADOtZwiKxZW1f+Frg=MYQ9U1K1qjh5T_GaA2vtJkt1Sw=cyxeg@mail.gmail.com>
	<20111130225328.51f3e806@tok>
Message-ID: <20111201085727.GA16154@videolan.org>

On Wed, Nov 30, 2011 at 10:53:28PM -0500, Rafa?l Carr? wrote :
> > +    memset(static_cast<NPClass*>(this), 0, sizeof(NPClass));
> 
> Why is that needed, doesn't C++ have default constructors?
> 

NPClass, as its name says, is a struct.


-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Thu Dec  1 09:59:31 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Thu, 1 Dec 2011 09:59:31 +0100
Subject: [vlc-devel] [PATCH] npapi: zero NPClass before using.
In-Reply-To: <CADOtZwiKxZW1f+Frg=MYQ9U1K1qjh5T_GaA2vtJkt1Sw=cyxeg@mail.gmail.com>
References: <CADOtZwiKxZW1f+Frg=MYQ9U1K1qjh5T_GaA2vtJkt1Sw=cyxeg@mail.gmail.com>
Message-ID: <20111201085931.GB16154@videolan.org>

On Thu, Dec 01, 2011 at 09:13:06AM +0700, Sergey Radionov wrote :
> 

Wouldn't setting the unset parameters to NULL be faster, and more
elegant?

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Thu Dec  1 10:00:54 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Thu, 1 Dec 2011 10:00:54 +0100
Subject: [vlc-devel] [RFC,
 PATCH] configure: Do not stop configure if component is not found.
In-Reply-To: <819cd0191398e013825a23e1c98bdefd@chewa.net>
References: <4ED6D93C.5010008@unixsol.org>
	<819cd0191398e013825a23e1c98bdefd@chewa.net>
Message-ID: <20111201090054.GC16154@videolan.org>

On Thu, Dec 01, 2011 at 08:04:46AM +0100, R?mi Denis-Courmont wrote :
>     Hello,
> 
> On Thu, 01 Dec 2011 03:32:44 +0200, Georgi Chorbadzhiyski <gf at unixsol.org>
> wrote:
> > I'm compiling vlc on a server with only minimal needed components
> > (libav, decklink, a52, x264) and without adding bunch of --disable-xxx
> > options configure stops with error on several occasions. IMHO if
> > component is not requested via --enable-xxx the configure should
> > mark it as not found and go on. The attached patch does this for
> > the components that I do not have but are generating errors.
> 
> It is far too easy to build a completely useless VLC with that approach.
> For instance, it is completely unreasonable NOT to fail by default if XCB
> is missing on Linux.

Indeed, but what about a52, for example?

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From sebastien-devel at celeos.eu  Thu Dec  1 10:04:14 2011
From: sebastien-devel at celeos.eu (=?ISO-8859-1?Q?S=E9bastien?= Escudier)
Date: Thu, 01 Dec 2011 10:04:14 +0100
Subject: [vlc-devel] [PATCH] live555 : enable/disable tracks on the fly
Message-ID: <1322730254.10961.3.camel@stim-desktop>

Hi

3 patches attached.
I did not test multiple audio or multiple video tracks case because I
don't have rtsp servers like that.

Regards,
S?bastien.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-live555-use-TAB_-macro-to-store-the-tracks.patch
Type: text/x-patch
Size: 2475 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111201/abbbe7e4/attachment.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0002-live555-when-we-switch-to-TCP-remember-it.patch
Type: text/x-patch
Size: 1268 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111201/abbbe7e4/attachment-0001.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0003-live555-enable-disable-tracks-on-the-fly.patch
Type: text/x-patch
Size: 3459 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111201/abbbe7e4/attachment-0002.bin>

From rsatom at gmail.com  Thu Dec  1 10:06:35 2011
From: rsatom at gmail.com (Sergey Radionov)
Date: Thu, 1 Dec 2011 16:06:35 +0700
Subject: [vlc-devel] [PATCH] npapi: zero NPClass before using.
In-Reply-To: <20111201085931.GB16154@videolan.org>
References: <CADOtZwiKxZW1f+Frg=MYQ9U1K1qjh5T_GaA2vtJkt1Sw=cyxeg@mail.gmail.com>
	<20111201085931.GB16154@videolan.org>
Message-ID: <CADOtZwgFz4j=LZjX35ipkNgTtakuAKLu_xDYYbGE-BCp3HzfUw@mail.gmail.com>

2011/12/1 Jean-Baptiste Kempf <jb at videolan.org>

> On Thu, Dec 01, 2011 at 09:13:06AM +0700, Sergey Radionov wrote :
> >
>
> Wouldn't setting the unset parameters to NULL be faster, and more
> elegant?
>
Maybe, but I am take care about possible future adding of new members to
this struct.

But I am not insist. So what you say, should I replace it with simple
assignment?



> Best regards,
>
> --
> Jean-Baptiste Kempf
> http://www.jbkempf.com/ - +33 672 704 734
> Sent from my Electronic Device
> _______________________________________________
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
> http://mailman.videolan.org/listinfo/vlc-devel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111201/649d2099/attachment.html>

From jb at videolan.org  Thu Dec  1 10:08:24 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Thu, 1 Dec 2011 10:08:24 +0100
Subject: [vlc-devel] [PATCH] npapi: zero NPClass before using.
In-Reply-To: <CADOtZwgFz4j=LZjX35ipkNgTtakuAKLu_xDYYbGE-BCp3HzfUw@mail.gmail.com>
References: <CADOtZwiKxZW1f+Frg=MYQ9U1K1qjh5T_GaA2vtJkt1Sw=cyxeg@mail.gmail.com>
	<20111201085931.GB16154@videolan.org>
	<CADOtZwgFz4j=LZjX35ipkNgTtakuAKLu_xDYYbGE-BCp3HzfUw@mail.gmail.com>
Message-ID: <20111201090824.GA19737@videolan.org>

On Thu, Dec 01, 2011 at 04:06:35PM +0700, Sergey Radionov wrote :
> 2011/12/1 Jean-Baptiste Kempf <jb at videolan.org>
> 
> > On Thu, Dec 01, 2011 at 09:13:06AM +0700, Sergey Radionov wrote :
> > >
> >
> > Wouldn't setting the unset parameters to NULL be faster, and more
> > elegant?
> >
> Maybe, but I am take care about possible future adding of new members to
> this struct.

The probability of extension of NPClass is quite low...

> But I am not insist. So what you say, should I replace it with simple
> assignment?

Well, seeing the doc and the headers, almost all members are init'd, but
a few ones. Setting them as NULL might be enough (and maybe not, so this
approach would be needed).

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From rsatom at gmail.com  Thu Dec  1 10:10:26 2011
From: rsatom at gmail.com (Sergey Radionov)
Date: Thu, 1 Dec 2011 16:10:26 +0700
Subject: [vlc-devel] [PATCH] npapi: zero NPClass before using.
In-Reply-To: <20111201090824.GA19737@videolan.org>
References: <CADOtZwiKxZW1f+Frg=MYQ9U1K1qjh5T_GaA2vtJkt1Sw=cyxeg@mail.gmail.com>
	<20111201085931.GB16154@videolan.org>
	<CADOtZwgFz4j=LZjX35ipkNgTtakuAKLu_xDYYbGE-BCp3HzfUw@mail.gmail.com>
	<20111201090824.GA19737@videolan.org>
Message-ID: <CADOtZwiw=N9H+Na9-SO92cJCrPeR=AEVh7+LfnuB18abjPvUzA@mail.gmail.com>

2011/12/1 Jean-Baptiste Kempf <jb at videolan.org>

> On Thu, Dec 01, 2011 at 04:06:35PM +0700, Sergey Radionov wrote :
> > 2011/12/1 Jean-Baptiste Kempf <jb at videolan.org>
> >
> > > On Thu, Dec 01, 2011 at 09:13:06AM +0700, Sergey Radionov wrote :
> > > >
> > >
> > > Wouldn't setting the unset parameters to NULL be faster, and more
> > > elegant?
> > >
> > Maybe, but I am take care about possible future adding of new members to
> > this struct.
>
> The probability of extension of NPClass is quite low...
>
> > But I am not insist. So what you say, should I replace it with simple
> > assignment?
>
> Well, seeing the doc and the headers, almost all members are init'd, but
> a few ones. Setting them as NULL might be enough (and maybe not, so this
> approach would be needed).
>
Ok, as you wish, I'll make new patch, not problem.


>
> Best regards,
>
> --
> Jean-Baptiste Kempf
> http://www.jbkempf.com/ - +33 672 704 734
> Sent from my Electronic Device
> _______________________________________________
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
> http://mailman.videolan.org/listinfo/vlc-devel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111201/7cfe641a/attachment.html>

From rsatom at gmail.com  Thu Dec  1 10:13:16 2011
From: rsatom at gmail.com (Sergey Radionov)
Date: Thu, 1 Dec 2011 16:13:16 +0700
Subject: [vlc-devel] [PATCH] npapi: zero NPClass before using.
In-Reply-To: <20111201085727.GA16154@videolan.org>
References: <CADOtZwiKxZW1f+Frg=MYQ9U1K1qjh5T_GaA2vtJkt1Sw=cyxeg@mail.gmail.com>
	<20111130225328.51f3e806@tok> <20111201085727.GA16154@videolan.org>
Message-ID: <CADOtZwjuLniD2E8_0nbBK3ocH_pbtmZ4_fXNyFv8m3ucJP0=Qg@mail.gmail.com>

2011/12/1 Jean-Baptiste Kempf <jb at videolan.org>

> On Wed, Nov 30, 2011 at 10:53:28PM -0500, Rafa?l Carr? wrote :
> > > +    memset(static_cast<NPClass*>(this), 0, sizeof(NPClass));
> >
> > Why is that needed, doesn't C++ have default constructors?
> >
>
> NPClass, as its name says, is a struct.
>
maybe you will be a bit surprised, but in C++ structs have constructors :)


>
> --
> Jean-Baptiste Kempf
> http://www.jbkempf.com/ - +33 672 704 734
> Sent from my Electronic Device
> _______________________________________________
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
> http://mailman.videolan.org/listinfo/vlc-devel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111201/fc23b406/attachment.html>

From mirsal at mirsal.fr  Thu Dec  1 11:34:04 2011
From: mirsal at mirsal.fr (mirsal)
Date: Thu, 01 Dec 2011 11:34:04 +0100
Subject: [vlc-devel] VLC API for UPNP service discovery
In-Reply-To: <CAA+KajG-HAipVj0-Tcxv_SHzULAFasG=QUzdHRFj6yUivSJr+A@mail.gmail.com>
References: <CAA+KajE1_oRPg9esBHVnM0dVuE73iyUOrWCqcm3RNmb3nfyOmw@mail.gmail.com>
	<CAA+KajG-HAipVj0-Tcxv_SHzULAFasG=QUzdHRFj6yUivSJr+A@mail.gmail.com>
Message-ID: <1322735644.18980.19.camel@mirsal-laptop1.mirsal.fr>

Hello,

On Thu, 2011-12-01 at 13:23 +0530, venkatesh k wrote:
>   I want to know the VLC API which searches all upnp servers connected in a
> network, so that I can add some more new servers manually.
> I am having VLC1.1.12 package.

The services discovery module uses the UpnpSearchAsync() function from
pupnp/libupnp in order to discover media servers.

If you want yours to show up in VLC, then it should broadcast proper
SSDP announcements.

Kind regards,

-- 
mirsal 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 490 bytes
Desc: This is a digitally signed message part
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111201/0b07fe18/attachment.sig>

From gf at unixsol.org  Thu Dec  1 11:47:14 2011
From: gf at unixsol.org (Georgi Chorbadzhiyski)
Date: Thu, 01 Dec 2011 12:47:14 +0200
Subject: [vlc-devel] [RFC,
 PATCH] configure: Do not stop configure if component is not found.
In-Reply-To: <819cd0191398e013825a23e1c98bdefd@chewa.net>
References: <4ED6D93C.5010008@unixsol.org>
	<819cd0191398e013825a23e1c98bdefd@chewa.net>
Message-ID: <4ED75B32.2000908@unixsol.org>

Around 12/01/2011 09:04 AM, R?mi Denis-Courmont scribbled:
> On Thu, 01 Dec 2011 03:32:44 +0200, Georgi Chorbadzhiyski <gf at unixsol.org>
> wrote:
>> I'm compiling vlc on a server with only minimal needed components
>> (libav, decklink, a52, x264) and without adding bunch of --disable-xxx
>> options configure stops with error on several occasions. IMHO if
>> component is not requested via --enable-xxx the configure should
>> mark it as not found and go on. The attached patch does this for
>> the components that I do not have but are generating errors.
> 
> It is far too easy to build a completely useless VLC with that approach.
> For instance, it is completely unreasonable NOT to fail by default if XCB
> is missing on Linux.
> 
> So I am very much against your proposal.

I can understand your point about xcb (and maybe alsa), but surely
dbus-control (I do not have dbus at all), lua (not needed) and a52 (just
warn possibly?) are not vital to vlc being usable.

-- 
Georgi Chorbadzhiyski
http://georgi.unixsol.org/


From remi at remlab.net  Thu Dec  1 12:00:13 2011
From: remi at remlab.net (=?UTF-8?Q?R=C3=A9mi_Denis-Courmont?=)
Date: Thu, 01 Dec 2011 12:00:13 +0100
Subject: [vlc-devel] [RFC,
 PATCH] configure: Do not stop configure if component is not found.
In-Reply-To: <4ED75B32.2000908@unixsol.org>
References: <4ED6D93C.5010008@unixsol.org>
	<819cd0191398e013825a23e1c98bdefd@chewa.net>
	<4ED75B32.2000908@unixsol.org>
Message-ID: <122f6dce5b7d51ea0e00381df60c8238@chewa.net>

On Thu, 01 Dec 2011 12:47:14 +0200, Georgi Chorbadzhiyski <gf at unixsol.org>
wrote:
> I can understand your point about xcb (and maybe alsa), but surely
> dbus-control (I do not have dbus at all), lua (not needed) and a52 (just
> warn possibly?) are not vital to vlc being usable.

LUA is very much needed nowadays.

And until we get a proper A/52 decoder plugin through libav, a52 is needed
as well.

-- 
R?mi Denis-Courmont
http://www.remlab.net/


From jb at videolan.org  Thu Dec  1 12:08:40 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Thu, 1 Dec 2011 12:08:40 +0100
Subject: [vlc-devel] [RFC,
 PATCH] configure: Do not stop configure if component is not found.
In-Reply-To: <122f6dce5b7d51ea0e00381df60c8238@chewa.net>
References: <4ED6D93C.5010008@unixsol.org>
	<819cd0191398e013825a23e1c98bdefd@chewa.net>
	<4ED75B32.2000908@unixsol.org>
	<122f6dce5b7d51ea0e00381df60c8238@chewa.net>
Message-ID: <20111201110840.GA11044@videolan.org>

On Thu, Dec 01, 2011 at 12:00:13PM +0100, R?mi Denis-Courmont wrote :
> On Thu, 01 Dec 2011 12:47:14 +0200, Georgi Chorbadzhiyski <gf at unixsol.org>
> wrote:
> > I can understand your point about xcb (and maybe alsa), but surely
> > dbus-control (I do not have dbus at all), lua (not needed) and a52 (just
> > warn possibly?) are not vital to vlc being usable.
> 
> LUA is very much needed nowadays.
> 
> And until we get a proper A/52 decoder plugin through libav, a52 is needed
> as well.

How so?
You might not want any A/52 decoding, no?

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From beauze.h at gmail.com  Thu Dec  1 12:15:07 2011
From: beauze.h at gmail.com (=?UTF-8?Q?Hugo_Beauz=C3=A9e=2DLuyssen?=)
Date: Thu, 1 Dec 2011 12:15:07 +0100
Subject: [vlc-devel] [Patches] Dash: mostly cosmetics
In-Reply-To: <4ED71FE0.5070003@web.de>
References: <CAK-C75SN=k0WL==GiO89Fm1VmiqapCCZ7k1uOefyg9brc=Aafw@mail.gmail.com>
	<CAK-C75TGwHkm5iTZpydXUe3JGeW5CASrxrsu7xe3CYc5veb9Eg@mail.gmail.com>
	<CAK-C75S4oXDKecXAxwZyP6qfFRh0Q+85iPgYtpt23Yx0SWF7mA@mail.gmail.com>
	<201111241914.49264.remi@remlab.net>
	<CAK-C75TcveJQ3LeKUxA4JzupLiN5JAyUMi5=tGC+QXiTrqFhPQ@mail.gmail.com>
	<CAK-C75QRRLSqe6EKSsNEfnJGwOnjA8b80_t-aa245=bq167Rrg@mail.gmail.com>
	<4ECFD86A.3050305@web.de>
	<CAK-C75QujbKk9EP2M4aDSw26ttq8+H2ALU=RhUiraUB-BzAu4g@mail.gmail.com>
	<4ED47565.1090602@web.de>
	<CAK-C75QcXCDoJF7cpjZQRDHf30soO4bK5-90K57r98ycNb11Kg@mail.gmail.com>
	<4ED4CC73.6010008@web.de>
	<CAK-C75SiVvUN_tnh+pvo+H5xCCNN+_Afi3qG+ZzsSMthAf8QaQ@mail.gmail.com>
	<4ED71FE0.5070003@web.de>
Message-ID: <CAK-C75TCOTO2dUUHz8HDPk4X395ifWNnO5bdzS2=JBJtOtkL_Q@mail.gmail.com>

On Thu, Dec 1, 2011 at 7:34 AM, Tobias G?ntner <fatbull at web.de> wrote:
> Am 30.11.2011 16:55, schrieb Hugo Beauz?e-Luyssen:
>
>> Well the argument stating that "you don't want to do this" is wrong to
>> me. Otherwise we may want to rewrite Qt to catch those exceptions, for
>> instance.
>
>
> Sure, but we have to resolve VLC ticket #35 first. ;) Let's focus on "our"
> code for now.
>

That's exactly what I'm trying to do. However I guess your point here
is STL is "our" code, while Qt is not. However the point stays, to me.

>
>> However, yes, we should check for std::bac_alloc everywhere,
>> or replacing new by new(nothrow) and check for NULL pointers.
>
>
> Ah, I forgot about nothrow. ;) But it's not that simple, unfortunately.
> new(nothrow) only applies to the allocation itself; it does not prevent the
> constructor from throwing an exception (including bad_alloc).
>

Indeed. We should check for bad_allow, bad_cast if we use references,
bad_typeid and ios_base::failure.
Saying that I've never encountered a code catching these exceptions
would be falacious, but true nevertheless.

>
>> My point about exceptions here was : they are not required, and can be
>> replaced by a (IMHO) cleaner solution, IE a proper return value.
>
>
> My point is: STL containers rely on exceptions, thus replacing exceptions
> with something else is not (easily) possible.
>

I'm not saying we should get read of the STL exceptions, just most of
those in Dash, that are useless most of the time.

> For example, could you rewrite DOMHelper::getElementByTagName without
> exceptions? And would it really be "cleaner" in some way?
> (Specifically, I am referring to the allocation of, insertions into, and
> return of the elements vector. All those operations can throw. The calls to
> at(i) should not be a problem, however.)
>

Again, I'm fine (and don't have much choice) with STL exceptions.

> I think we agree that errors must be handled, but we disagree regarding the
> error *reporting* mechanism. I look at it this way: Banning exceptions is
> like shooting the proverbial messenger, just so that you can deliver the bad
> news yourself. ;)
>
> What we "gain" by this is that the compiler no longer takes care of error
> reporting for us; we need to do it ourselves. I find that extremely tedious
> and annoying. And since we cannot change the STL, we *still* have to deal
> with its exceptions. IMHO this is a lot of trouble for very little benefit,
> if any.
>

Let's take a code sample :

Representation.cpp :

SegmentInfo*        Representation::getSegmentInfo          () //...
{
    if(this->segmentInfo == NULL)
        throw ElementNotPresentException();

    return this->trickModeType;
}

BasicCMManager.cpp:

try
    {
        SegmentInfo* info = rep->getSegmentInfo();
        //....
    }
    catch(ElementNotPresentException &e)
    {
        /*TODO Debug */
    }

The error is handled twice. One so we can raise, and one to catch the error.
It looks clearer to me just to return the pointer, regardless of it's
value, and check it just once.

Again, I'm not saying anything about STL's exceptions. They're not
handled for now anyway.

>
> Regards,
> Tobias
>

Regards,

-- 
Hugo Beauz?e-Luyssen


From beauze.h at gmail.com  Thu Dec  1 12:20:08 2011
From: beauze.h at gmail.com (=?UTF-8?Q?Hugo_Beauz=C3=A9e=2DLuyssen?=)
Date: Thu, 1 Dec 2011 12:20:08 +0100
Subject: [vlc-devel] [PATCH] npapi: zero NPClass before using.
In-Reply-To: <20111201085931.GB16154@videolan.org>
References: <CADOtZwiKxZW1f+Frg=MYQ9U1K1qjh5T_GaA2vtJkt1Sw=cyxeg@mail.gmail.com>
	<20111201085931.GB16154@videolan.org>
Message-ID: <CAK-C75QY49p=ggTKJfPOdF5wsL7esnNuv=fq5c-rKm5Jqc7hzQ@mail.gmail.com>

On Thu, Dec 1, 2011 at 9:59 AM, Jean-Baptiste Kempf <jb at videolan.org> wrote:
> On Thu, Dec 01, 2011 at 09:13:06AM +0700, Sergey Radionov wrote :
>>
>
> Wouldn't setting the unset parameters to NULL be faster, and more
> elegant?
>
> Best regards,
>

Faster, I don't know, but if at some point this class/struct has some
virtual methods, you're pretty much doomed. (I haven't read this code,
it was just my 2 cents :) )

-- 
Hugo Beauz?e-Luyssen


From gf at unixsol.org  Thu Dec  1 12:21:43 2011
From: gf at unixsol.org (Georgi Chorbadzhiyski)
Date: Thu, 01 Dec 2011 13:21:43 +0200
Subject: [vlc-devel] [RFC,
 PATCH] configure: Do not stop configure if component is not found.
In-Reply-To: <20111201110840.GA11044@videolan.org>
References: <4ED6D93C.5010008@unixsol.org>
	<819cd0191398e013825a23e1c98bdefd@chewa.net>
	<4ED75B32.2000908@unixsol.org>
	<122f6dce5b7d51ea0e00381df60c8238@chewa.net>
	<20111201110840.GA11044@videolan.org>
Message-ID: <4ED76347.7000405@unixsol.org>

Around 12/01/2011 01:08 PM, Jean-Baptiste Kempf scribbled:
> On Thu, Dec 01, 2011 at 12:00:13PM +0100, R?mi Denis-Courmont wrote :
>> On Thu, 01 Dec 2011 12:47:14 +0200, Georgi Chorbadzhiyski <gf at unixsol.org>
>> wrote:
>>> I can understand your point about xcb (and maybe alsa), but surely
>>> dbus-control (I do not have dbus at all), lua (not needed) and a52 (just
>>> warn possibly?) are not vital to vlc being usable.
>>
>> LUA is very much needed nowadays.
>>
>> And until we get a proper A/52 decoder plugin through libav, a52 is needed
>> as well.
> 
> How so?
> You might not want any A/52 decoding, no?

My point exactly, configure should not require --disable-xxxxx just to silence
an error that if xxxx is not found. If it is not found just mark it so and go on.

Otherwise it is cat and mouse game. ./configure ... wait....aaaa an error, whats
the disable option (grep for it)...run configure --disable-abc ... wait....loop
if an error is found.

And lets not forget that configure is huge and it is taking lots of time to run.

-- 
Georgi Chorbadzhiyski
http://georgi.unixsol.org/


From beauze.h at gmail.com  Thu Dec  1 12:40:22 2011
From: beauze.h at gmail.com (=?UTF-8?Q?Hugo_Beauz=C3=A9e=2DLuyssen?=)
Date: Thu, 1 Dec 2011 12:40:22 +0100
Subject: [vlc-devel] [Patch] Adding an helper to convert iso8601
 durations to seconds
In-Reply-To: <CAK-C75TPsc1Lh07NNXFnmzW5ZQgSQbkfm1S2Vfb=e2Nc11Z_Tw@mail.gmail.com>
References: <CAK-C75R5BPUud6z2QxrFNNnh1wMFktu48O59oXi2bAUqZ0mYMw@mail.gmail.com>
	<4ED62AB1.4080208@web.de>
	<CAK-C75RzQhAOxdHZfqijb2G_p27GdVCSYu4nhcakwwW1dD+kuQ@mail.gmail.com>
	<CAK-C75TPsc1Lh07NNXFnmzW5ZQgSQbkfm1S2Vfb=e2Nc11Z_Tw@mail.gmail.com>
Message-ID: <CAK-C75S8UbGf+j7dU10ivFGSQvQFCVcg+MK3GiG5fJdbDOQ-PQ@mail.gmail.com>

On Wed, Nov 30, 2011 at 4:41 PM, Hugo Beauz?e-Luyssen
<beauze.h at gmail.com> wrote:
> On Wed, Nov 30, 2011 at 2:45 PM, Hugo Beauz?e-Luyssen
> <beauze.h at gmail.com> wrote:
>> On Wed, Nov 30, 2011 at 2:08 PM, Tobias G?ntner <fatbull at web.de> wrote:
>>> Am 30.11.2011 11:05, schrieb Hugo Beauz?e-Luyssen:
>>>> +/*
>>>> + ?Decodes a duration as defined by ISO 8601
>>>> + ?http://en.wikipedia.org/wiki/ISO_8601#Durations
>>>> + ?@param str A null-terminated string to convert
>>>
>>> Perhaps a short example of a valid input string would be useful, so
>>> readers can see at first glance what kind of data this function expects,
>>> without looking it up.
>>>
>>
>> Yep I'm going to write some tests in a this afternoon or tomorrow, but
>> I could add a link to the format description and an exemple in the
>> comments.
>>
>>>> + ?@return: The duration in seconds. 0 if an error occured.
>>>
>>> This function usually does not return 0 if the input is invalid.
>>>
>>
>> Yep i'm going to change the return type as a signed int 64, so -1 is an error.
>>
>>>> + */
>>>> +uint64_t str_iso8601_duration_to_seconds( const char *psz_duration )
>>>> +{
>>>> + ? ?bool ? ? ? ?timeDesignatorReached = false;
>>>> + ? ?double ? ? ?mul = 0;
>>>
>>> I think an integer should be fine.
>>>
>>
>> Well in the exemple file I've got
>> (http://www-itec.uni-klu.ac.at/ftp/datasets/mmsys12/Selected_VLC_Plugin_MPDs/BigBuckBunny_15s_DASH_VLC_Compatible_4-Representations_720p.mpd)
>> there are some decimal values, so I try to reduce the loss of
>> precision when converting by using doubles.
>>
>>>> + ? ?mtime_t ? ? res = 0;
>>>
>>> Shouldn't this type and the return type be the same?
>>>
>>
>> I was using mtime_t at first, it seems I forgot to change this one, my bad.
>>
>> Since the line below is a direct pointer dereferencement without any
>> prior check, this was intended, but a check could be added.
>>
>>>> +
>>>> + ? ?if ( *psz_duration++ != 'P' )
>>>> + ? ? ? ?return 0;
>>>> + ? ?for ( const char* nptr = psz_duration; *psz_duration; ++psz_duration )
>>>> + ? ?{
>>>> + ? ? ? ?switch( *psz_duration )
>>>> + ? ? ? ?{
>>>> + ? ? ? ? ? ?case 'M':
>>>> + ? ? ? ? ? ?{
>>>
>>> I find this a little confusing because letters normally follow numbers
>>> in the input, yet you somehow parse letters first. ;)
>>>
>>
>> That is absolutely right...
>>
>>>> + ? ? ? ? ? ? ? ?//M can mean month or minutes, if the 'T' flag has been reached.
>>>> + ? ? ? ? ? ? ? ?//We don't handle months though.
>>>> + ? ? ? ? ? ? ? ?if ( timeDesignatorReached == true )
>>>> + ? ? ? ? ? ? ? ? ? ?mul = 60.0;
>>>> + ? ? ? ? ? ? ? ?break ;
>>>> + ? ? ? ? ? ?}
>>>> + ? ? ? ? ? ?case 'Y':
>>>> + ? ? ? ? ? ?case 'W':
>>>> + ? ? ? ? ? ? ? ?break ; //Don't handle this duration.
>>>
>>> Maybe return an error here?
>>>
>>
>> I don't really know... warning should be mandatory, an error may be to
>> agressive...
>>
>>>> + ? ? ? ? ? ?case 'D':
>>>> + ? ? ? ? ? ? ? ?mul = 86400.0;
>>>> + ? ? ? ? ? ? ? ?break ;
>>>> + ? ? ? ? ? ?case 'T':
>>>> + ? ? ? ? ? ? ? ?timeDesignatorReached = true;
>>>> + ? ? ? ? ? ? ? ?break ;
>>>> + ? ? ? ? ? ?case 'H':
>>>> + ? ? ? ? ? ? ? ?mul = 3600.0;
>>>> + ? ? ? ? ? ? ? ?break ;
>>>> + ? ? ? ? ? ?case 'S':
>>>> + ? ? ? ? ? ? ? ?mul = 1.0;
>>>> + ? ? ? ? ? ? ? ?break ;
>>>> + ? ? ? ? ? ?default:
>>>> + ? ? ? ? ? ? ? ?continue ;
>>>> + ? ? ? ?}
>>>> + ? ? ? ?//It means we went trough some digits. So we must compute a number
>>>> + ? ? ? ?if ( nptr != psz_duration )
>>>> + ? ? ? ?{
>>>> + ? ? ? ? ? ?double tmp = strtold( nptr, NULL );
>>>
>>> I think you can achieve a slightly cleaner design if you utilize the
>>> second parameter of that function. The strtoX functions store the
>>> address of the first invalid character in there, so all you have to do
>>> is basically:
>>>
>>
>> Yep since I'm doing it in the wrong way, using the second parameter
>> makes a lot of sense now :D
>>
>>> 1. Call strtoul(psz_duration, &endptr, 10)
>>> 2. Check the character at *endptr
>>> 3. Repeat with psz_duration = endptr + 1
>>>
>>>> + ? ? ? ? ? ?res += (uint64_t)(tmp * mul);
>>>> + ? ? ? ? ? ?nptr = psz_duration;
>>>> + ? ? ? ?}
>>>> + ? ? ? ?nptr++;
>>>> + ? ?}
>>>> + ? ?return res;
>>>> +}
>>>
>>> Regards,
>>> Tobias
>>>
>>
>> Thanks for the review!
>> Regards,
>>
>
> Here is a new version, along with some unit tests.
>
> Regards,
>
>

Ping for review!

Regards,

-- 
Hugo Beauz?e-Luyssen


From remi at remlab.net  Thu Dec  1 12:42:11 2011
From: remi at remlab.net (=?UTF-8?Q?R=C3=A9mi_Denis-Courmont?=)
Date: Thu, 01 Dec 2011 12:42:11 +0100
Subject: [vlc-devel] [RFC,
 PATCH] configure: Do not stop configure if component is not found.
In-Reply-To: <4ED76347.7000405@unixsol.org>
References: <4ED6D93C.5010008@unixsol.org>
	<819cd0191398e013825a23e1c98bdefd@chewa.net>
	<4ED75B32.2000908@unixsol.org>
	<122f6dce5b7d51ea0e00381df60c8238@chewa.net>
	<20111201110840.GA11044@videolan.org>
	<4ED76347.7000405@unixsol.org>
Message-ID: <7ef8e63f1359b22a8f302edb2355891b@chewa.net>

On Thu, 01 Dec 2011 13:21:43 +0200, Georgi Chorbadzhiyski <gf at unixsol.org>
wrote:
>> How so? You might not want any A/52 decoding, no?

A/52 is one of the most common audio codec, perhaps the most common after
MPEG-2 Audio.

> My point exactly, configure should not require --disable-xxxxx just to
> silence an error that if xxxx is not found. If it is not found just mark
> it so and go on.

And who will read the warning among the flood of messages?

> Otherwise it is cat and mouse game. ./configure ... wait....aaaa an
error,
> whats the disable option (grep for it)...run configure --disable-abc ...
> wait....loop if an error is found.
> 
> And lets not forget that configure is huge and it is taking lots of time
> to run.

It's much less time than tracking why a VLC build does not work. And it's
also taking less time for us to support people who screwed their builds due
to missing dependencies.

You're not the one doing the build support on the forum.

-- 
R?mi Denis-Courmont
http://www.remlab.net/


From beauze.h at gmail.com  Thu Dec  1 12:43:02 2011
From: beauze.h at gmail.com (=?UTF-8?Q?Hugo_Beauz=C3=A9e=2DLuyssen?=)
Date: Thu, 1 Dec 2011 12:43:02 +0100
Subject: [vlc-devel] [Patch] Dash: Don't try to do something without a
 valid MPD manager
In-Reply-To: <20111128224245.GA29913@videolan.org>
References: <CAK-C75TOrnUGWki6HDVk0iJ0_xE1ZO4Swd8FGAoCraupTXQW8Q@mail.gmail.com>
	<4ed3d2c2.c27f980a.2078.19a3SMTPIN_ADDED@mx.google.com>
	<CAK-C75QjQ05NLFG25mrJ7drozaSdXkv-gGzQFPsuSCerLQ5V7Q@mail.gmail.com>
	<20111128224245.GA29913@videolan.org>
Message-ID: <CAK-C75Rsjt92O5ho3jmrturf-ugdzHk38hEY6HMgyA4+mKTDeg@mail.gmail.com>

On Mon, Nov 28, 2011 at 11:42 PM, Jean-Baptiste Kempf <jb at videolan.org> wrote:
> On Mon, Nov 28, 2011 at 10:54:07PM +0100, Hugo Beauz?e-Luyssen wrote :
>> much, but feels more natural in my opinion.
>
> Natural and C++ ?
>
> pff

Ping for review!

Regards,

-- 
Hugo Beauz?e-Luyssen


From beauze.h at gmail.com  Thu Dec  1 12:37:19 2011
From: beauze.h at gmail.com (=?UTF-8?Q?Hugo_Beauz=C3=A9e=2DLuyssen?=)
Date: Thu, 1 Dec 2011 12:37:19 +0100
Subject: [vlc-devel] [Patches] Dash
Message-ID: <CAK-C75QQttr4Mpb=B=YAVKgTpMrOeFpDVU9a4wf_6fRAHxFA9A@mail.gmail.com>

Hi,

As the patch has been submitted randomly in two other threads, I'll
post all the dash patches that I have in local to ease up the
follow-up.
The patches that are still being discussed are patches 6 through 8

Regards,


-- 
Hugo Beauz?e-Luyssen
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-dash-MPD-manager-factory-fixing-a-potential-memory-l.patch
Type: application/octet-stream
Size: 1205 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111201/8eff8e0d/attachment.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0002-dash-Don-t-hardcode-the-isLive-information.patch
Type: application/octet-stream
Size: 5549 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111201/8eff8e0d/attachment-0001.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0003-dash-Representation-Cosmetics.patch
Type: application/octet-stream
Size: 12329 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111201/8eff8e0d/attachment-0002.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0004-dash-Representation-Return-a-usable-bandwith-value.patch
Type: application/octet-stream
Size: 3845 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111201/8eff8e0d/attachment-0003.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0005-dash-Adding-a-RepresentationAttributesElements-to-re.patch
Type: application/octet-stream
Size: 17434 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111201/8eff8e0d/attachment-0004.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0006-dash-Removing-EOFException.patch
Type: application/octet-stream
Size: 12566 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111201/8eff8e0d/attachment-0005.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0007-dash-Don-t-try-to-do-anything-if-we-can-t-get-a-vali.patch
Type: application/octet-stream
Size: 10145 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111201/8eff8e0d/attachment-0006.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0008-dash-Removing-NullAdapationLogic.patch
Type: application/octet-stream
Size: 7126 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111201/8eff8e0d/attachment-0007.obj>

From rsatom at gmail.com  Thu Dec  1 12:53:28 2011
From: rsatom at gmail.com (Sergey Radionov)
Date: Thu, 1 Dec 2011 18:53:28 +0700
Subject: [vlc-devel] [PATCH] npapi: zero unused members of NPClass
Message-ID: <CADOtZwhOwBei-Bd_naA0cgjRay4YQR4pRCujKFydocyYoucHtA@mail.gmail.com>


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111201/79fcb2ad/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-npapi-zero-unused-members-of-NPClass.patch
Type: application/octet-stream
Size: 788 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111201/79fcb2ad/attachment.obj>

From jb at videolan.org  Thu Dec  1 12:54:45 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Thu, 1 Dec 2011 12:54:45 +0100
Subject: [vlc-devel] [PATCH] npapi: zero unused members of NPClass
In-Reply-To: <CADOtZwhOwBei-Bd_naA0cgjRay4YQR4pRCujKFydocyYoucHtA@mail.gmail.com>
References: <CADOtZwhOwBei-Bd_naA0cgjRay4YQR4pRCujKFydocyYoucHtA@mail.gmail.com>
Message-ID: <20111201115444.GA20539@videolan.org>

On Thu, Dec 01, 2011 at 06:53:28PM +0700, Sergey Radionov wrote :
> 

If that still fixes the 3880, perfect for me.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From rsatom at gmail.com  Thu Dec  1 12:56:12 2011
From: rsatom at gmail.com (Sergey Radionov)
Date: Thu, 1 Dec 2011 18:56:12 +0700
Subject: [vlc-devel] [PATCH] npapi: zero unused members of NPClass
In-Reply-To: <20111201115444.GA20539@videolan.org>
References: <CADOtZwhOwBei-Bd_naA0cgjRay4YQR4pRCujKFydocyYoucHtA@mail.gmail.com>
	<20111201115444.GA20539@videolan.org>
Message-ID: <CADOtZwgsWeUZCtsHgjs6snyYxLrUkp5ibrOGqa_N_fYRP6rmug@mail.gmail.com>

2011/12/1 Jean-Baptiste Kempf <jb at videolan.org>

> On Thu, Dec 01, 2011 at 06:53:28PM +0700, Sergey Radionov wrote :
> >
>
> If that still fixes the 3880, perfect for me.
>
yes, it is.


>
> Best regards,
>
> --
> Jean-Baptiste Kempf
> http://www.jbkempf.com/ - +33 672 704 734
> Sent from my Electronic Device
> _______________________________________________
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
> http://mailman.videolan.org/listinfo/vlc-devel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111201/1d54920d/attachment.html>

From alexey at alexeysokolov.co.cc  Thu Dec  1 13:14:15 2011
From: alexey at alexeysokolov.co.cc (Alexey Sokolov)
Date: Thu, 01 Dec 2011 19:14:15 +0700
Subject: [vlc-devel] [RFC,
 PATCH] configure: Do not stop configure if component is not found.
In-Reply-To: <7ef8e63f1359b22a8f302edb2355891b@chewa.net>
References: <4ED6D93C.5010008@unixsol.org>
	<819cd0191398e013825a23e1c98bdefd@chewa.net>
	<4ED75B32.2000908@unixsol.org>
	<122f6dce5b7d51ea0e00381df60c8238@chewa.net>
	<20111201110840.GA11044@videolan.org>
	<4ED76347.7000405@unixsol.org>
	<7ef8e63f1359b22a8f302edb2355891b@chewa.net>
Message-ID: <4ED76F97.7090405@alexeysokolov.co.cc>

01.12.2011 18:42, R?mi Denis-Courmont ?????:
> On Thu, 01 Dec 2011 13:21:43 +0200, Georgi Chorbadzhiyski <gf at unixsol.org>
> wrote:
>>> How so? You might not want any A/52 decoding, no?
> 
> A/52 is one of the most common audio codec, perhaps the most common after
> MPEG-2 Audio.
> 
>> My point exactly, configure should not require --disable-xxxxx just to
>> silence an error that if xxxx is not found. If it is not found just mark
>> it so and go on.
> 
> And who will read the warning among the flood of messages?
> 
Just show the table in the end of configure with the list of
enabled/disabled codecs.

>> Otherwise it is cat and mouse game. ./configure ... wait....aaaa an
> error,
>> whats the disable option (grep for it)...run configure --disable-abc ...
>> wait....loop if an error is found.
>>
>> And lets not forget that configure is huge and it is taking lots of time
>> to run.
> 
> It's much less time than tracking why a VLC build does not work. And it's
> also taking less time for us to support people who screwed their builds due
> to missing dependencies.
> 
> You're not the one doing the build support on the forum.
> 
Probably they can paste the same table from the end of configure,
besides ./configure cmdline?


-- 
Best regards,
Alexey "DarthGandalf" Sokolov


From rsatom at gmail.com  Thu Dec  1 13:39:29 2011
From: rsatom at gmail.com (Sergey Radionov)
Date: Thu, 1 Dec 2011 19:39:29 +0700
Subject: [vlc-devel] [PATCH] ActiveX,
	npapi win32: fixed mouse event handling in "stop" mode.
Message-ID: <CADOtZwgjEfeU0eoOVCp+N9WN3K6AEbZWJ=jYxyb-hUyqFGebSQ@mail.gmail.com>


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111201/64609856/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-ActiveX-npapi-win32-fixed-mouse-event-handling-in-st.patch
Type: application/octet-stream
Size: 1342 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111201/64609856/attachment.obj>

From komh78 at gmail.com  Thu Dec  1 14:09:51 2011
From: komh78 at gmail.com (KO Myung-Hun)
Date: Thu, 01 Dec 2011 22:09:51 +0900
Subject: [vlc-devel] [PATCH 1/9] Workaround for OS/2 iconv(),
 which 1. supports UCS-2 only not UTF-16 2. does not treat a string
 as a pathname by default
In-Reply-To: <201111261639.13172.remi@remlab.net>
References: <1322287529-7242-1-git-send-email-komh@chollian.net>	<1322287529-7242-2-git-send-email-komh@chollian.net>
	<201111261639.13172.remi@remlab.net>
Message-ID: <4ED77C9F.1050502@chollian.net>



R?mi Denis-Courmont wrote:
> Le samedi 26 novembre 2011 08:05:21 KO Myung-Hun, vous avez ?crit :
>> ---
>>  src/extras/libc.c |   44 ++++++++++++++++++++++++++++++++++++++++++++
>>  1 files changed, 44 insertions(+), 0 deletions(-)
>>
>> diff --git a/src/extras/libc.c b/src/extras/libc.c
>> index 5b028a6..ff40aa0 100644
>> --- a/src/extras/libc.c
>> +++ b/src/extras/libc.c
>> @@ -37,6 +37,15 @@
>>  #   include <iconv.h>
>>  #endif
>>
>> +#if defined(__OS2__) && defined(__INNOTEK_LIBC__)
>> +#   include <uconv.h>
>> +
>> +typedef struct os2_iconv_t
>> +{
>> +    UconvObject from;
>> +} os2_iconv_t;
>> +#endif
>> +
>>  /*************************************************************************
>> **** * Local conversion routine from ISO_6937 to UTF-8 charset. Support for
>> this * is still missing in libiconv, hence multiple operating systems lack
>> it. @@ -326,7 +335,42 @@ vlc_iconv_t vlc_iconv_open( const char *tocode,
>> const char *fromcode ) return (vlc_iconv_t)(-2);
>>  #endif
>>  #if defined(HAVE_ICONV)
>> +# if defined(__OS2__) && defined(__INNOTEK_LIBC__)
>> +    char tocode_ucs2[] = "UCS-2LE";
>> +    char fromcode_ucs2[] = "UCS-2LE";
> 
> This should be const char *. There is no point in copying the strings on the 
> stack every time.
> 


Ooops. This generates the following warning.

-----
  CC     extras/libc.lo
extras/libc.c: In function 'vlc_iconv_open':
extras/libc.c:346: warning: passing argument 1 of 'strncpy' discards
qualifiers
from pointer target type
f:/lang/gcc/usr/include/string.h:104: note: expected 'char * restrict' but
argum
ent is of type 'const char *'
extras/libc.c:352: warning: passing argument 1 of 'strncpy' discards
qualifiers
from pointer target type
f:/lang/gcc/usr/include/string.h:104: note: expected 'char * restrict' but
argum
ent is of type 'const char *'
-----

And it causes SIGSEGV.

So I attach my original sources, again.

-- 
KO Myung-Hun

Using Mozilla SeaMonkey 2.0.14
Under OS/2 Warp 4 for Korean with FixPak #15
On AMD ThunderBird 1GHz with 512 MB RAM

Korean OS/2 User Community : http://www.ecomstation.co.kr

-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: 0001-Workaround-for-OS-2-iconv.patch
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111201/dda9a4f7/attachment.ksh>

From komh78 at gmail.com  Thu Dec  1 14:10:42 2011
From: komh78 at gmail.com (KO Myung-Hun)
Date: Thu, 01 Dec 2011 22:10:42 +0900
Subject: [vlc-devel] [PATCH 6/9] Add missing getaddrinfo(),
 freeaddrinfo(), getnameinfo() and gai_strerror() for OS/2
In-Reply-To: <201111291655.40201.remi@remlab.net>
References: <1322287529-7242-1-git-send-email-komh@chollian.net>	<201111261640.41937.remi@remlab.net>
	<4ED4D8A3.80808@chollian.net> <201111291655.40201.remi@remlab.net>
Message-ID: <4ED77CD2.5090707@chollian.net>



R?mi Denis-Courmont wrote:
> I'm neither allowed nor willing to merge GPL stuff into LibVLC as of now.
> 

How about this ?

-- 
KO Myung-Hun

Using Mozilla SeaMonkey 2.0.14
Under OS/2 Warp 4 for Korean with FixPak #15
On AMD ThunderBird 1GHz with 512 MB RAM

Korean OS/2 User Community : http://www.ecomstation.co.kr

-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: 0006-Add-missing-getaddrinfo-freeaddrinfo-getnameinfo-and.patch
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111201/24823d54/attachment.ksh>

From hjohn at xs4all.nl  Thu Dec  1 14:12:22 2011
From: hjohn at xs4all.nl (John Hendrikx)
Date: Thu, 01 Dec 2011 14:12:22 +0100
Subject: [vlc-devel] [PATCH] Added libvlc_video_get_spu_delay and
	libvlc_video_set_spu_delay.
Message-ID: <4ED77D36.30105@xs4all.nl>

Added the debug prints as requested by Alexey Sokolov.

 From 38077d512825375cd27bb40670f56113ce190c93 Mon Sep 17 00:00:00 2001
From: John Hendrikx <hjohn at xs4all.nl>
Date: Thu, 1 Dec 2011 14:08:33 +0100
Subject: [PATCH] Added libvlc_video_get_spu_delay and 
libvlc_video_set_spu_delay

Allows for controlling the subtitle delay through libvlc.
---
 include/vlc/libvlc_media_player.h |   24 +++++++++++++++++++++++
 lib/libvlc.sym                    |    2 +
 lib/video.c                       |   38 
+++++++++++++++++++++++++++++++++++++
 3 files changed, 64 insertions(+), 0 deletions(-)

diff --git a/include/vlc/libvlc_media_player.h 
b/include/vlc/libvlc_media_player.h
index a2b5748..152a95d 100644
--- a/include/vlc/libvlc_media_player.h
+++ b/include/vlc/libvlc_media_player.h
@@ -1039,6 +1039,30 @@ LIBVLC_API int libvlc_video_set_spu( 
libvlc_media_player_t *p_mi, unsigned i_spu
 LIBVLC_API int libvlc_video_set_subtitle_file( libvlc_media_player_t 
*p_mi, const char *psz_subtitle );
 
 /**
+ * Get the current subtitle delay. Positive values means subtitles are 
being
+ * displayed later, negative values earlier.
+ *
+ * \param p_mi media player
+ * \return time (in microseconds) the display of subtitles is being delayed
+ * \version LibVLC 1.2.0 or later
+ */
+LIBVLC_API int64_t libvlc_video_get_spu_delay( libvlc_media_player_t 
*p_mi );
+
+/**
+ * Set the subtitle delay. This affects the timing of when the subtitle 
will
+ * be displayed. Positive values result in subtitles being displayed later,
+ * while negative values will result in subtitles being displayed earlier.
+ *
+ * The subtitle delay will be reset to zero each time the media changes.
+ *
+ * \param p_mi media player
+ * \param i_delay time (in microseconds) the display of subtitles 
should be delayed
+ * \return 0 on success, -1 on error
+ * \version LibVLC 1.2.0 or later
+ */
+LIBVLC_API int libvlc_video_set_spu_delay( libvlc_media_player_t *p_mi, 
int64_t i_delay );
+
+/**
  * Get the description of available titles.
  *
  * \param p_mi the media player
diff --git a/lib/libvlc.sym b/lib/libvlc.sym
index 6582a96..45327d5 100644
--- a/lib/libvlc.sym
+++ b/lib/libvlc.sym
@@ -189,6 +189,7 @@ libvlc_video_get_marquee_string
 libvlc_video_get_scale
 libvlc_video_get_spu
 libvlc_video_get_spu_count
+libvlc_video_get_spu_delay
 libvlc_video_get_spu_description
 libvlc_video_get_teletext
 libvlc_video_get_title_description
@@ -212,6 +213,7 @@ libvlc_video_set_marquee_string
 libvlc_video_set_mouse_input
 libvlc_video_set_scale
 libvlc_video_set_spu
+libvlc_video_set_spu_delay
 libvlc_video_set_subtitle_file
 libvlc_video_set_teletext
 libvlc_video_set_track
diff --git a/lib/video.c b/lib/video.c
index b659c88..163bca3 100644
--- a/lib/video.c
+++ b/lib/video.c
@@ -365,6 +365,44 @@ int libvlc_video_set_subtitle_file( 
libvlc_media_player_t *p_mi,
     return b_ret;
 }
 
+int64_t libvlc_video_get_spu_delay( libvlc_media_player_t *p_mi )
+{
+    input_thread_t *p_input_thread = libvlc_get_input_thread( p_mi );
+    int64_t val = 0;
+
+    if( p_input_thread )
+    {
+        val = var_GetTime( p_input_thread, "spu-delay" );
+        vlc_object_release( p_input_thread );
+    }
+    else
+    {
+        libvlc_printerr( "No active input" );
+    }
+
+    return val;
+}
+
+int libvlc_video_set_spu_delay( libvlc_media_player_t *p_mi,
+                                int64_t i_delay )
+{
+    input_thread_t *p_input_thread = libvlc_get_input_thread( p_mi );
+    int ret = -1;
+
+    if( p_input_thread )
+    {
+        var_SetTime( p_input_thread, "spu-delay", i_delay );
+        vlc_object_release( p_input_thread );
+        ret = 0;
+    }
+    else
+    {
+        libvlc_printerr( "No active input" );
+    }
+
+    return ret;
+}
+
 libvlc_track_description_t *
         libvlc_video_get_title_description( libvlc_media_player_t *p_mi )
 {
-- 
1.7.4.1




From yann.lehir at gmail.com  Thu Dec  1 14:18:03 2011
From: yann.lehir at gmail.com (yann LE HIR)
Date: Thu, 1 Dec 2011 14:18:03 +0100
Subject: [vlc-devel] Getting the size of a video
In-Reply-To: <201111301955.51485.remi@remlab.net>
References: <CAGvrb0SGQdXPSvhTXtOQzqjuu9pf7L=wh0FET05GsHLGdQSmDw@mail.gmail.com>
	<201111301955.51485.remi@remlab.net>
Message-ID: <CAGvrb0S72_QAO_tCa+Ncp0P1GdSm92CpwzxmCBdF6BJD4ZAM6A@mail.gmail.com>

2011/11/30 R?mi Denis-Courmont <remi at remlab.net>:
> ? ? ? ?Hello,
>
> Le mercredi 30 novembre 2011 15:58:52 yann LE HIR, vous avez ?crit :
>> libvlc_media_get_tracks_info() works fine on files but sometimes gives
>> bad data with streams (TS streams with mpeg2 video in that case).
>> However, X11 video_output module get the good size when creating the
>> window.
>
> Sure. The video output gets the picture format so that it can setup picture
> buffers correctly.
>
>> My software embed a vlc_media_player in a GTK Window (with
>> set_xwindow()). And I would like to resize the widget that displays
>> VLC when I load a file.
>
> Such a design is intrinsically broken. Some media can have multiple video
> tracks of different resolutions, simultaneously or consecutively.
>
> Even if you ignore those corner cases, there will be picture resize glitches.

Anyway, I find odd that there is no method to get the size of the
native video resolution.

>
> If you want the to set windows dimensions according to the native video
> resolution, you need to create a dedicated window provider plug-in. Only then
> you can get the correct dimension all the time, and create/resize the window
> before the pipeline starts displaying pictures.

I was afraid you'd say that.
I was trying to avoid the pain of reading/understanding all of the
video_output process, but I guess this is not an option.

>> The video_output module used in that case is xvideo, that
>> automatically resize the vlc window to the size of the xwindow given
>> and replace the size values in the config of the video output thread.
>
> The X11 plugins (XVideo, GLX, X11 and GL/EGL) all request a window of the
> correct size. They don't actually resize the window. That is the
> responsibility of the window provider.
>
>> I managed to make it work, but I had to modify the way the xvideo
>> module works, and to hack around in libvlc_video_get_size() to access
>> the private part of video output thread.
>
> You cannot do that. Besides breaking the encapsulation principle, this patch
> violates the threading model surrounding the video output thread. In other
> words, sooner or later, it will crash the process or return corrupt values.

Anyway,
thank you for your time.


From remi at remlab.net  Thu Dec  1 14:52:59 2011
From: remi at remlab.net (=?UTF-8?Q?R=C3=A9mi_Denis-Courmont?=)
Date: Thu, 01 Dec 2011 14:52:59 +0100
Subject: [vlc-devel] Getting the size of a video
In-Reply-To: <CAGvrb0S72_QAO_tCa+Ncp0P1GdSm92CpwzxmCBdF6BJD4ZAM6A@mail.gmail.com>
References: <CAGvrb0SGQdXPSvhTXtOQzqjuu9pf7L=wh0FET05GsHLGdQSmDw@mail.gmail.com>
	<201111301955.51485.remi@remlab.net>
	<CAGvrb0S72_QAO_tCa+Ncp0P1GdSm92CpwzxmCBdF6BJD4ZAM6A@mail.gmail.com>
Message-ID: <b46981c7aadf8f6dff007325758f5808@chewa.net>

On Thu, 1 Dec 2011 14:18:03 +0100, yann LE HIR <yann.lehir at gmail.com>
wrote:
> Anyway, I find odd that there is no method to get the size of the
> native video resolution.

Eh? There are two ways: track infos and the size getter.

The first one should be working already. The second one is waiting for
someone to fix it.

-- 
R?mi Denis-Courmont
http://www.remlab.net/


From beauze.h at gmail.com  Thu Dec  1 16:02:46 2011
From: beauze.h at gmail.com (=?UTF-8?Q?Hugo_Beauz=C3=A9e=2DLuyssen?=)
Date: Thu, 1 Dec 2011 16:02:46 +0100
Subject: [vlc-devel] [Patches] Dash
In-Reply-To: <CAK-C75QQttr4Mpb=B=YAVKgTpMrOeFpDVU9a4wf_6fRAHxFA9A@mail.gmail.com>
References: <CAK-C75QQttr4Mpb=B=YAVKgTpMrOeFpDVU9a4wf_6fRAHxFA9A@mail.gmail.com>
Message-ID: <CAK-C75SPp=18hNdReif+AbxuZR8xT2zwL-9jSYCj1dw4JsBGew@mail.gmail.com>

On Thu, Dec 1, 2011 at 12:37 PM, Hugo Beauz?e-Luyssen
<beauze.h at gmail.com> wrote:
> Hi,
>
> As the patch has been submitted randomly in two other threads, I'll
> post all the dash patches that I have in local to ease up the
> follow-up.
> The patches that are still being discussed are patches 6 through 8
>
> Regards,
>

Hello again, please consider this patch instead of the previous 0004.

Regards,


-- 
Hugo Beauz?e-Luyssen
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0004-dash-Representation-Return-a-usable-bandwith-value.patch
Type: text/x-patch
Size: 4074 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111201/ed770b24/attachment.bin>

From gf at unixsol.org  Thu Dec  1 17:43:16 2011
From: gf at unixsol.org (Georgi Chorbadzhiyski)
Date: Thu,  1 Dec 2011 18:43:16 +0200
Subject: [vlc-devel] [PATCH 1/3] m4: Add VLC_ARG_ENABLE function.
In-Reply-To: <1322757798-19213-1-git-send-email-gf@unixsol.org>
References: <1322757798-19213-1-git-send-email-gf@unixsol.org>
Message-ID: <1322757798-19213-2-git-send-email-gf@unixsol.org>

It does the same as AC_ARG_ENABLE but checks for if disable_all
variable is set.
---

Somebody with better m4 knowledge should check this function. It
probably could be written better.

 m4/with_pkg.m4 |   13 +++++++++++++
 1 files changed, 13 insertions(+), 0 deletions(-)

diff --git a/m4/with_pkg.m4 b/m4/with_pkg.m4
index 39439b7..340c695 100644
--- a/m4/with_pkg.m4
+++ b/m4/with_pkg.m4
@@ -31,6 +31,19 @@ dnl Prepare a --with-variable-prefix triggered check for module,
 dnl disable by default.
 dnl
 
+dnl VLC_ARG_ENABLE have same parameters as AC_ARG_ENABLE but checks for disable_all
+AC_DEFUN([VLC_ARG_ENABLE],
+[
+    var=enable_$(echo $1 | tr .+- ___)
+    eval xvar=\$${var}
+    AS_IF(
+      [test x"${disable_all}" = "xyes" -a x"${xvar}" != "xyes" ],
+      [eval ${var}=no
+       AC_MSG_NOTICE([$1 disabled. Enable it with --enable-$1])],
+      [AC_ARG_ENABLE([$1], [$2], [$3])]
+    )
+])
+
 AC_DEFUN([PKG_WITH_MODULES],
 [
 AC_REQUIRE([PKG_PROG_PKG_CONFIG])
-- 
1.7.5.1



From gf at unixsol.org  Thu Dec  1 17:43:15 2011
From: gf at unixsol.org (Georgi Chorbadzhiyski)
Date: Thu,  1 Dec 2011 18:43:15 +0200
Subject: [vlc-devel] [PATCH 0/3]i configure: Add --disable-all.
Message-ID: <1322757798-19213-1-git-send-email-gf@unixsol.org>

The following patch series add --disable-all switch to configure and
show enabled plugins when configure is over.

This allows me to compile minimal vlc with only:
./configure \
  --disable-all \
  --enable-vlc \
  --enable-a52 \
  --enable-x264 \
  --enable-avcodec \
  --enable-avformat \
  --enable-swscale \
  --enable-sout \
  --enable-dvbpsi

There are some plugins in configure that are always enabled. For example
./configure --disable-all results in the following plugins:

 access_shm, dynamicoverlay and fb

I'll fix these (add configure options for them) if the current aproach is
acceptable.

Georgi Chorbadzhiyski (3):
  m4: Add VLC_ARG_ENABLE function.
  configure.ac: Replace AC_ARG_ENABLE with VLC_ARG_ENABLE.
  configure: Show enabled plugins.

 configure.ac   |  222 +++++++++++++++++++++++++++++--------------------------
 m4/with_pkg.m4 |   15 ++++-
 2 files changed, 131 insertions(+), 106 deletions(-)

-- 
1.7.5.1



From gf at unixsol.org  Thu Dec  1 17:43:17 2011
From: gf at unixsol.org (Georgi Chorbadzhiyski)
Date: Thu,  1 Dec 2011 18:43:17 +0200
Subject: [vlc-devel] [PATCH 2/3] configure.ac: Replace AC_ARG_ENABLE with
	VLC_ARG_ENABLE.
In-Reply-To: <1322757798-19213-1-git-send-email-gf@unixsol.org>
References: <1322757798-19213-1-git-send-email-gf@unixsol.org>
Message-ID: <1322757798-19213-3-git-send-email-gf@unixsol.org>

This allows --disable-all to work.
---
 configure.ac   |  219 +++++++++++++++++++++++++++++--------------------------
 m4/with_pkg.m4 |    2 +-
 2 files changed, 116 insertions(+), 105 deletions(-)

diff --git a/configure.ac b/configure.ac
index 09766ac..a44f447 100644
--- a/configure.ac
+++ b/configure.ac
@@ -764,7 +764,7 @@ AS_IF([test "${enable_non_utf8}" != "no"], [
 
 
 dnl Check for dbus
-AC_ARG_ENABLE(dbus,
+VLC_ARG_ENABLE(dbus,
   [AS_HELP_STRING([--enable-dbus],
     [compile D-Bus message bus support (default enabled)])])
 case "${SYS}" in
@@ -777,7 +777,7 @@ then
       VLC_ADD_LIBS([libvlccore],[$DBUS_LIBS])
       VLC_ADD_CFLAGS([libvlccore],[$DBUS_CFLAGS])
       dnl Check for dbus control interface
-        AC_ARG_ENABLE(dbus-control,
+        VLC_ARG_ENABLE(dbus-control,
           [AS_HELP_STRING([--disable-dbus-control],
             [D-Bus control interface (default enabled)])])
         if test "${enable_dbus_control}" != "no"
@@ -787,7 +787,7 @@ then
           VLC_ADD_CFLAGS([dbus],[$DBUS_CFLAGS])
         fi
       dnl Check for Telepathy
-        AC_ARG_ENABLE(telepathy,
+        VLC_ARG_ENABLE(telepathy,
           AS_HELP_STRING([--enable-telepathy],[Telepathy Presence plugin through DBus(default enabled)]))
         if test "${enable_telepathy}" != "no"; then
           VLC_ADD_PLUGIN([telepathy])
@@ -799,7 +799,9 @@ then
         VLC_ADD_LIBS([inhibit],[$DBUS_LIBS])
         VLC_ADD_CFLAGS([inhibit],[$DBUS_CFLAGS])
     ],
+    [AS_IF([test "${enable_dbus_control}" = "yes"],
     [AC_MSG_ERROR([${DBUS_PKG_ERRORS}.])]
+    )]
   )
 fi
 ;;
@@ -1518,9 +1520,18 @@ if test "${enable_optimize_memory}" = "yes"; then
 fi
 
 dnl
+dnl Disable all plugins except explicitly enabled via --enable-XXXX
+dnl
+AC_ARG_ENABLE(all,
+  [AS_HELP_STRING([--disable-all],
+    [disable all plugins except explicitly enabled via --enable-xxx.])],
+  [disable_all="yes"],
+  [disable_all="no"])
+
+dnl
 dnl Allow running as root (useful for people running on embedded platforms)
 dnl
-AC_ARG_ENABLE(run-as-root,
+VLC_ARG_ENABLE(run-as-root,
   [AS_HELP_STRING([--enable-run-as-root],
     [allow running VLC as root (default disabled)])])
 AS_IF([test "${enable_run_as_root}" = "yes"],[
@@ -1531,7 +1542,7 @@ AS_IF([test "${enable_run_as_root}" = "yes"],[
 dnl
 dnl Stream output
 dnl
-AC_ARG_ENABLE(sout,
+VLC_ARG_ENABLE(sout,
   [AS_HELP_STRING([--disable-sout],
     [disable streaming output (default enabled)])])
 AS_IF([test "${enable_sout}" != "no"], [
@@ -1540,7 +1551,7 @@ AS_IF([test "${enable_sout}" != "no"], [
 AM_CONDITIONAL(ENABLE_SOUT, [test "${enable_sout}" != "no"])
 
 dnl Lua modules
-AC_ARG_ENABLE(lua,
+VLC_ARG_ENABLE(lua,
   [AS_HELP_STRING([--disable-lua],
     [disable LUA scripting support (default enabled)])])
 if test "${enable_lua}" != "no"
@@ -1587,7 +1598,7 @@ AM_CONDITIONAL(BUILD_LUA, [test "${have_lua}" = "yes"])
 dnl
 dnl HTTP daemon
 dnl
-AC_ARG_ENABLE(httpd,
+VLC_ARG_ENABLE(httpd,
   [AS_HELP_STRING([--disable-httpd],
     [disable the built-in HTTP server (default enabled)])])
 if test "${enable_httpd}" != "no"
@@ -1600,7 +1611,7 @@ AM_CONDITIONAL(BUILD_HTTPD, [test "${enable_httpd}" != "no"])
 dnl
 dnl VideoLAN manager
 dnl
-AC_ARG_ENABLE(vlm,
+VLC_ARG_ENABLE(vlm,
   [AS_HELP_STRING([--disable-vlm],
     [disable the stream manager (default enabled)])],,
   [enable_vlm="${enable_sout}"])
@@ -1622,7 +1633,7 @@ EXTEND_HELP_STRING([Input plugins:])
 dnl
 dnl libproxy support
 dnl
-AC_ARG_ENABLE(libproxy,
+VLC_ARG_ENABLE(libproxy,
   [AS_HELP_STRING([--enable-libproxy],[support libproxy (default auto)])])
 AS_IF([test "${enable_libproxy}" != "no"], [
   PKG_CHECK_MODULES(LIBPROXY, libproxy-1.0, [
@@ -1639,7 +1650,7 @@ AS_IF([test "${enable_libproxy}" != "no"], [
 dnl
 dnl  live555 input
 dnl
-AC_ARG_ENABLE(live555,
+VLC_ARG_ENABLE(live555,
   [AS_HELP_STRING([--enable-live555],
     [enable RTSP input through live555 (default enabled)])])
 AS_IF([test "${enable_live555}" != "no" -a -n "${CXX}"], [
@@ -1736,7 +1747,7 @@ dnl
 PKG_ENABLE_MODULES_VLC([DC1394], [], [libraw1394 >= 2.0.1 libdc1394-2 >= 2.1.0], [dc1394 access module], [auto])
 PKG_ENABLE_MODULES_VLC([DV], [access_dv], [libraw1394 >= 2.0.1 libavc1394 >= 0.5.3], [DV input module], [auto])
 
-AC_ARG_ENABLE(linsys,
+VLC_ARG_ENABLE(linsys,
   [AS_HELP_STRING([--enable-linsys],
     [Linux Linear Systems Ltd. SDI and HD-SDI input cards (default enabled)])])
 if test "$SYS" = "linux" -a "${enable_linsys}" != "no"; then
@@ -1757,7 +1768,7 @@ PKG_ENABLE_MODULES_VLC([DVDREAD], [], [dvdread], [dvdread input module], [auto])
 dnl
 dnl libdvdnav plugin
 dnl
-AC_ARG_ENABLE(dvdnav,
+VLC_ARG_ENABLE(dvdnav,
   [AS_HELP_STRING([--disable-dvdnav],
     [disable DVD navigation with libdvdnav (default auto)])])
 if test "${enable_dvdnav}" != "no"
@@ -1778,7 +1789,7 @@ fi
 dnl
 dnl  Windows DirectShow access module
 dnl
-AC_ARG_ENABLE(dshow,
+VLC_ARG_ENABLE(dshow,
   [AS_HELP_STRING([--disable-dshow],
     [support DirectShow (default auto)])])
 if test "${enable_dshow}" != "no"
@@ -1808,7 +1819,7 @@ PKG_ENABLE_MODULES_VLC([OPENCV], [opencv_example opencv_wrapper], [opencv], (Ope
 dnl
 dnl  libsmbclient plugin
 dnl
-AC_ARG_ENABLE(smb,
+VLC_ARG_ENABLE(smb,
   [AS_HELP_STRING([--disable-smb], [disable SMB/CIFS support (default auto)])])
 if test "${enable_smb}" != "no"; then
   AC_CHECK_HEADERS(libsmbclient.h,
@@ -1823,7 +1834,7 @@ fi
 dnl
 dnl sftp access support
 dnl
-AC_ARG_ENABLE(sftp,
+VLC_ARG_ENABLE(sftp,
   [AS_HELP_STRING([--enable-sftp],
     [support SFTP file transfer via libssh2 (default disabled)])])
 if test "${enable_sftp}" = "yes"; then
@@ -1836,9 +1847,9 @@ fi
 dnl
 dnl  Video4Linux 2
 dnl
-AC_ARG_ENABLE(libv4l2, [AS_HELP_STRING([--disable-libv4l2],
+VLC_ARG_ENABLE(libv4l2, [AS_HELP_STRING([--disable-libv4l2],
   [disable userspace V4L2 library (default auto)])])
-AC_ARG_ENABLE(pvr, [AS_HELP_STRING([--enable-pvr],
+VLC_ARG_ENABLE(pvr, [AS_HELP_STRING([--enable-pvr],
   [support PVR V4L2 cards (default disabled)])])
 have_v4l2="no"
 AC_CHECK_HEADERS([linux/videodev2.h sys/videoio.h], [
@@ -1861,7 +1872,7 @@ AM_CONDITIONAL(HAVE_V4L2, [test "${have_v4l2}" != "no"])
 dnl
 dnl special access module for Blackmagic SDI cards
 dnl
-AC_ARG_ENABLE(decklink,
+VLC_ARG_ENABLE(decklink,
   [AS_HELP_STRING([--disable-decklink],
     [disable Blackmagic DeckLink SDI input (default auto)])])
 AC_ARG_WITH(decklink_sdk,
@@ -1892,7 +1903,7 @@ PKG_ENABLE_MODULES_VLC([GNOMEVFS], [access_gnomevfs], [gnome-vfs-2.0], [GnomeVFS
 dnl
 dnl  VCDX modules
 dnl
-AC_ARG_ENABLE(vcdx,
+VLC_ARG_ENABLE(vcdx,
   [AS_HELP_STRING([--enable-vcdx],
     [navigate VCD with libvcdinfo (default disabled)])])
 if test "${enable_vcdx}" = "yes"
@@ -1911,11 +1922,11 @@ fi
 dnl
 dnl  Built-in CD-DA and VCD module
 dnl
-AC_ARG_ENABLE(vcd,
+VLC_ARG_ENABLE(vcd,
   [AS_HELP_STRING([--disable-vcd],
     [disable built-in VCD and CD-DA support (default enabled)])])
 
-AC_ARG_ENABLE(libcddb,
+VLC_ARG_ENABLE(libcddb,
   [AS_HELP_STRING([--disable-libcddb],
     [disable CDDB for Audio CD (default enabled)])])
 
@@ -1989,7 +2000,7 @@ AM_CONDITIONAL([HAVE_LINUX_DVB], [test "$ac_cv_linux_s2api" = "yes"])
 dnl
 dnl  Screen capture module
 dnl
-AC_ARG_ENABLE(screen,
+VLC_ARG_ENABLE(screen,
   [AS_HELP_STRING([--enable-screen],
     [disable screen capture (default enabled)])])
 if test "${enable_screen}" != "no"; then
@@ -2009,7 +2020,7 @@ fi
 dnl
 dnl  Real RTSP plugin
 dnl
-AC_ARG_ENABLE(realrtsp,
+VLC_ARG_ENABLE(realrtsp,
   [  --enable-realrtsp       Real RTSP module (default disabled)])
 if test "${enable_realrtsp}" = "yes"; then
   VLC_ADD_PLUGIN([access_realrtsp])
@@ -2017,7 +2028,7 @@ fi
 
 dnl
 dnl MacOS eyeTV
-AC_ARG_ENABLE(macosx-eyetv,
+VLC_ARG_ENABLE(macosx-eyetv,
   [  --enable-macosx-eyetv   Mac OS X EyeTV (TNT Tuner) module (default enabled on Mac OS X)])
 if test "x${enable_macosx_eyetv}" != "xno" &&
   (test "${SYS}" = "darwin" || test "${enable_macosx_eyetv}" = "yes")
@@ -2028,7 +2039,7 @@ fi
 
 dnl
 dnl QTKit
-AC_ARG_ENABLE(macosx-qtkit,
+VLC_ARG_ENABLE(macosx-qtkit,
   [  --enable-macosx-qtkit Mac OS X qtcapture (video) and qtsound (audio) module (default enabled on Mac OS X)])
 if test "x${enable_macosx_qtkit}" != "xno" &&
   (test "${SYS}" = "darwin" || test "${enable_macosx_qtkit}" = "yes")
@@ -2056,7 +2067,7 @@ AM_CONDITIONAL(HAVE_DVBPSI, [test "${have_dvbpsi}" = "yes"])
 dnl
 dnl  GME demux plugin
 dnl
-AC_ARG_ENABLE(gme,
+VLC_ARG_ENABLE(gme,
   [AS_HELP_STRING([--enable-gme],
     [use Game Music Emu (default auto)])])
 AS_IF([test "${enable_gme}" != "no"], [
@@ -2074,7 +2085,7 @@ AS_IF([test "${enable_gme}" != "no"], [
 dnl
 dnl  SIDPlay plugin
 dnl
-AC_ARG_ENABLE(sid,
+VLC_ARG_ENABLE(sid,
     [AS_HELP_STRING([--enable-sid],[C64 sid demux support (default auto)])])
 AS_IF([test "${enable_sid}" != "no"], [
     PKG_CHECK_MODULES(SID, [libsidplay2], [
@@ -2115,7 +2126,7 @@ fi
 dnl
 dnl  matroska demux plugin
 dnl
-AC_ARG_ENABLE(mkv,
+VLC_ARG_ENABLE(mkv,
   [AS_HELP_STRING([--disable-mkv],
     [do not use libmatroska (default auto)])])
 if test "${enable_mkv}" != "no" -a "${CXX}" != ""; then
@@ -2170,7 +2181,7 @@ fi
 dnl
 dnl  modplug demux plugin
 dnl
-AC_ARG_ENABLE(mod,
+VLC_ARG_ENABLE(mod,
   [AS_HELP_STRING([--disable-mod],
     [do not use libmodplug (default auto)])])
 if test "${enable_mod}" != "no" ; then
@@ -2188,7 +2199,7 @@ fi
 dnl
 dnl  mpc demux plugin
 dnl
-AC_ARG_ENABLE(mpc,
+VLC_ARG_ENABLE(mpc,
   [AS_HELP_STRING([--disable-mpc], [do not use libmpcdec (default auto)])])
 if test "${enable_mpc}" != "no"
 then
@@ -2209,7 +2220,7 @@ EXTEND_HELP_STRING([Codec plugins:])
 dnl
 dnl wmafixed plugin
 dnl
-AC_ARG_ENABLE(wma-fixed,
+VLC_ARG_ENABLE(wma-fixed,
   [  --enable-wma-fixed      libwma-fixed module (default disabled)])
 if test "${enable_wma_fixed}" = "yes"
 then
@@ -2219,7 +2230,7 @@ fi
 dnl
 dnl shine fixed point mp3 encoder
 dnl
-AC_ARG_ENABLE(shine,
+VLC_ARG_ENABLE(shine,
   [  --enable-shine          shine mp3 encoding module (default disabled)])
 if test "${enable_shine}" = "yes"
 then
@@ -2229,7 +2240,7 @@ fi
 dnl
 dnl openmax il codec plugin
 dnl
-AC_ARG_ENABLE(omxil,
+VLC_ARG_ENABLE(omxil,
   [  --enable-omxil          openmax il codec module (default disabled)])
 if test "${enable_omxil}" = "yes"
 then
@@ -2240,7 +2251,7 @@ fi
 dnl
 dnl iomx codec plugin
 dnl
-AC_ARG_ENABLE(iomx,
+VLC_ARG_ENABLE(iomx,
   [  --enable-iomx           iomx codec module (default disabled)])
 if test "${enable_iomx}" = "yes"
 then
@@ -2261,7 +2272,7 @@ fi
 dnl
 dnl CrystalHD codec plugin
 dnl
-AC_ARG_ENABLE(crystalhd,
+VLC_ARG_ENABLE(crystalhd,
   [  --enable-crystalhd      crystalhd codec plugin (default auto)])
 if test "${enable_crystalhd}" != "no"; then
     AC_CHECK_HEADER(libcrystalhd/libcrystalhd_if.h, [
@@ -2290,7 +2301,7 @@ fi
 dnl
 dnl  mad plugin
 dnl
-AC_ARG_ENABLE(mad,
+VLC_ARG_ENABLE(mad,
   [  --enable-mad            libmad module (default enabled)])
 if test "${enable_mad}" != "no"
 then
@@ -2347,7 +2358,7 @@ then
 fi
 
 
-AC_ARG_ENABLE(merge-ffmpeg,
+VLC_ARG_ENABLE(merge-ffmpeg,
 [  --enable-merge-ffmpeg   merge FFmpeg-based plugins (default disabled)],, [
   enable_merge_ffmpeg="no"
 ])
@@ -2363,7 +2374,7 @@ VLC_RESTORE_FLAGS
 dnl
 dnl  avcodec decoder/encoder plugin
 dnl
-AC_ARG_ENABLE(avcodec,
+VLC_ARG_ENABLE(avcodec,
 [  --enable-avcodec        libavcodec codec (default enabled)])
 AS_IF([test "${enable_avcodec}" != "no"], [
   PKG_CHECK_MODULES(AVCODEC,[libavcodec >= 52.25.0 libavutil],
@@ -2391,7 +2402,7 @@ AS_IF([test "${enable_avcodec}" != "no"], [
 dnl
 dnl libva needs avcodec
 dnl
-AC_ARG_ENABLE(libva,
+VLC_ARG_ENABLE(libva,
   [  --enable-libva          VAAPI GPU decoding support (libVA) (default auto)])
 
 AS_IF([test "${enable_libva}" = "yes" -a "${have_avcodec}" != "yes" ], [
@@ -2427,7 +2438,7 @@ AS_IF([test "${enable_libva}" != "no"], [
 dnl
 dnl dxva2 needs avcodec
 dnl
-AC_ARG_ENABLE(dxva2,
+VLC_ARG_ENABLE(dxva2,
   [  --enable-dxva2          DxVA2 GPU decoding support (default auto)])
 
 AS_IF([test "${enable_dxva2}" != "no"], [
@@ -2461,7 +2472,7 @@ AS_IF([test "${enable_dxva2}" != "no"], [
 dnl
 dnl stream_out switcher needs libavcodec
 dnl
-AC_ARG_ENABLE(switcher,
+VLC_ARG_ENABLE(switcher,
   [  --enable-switcher       Stream-out switcher plugin (default disabled)])
 AS_IF([test "${enable_switcher}" = "yes"], [
   AS_IF([test "x${have_avcodec}" = "xyes"], [
@@ -2477,7 +2488,7 @@ dnl
 dnl  avformat demuxer/muxer plugin
 dnl
 
-AC_ARG_ENABLE(avformat,
+VLC_ARG_ENABLE(avformat,
 [  --enable-avformat       libavformat containers (default enabled)],, [
   enable_avformat="${have_avcodec}"
 ])
@@ -2511,7 +2522,7 @@ dnl
 dnl  swscale image scaling and conversion plugin
 dnl
 
-AC_ARG_ENABLE(swscale,
+VLC_ARG_ENABLE(swscale,
   AS_HELP_STRING([--enable-swscale],[libswscale image scaling and conversion
           (default enabled)]))
 if test "${enable_swscale}" != "no"
@@ -2538,7 +2549,7 @@ dnl
 dnl  postproc plugin
 dnl
 
-AC_ARG_ENABLE(postproc,
+VLC_ARG_ENABLE(postproc,
 [  --enable-postproc       libpostproc image post-processing (default enabled)])
 if test "${enable_postproc}" != "no"
 then
@@ -2560,7 +2571,7 @@ fi
 dnl
 dnl  faad decoder plugin
 dnl
-AC_ARG_ENABLE(faad,
+VLC_ARG_ENABLE(faad,
 [  --enable-faad           faad codec (default disabled)])
 if test "${enable_faad}" = "yes"
 then
@@ -2613,7 +2624,7 @@ PKG_ENABLE_MODULES_VLC([TWOLAME], [], [twolame], [MPEG Audio Layer 2 encoder], [
 dnl
 dnl  QuickTime plugin
 dnl
-AC_ARG_ENABLE(quicktime,
+VLC_ARG_ENABLE(quicktime,
   [  --enable-quicktime      QuickTime module (deprecated)])
 if test "${enable_quicktime}" = "yes"; then
   if test "${SYS}" = "mingw32"; then
@@ -2629,7 +2640,7 @@ fi
 dnl
 dnl  Real plugin
 dnl
-AC_ARG_ENABLE(real,
+VLC_ARG_ENABLE(real,
   [  --enable-real           Real video decoder module (default disabled)])
 if test "${enable_real}" = "yes"; then
   VLC_ADD_PLUGIN([realvideo])
@@ -2638,7 +2649,7 @@ fi
 dnl
 dnl A52/AC3 decoder plugin
 dnl
-AC_ARG_ENABLE(a52,
+VLC_ARG_ENABLE(a52,
   [  --enable-a52            A/52 support with liba52 (default enabled)])
 if test "${enable_a52}" != "no"
 then
@@ -2735,7 +2746,7 @@ PKG_ENABLE_MODULES_VLC([VORBIS], [], [ogg vorbis >= 1.1 vorbisenc >= 1.1], [Vorb
 dnl
 dnl  Tremor plugin
 dnl
-AC_ARG_ENABLE(tremor,
+VLC_ARG_ENABLE(tremor,
   [  --enable-tremor         Tremor decoder support (default disabled)])
 if test "${enable_tremor}" = "yes"
 then
@@ -2781,7 +2792,7 @@ PKG_ENABLE_MODULES_VLC([SCHROEDINGER], [], [schroedinger-1.0 >= 1.0.10], [dirac
 dnl
 dnl  PNG decoder module
 dnl
-AC_ARG_ENABLE(png,
+VLC_ARG_ENABLE(png,
   [  --enable-png            PNG support (default enabled)])
 if test "${enable_png}" != "no"; then
 AC_CHECK_HEADERS(png.h, [
@@ -2799,7 +2810,7 @@ AM_CONDITIONAL(BUILD_OSDMENU, [test "${enable_png}" != "no"])
 dnl
 dnl H264 encoder plugin (using libx264)
 dnl
-AC_ARG_ENABLE(x264,
+VLC_ARG_ENABLE(x264,
   [  --enable-x264           H264 encoding support with libx264 (default enabled)])
 if test "${enable_x264}" != "no"; then
   AC_ARG_WITH(x264-tree,
@@ -2866,10 +2877,10 @@ dnl vbi decoder plugin (using libzbvi)
 dnl telx module
 dnl uncompatible
 dnl
-AC_ARG_ENABLE(zvbi,
+VLC_ARG_ENABLE(zvbi,
   AS_HELP_STRING([--enable-zvbi],[VBI (inc. Teletext) decoding support with
                   libzvbi (default enabled)]))
-AC_ARG_ENABLE(telx,
+VLC_ARG_ENABLE(telx,
   AS_HELP_STRING([--enable-telx],[Teletext decoding module (conflicting with
                   zvbi) (default enabled if zvbi is absent)]))
 
@@ -2897,7 +2908,7 @@ AS_IF( [test "${enable_telx}" != "no" ],[
 dnl
 dnl libass subtitle rendering module
 dnl
-AC_ARG_ENABLE(libass,
+VLC_ARG_ENABLE(libass,
   [  --enable-libass         Subtitle support using libass (default enabled)])
 AS_IF( [test "${enable_libass}" != "no"], [
   PKG_CHECK_MODULES(LIBASS, [libass >= 0.9.8],
@@ -2923,7 +2934,7 @@ PKG_ENABLE_MODULES_VLC([KATE], [], [kate >= 0.3.0], [kate codec], [auto])
 dnl
 dnl  tiger rendering for kate decoder plugin
 dnl
-AC_ARG_ENABLE(tiger,
+VLC_ARG_ENABLE(tiger,
 [  --enable-tiger          Tiger rendering library for Kate streams (default auto)])
 AS_IF([test "${enable_tiger}" != "no"], [
   PKG_CHECK_MODULES(TIGER,[tiger >= 0.3.1], [
@@ -2963,7 +2974,7 @@ dnl PKG_ENABLE_MODULES_VLC([GLES2], [], [glesv2], [OpenGL ES v2 support], [auto]
 dnl OpenGL ES 1: depends on EGL 1.0 and is currently broken
 dnl PKG_ENABLE_MODULES_VLC([GLES1], [], [glesv1_cm], [OpenGL ES v1 support], [auto])
 
-AC_ARG_ENABLE(egl,
+VLC_ARG_ENABLE(egl,
   [  --enable-egl            OpenGL support through EGL (default disabled)],, [
   enable_egl="no"
 ])
@@ -2984,7 +2995,7 @@ AC_CHECK_HEADERS(X11/Xlib.h)
 dnl
 dnl  X C Bindings modules
 dnl
-AC_ARG_ENABLE(xcb,
+VLC_ARG_ENABLE(xcb,
   [  --enable-xcb            X11 support with XCB (default enabled)],, [
   AS_IF([test "${SYS}" != "mingw32" -a "${SYS}" != "mingwce" -a "${SYS}" != "darwin" -a "${SYS}" != "symbian"], [
     enable_xcb="yes"
@@ -2992,11 +3003,11 @@ AC_ARG_ENABLE(xcb,
     enable_xcb="no"
   ])
 ])
-AC_ARG_ENABLE(xvideo,
+VLC_ARG_ENABLE(xvideo,
   [  --enable-xvideo         XVideo support (default enabled)],, [
     enable_xvideo="$enable_xcb"
 ])
-AC_ARG_ENABLE(glx,
+VLC_ARG_ENABLE(glx,
   [  --enable-glx            OpenGL support through GLX (default enabled)],, [
   enable_glx="$enable_xcb"
 ])
@@ -3050,9 +3061,9 @@ AM_CONDITIONAL([HAVE_XCB], [test "${have_xcb}" = "yes"])
 dnl
 dnl  SDL module
 dnl
-AC_ARG_ENABLE(sdl,
+VLC_ARG_ENABLE(sdl,
   [  --enable-sdl            SDL support (default enabled)])
-AC_ARG_ENABLE(sdl-image,
+VLC_ARG_ENABLE(sdl-image,
   [  --enable-sdl-image      SDL image support (default enabled)])
 if test "${enable_sdl}" != "no"
 then
@@ -3084,7 +3095,7 @@ fi
 
 dnl
 dnl Mac Vout
-AC_ARG_ENABLE(macosx-vout,
+VLC_ARG_ENABLE(macosx-vout,
   [  --enable-macosx-vout    Mac OS X video output module (default enabled on Mac OS X)])
 if test "x${enable_macosx_vout}" != "xno" &&
   (test "${SYS}" = "darwin" || test "${enable_macosx_vout}" = "yes")
@@ -3098,11 +3109,11 @@ fi
 dnl
 dnl  freetype module
 dnl
-AC_ARG_ENABLE(freetype,
+VLC_ARG_ENABLE(freetype,
   [  --enable-freetype       freetype support   (default auto)])
-AC_ARG_ENABLE(fribidi,
+VLC_ARG_ENABLE(fribidi,
   [  --enable-fribidi        fribidi support    (default auto)])
-AC_ARG_ENABLE(fontconfig,
+VLC_ARG_ENABLE(fontconfig,
   [  --enable-fontconfig     fontconfig support (default auto)])
 
 if test "${enable_freetype}" != "no"; then
@@ -3150,7 +3161,7 @@ fi
 dnl
 dnl  QuartzText vout module (iOS/Mac OS)
 dnl
-AC_ARG_ENABLE(macosx-quartztext,
+VLC_ARG_ENABLE(macosx-quartztext,
   [  --enable-macosx-quartztext   Mac OS X quartz text module (default enabled on Mac OS X)])
 if test "x${enable_macosx_quartztext}" != "xno" &&
   (test "${SYS}" = "darwin" || test "${enable_macosx_quartztext}" = "yes")
@@ -3167,7 +3178,7 @@ PKG_ENABLE_MODULES_VLC([SVG], [], [librsvg-2.0 >= 2.9.0], [SVG rendering library
 dnl
 dnl  android surface module
 dnl
-AC_ARG_ENABLE(android-surface,
+VLC_ARG_ENABLE(android-surface,
   [  --enable-android-surface   Android Surface video output module (default disabled)])
 if test "${enable_android_surface}" = "yes"; then
   if test "${HAVE_ANDROID}" = "1"; then
@@ -3180,7 +3191,7 @@ fi
 dnl
 dnl  iOS vout module
 dnl
-AC_ARG_ENABLE(ios-vout,
+VLC_ARG_ENABLE(ios-vout,
   [  --enable-ios-vout    iOS video output module (default disabled)])
 if test "${enable_ios_vout}" = "yes"
 then
@@ -3198,7 +3209,7 @@ then
   VLC_ADD_PLUGIN([panoramix])
 fi
 
-AC_ARG_ENABLE(directx,
+VLC_ARG_ENABLE(directx,
   [  --enable-directx        Win32 DirectX support (default enabled on Win32)])
 if test "${enable_directx}" != "no"; then
   if test "${SYS}" = "mingw32" -o "${SYS}" = "mingwce"; then
@@ -3230,7 +3241,7 @@ fi
 dnl
 dnl  Windows Direct2D plugin
 dnl
-AC_ARG_ENABLE(direct2d,
+VLC_ARG_ENABLE(direct2d,
   [  --enable-direct2d       Win7/VistaPU Direct2D support (default auto on Win32)],, [
   AS_IF([test "${SYS}" != "mingw32" -a "${SYS}" != "mingwce"], [
     enable_direct2d="no"
@@ -3248,7 +3259,7 @@ AS_IF([test "${enable_direct2d}" != "no"], [
 dnl
 dnl  win32 GDI plugin
 dnl
-AC_ARG_ENABLE(wingdi,
+VLC_ARG_ENABLE(wingdi,
   [  --enable-wingdi         Win32 GDI module (default enabled on Win32)])
 if test "${enable_wingdi}" != "no"; then
   if test "${SYS}" = "mingw32"; then
@@ -3276,7 +3287,7 @@ dnl  DirectFB module
 dnl  try to find using: 1 - given location; 2 - directfb-config; 3 - pkg-config
 dnl  TODO: support for static linking
 dnl
-AC_ARG_ENABLE(directfb,
+VLC_ARG_ENABLE(directfb,
   [  --enable-directfb       DirectFB support (default disabled)])
 AC_ARG_WITH(directfb,
   [  --with-directfb=PATH    path to DirectFB headers and libraries])
@@ -3338,7 +3349,7 @@ fi
 dnl
 dnl  AA plugin
 dnl
-AC_ARG_ENABLE(aa,
+VLC_ARG_ENABLE(aa,
   [  --enable-aa             aalib output (default disabled)])
 if test "${enable_aa}" = "yes"
 then
@@ -3361,7 +3372,7 @@ PKG_ENABLE_MODULES_VLC([CACA], [], [caca >= 0.99.beta14], [libcaca output],[auto
 dnl
 dnl  OS/2 KVA plugin
 dnl
-AC_ARG_ENABLE(kva,
+VLC_ARG_ENABLE(kva,
   [AS_HELP_STRING([--enable-kva],
     [support the K Video Accelerator KVA (default enabled on OS/2)])],, [
   AS_IF([test "$SYS" = "os2"], [
@@ -3390,7 +3401,7 @@ EXTEND_HELP_STRING([Audio plugins:])
 dnl
 dnl  Pulseaudio module
 dnl
-AC_ARG_ENABLE(pulse,
+VLC_ARG_ENABLE(pulse,
   [AS_HELP_STRING([--enable-pulse],
     [use the PulseAudio client library (default auto)])])
 have_pulse="no"
@@ -3420,7 +3431,7 @@ AM_CONDITIONAL([HAVE_PULSE], [test "${have_pulse}" = "yes"])
 dnl
 dnl  ALSA module
 dnl
-AC_ARG_ENABLE(alsa,
+VLC_ARG_ENABLE(alsa,
   [AS_HELP_STRING([--enable-alsa],
     [support the Advanced Linux Sound Architecture (default auto)])],, [
   AS_IF([test "$SYS" = "linux" -a "${have_pulse}" = "no"], [
@@ -3442,7 +3453,7 @@ AM_CONDITIONAL([HAVE_ALSA], [test "${have_alsa}" = "yes"])
 dnl
 dnl  Open Sound System module
 dnl
-AC_ARG_ENABLE(oss,
+VLC_ARG_ENABLE(oss,
   [AS_HELP_STRING([--enable-oss],
     [support the Open Sound System OSS (default enabled on BSD)])],, [
   AS_IF([test "$SYS" = "mingw32" -o "$SYS" = "mingwce" -o "$SYS" = "linux"], [
@@ -3470,7 +3481,7 @@ PKG_ENABLE_MODULES_VLC([PORTAUDIO], [], [portaudio-2.0], [Portaudio library supp
 dnl
 dnl  win32 waveOut plugin
 dnl
-AC_ARG_ENABLE(waveout,
+VLC_ARG_ENABLE(waveout,
   [  --enable-waveout        Win32 waveOut module (default enabled on Win32)])
 if test "${enable_waveout}" != "no"; then
   if test "${SYS}" = "mingw32"; then
@@ -3485,7 +3496,7 @@ fi
 dnl
 dnl  CoreAudio plugin
 dnl
-AC_ARG_ENABLE(macosx-audio,
+VLC_ARG_ENABLE(macosx-audio,
   [  --enable-macosx-audio   Mac OS X audio module (default enabled on MacOS X)])
 if test "x${enable_macosx_audio}" != "xno" &&
   (test "${SYS}" = "darwin" || test "${enable_macosx_audio}" = "yes")
@@ -3499,7 +3510,7 @@ fi
 dnl
 dnl  AudioQueue plugin
 dnl
-AC_ARG_ENABLE(audioqueue,
+VLC_ARG_ENABLE(audioqueue,
   [  --enable-audioqueue     AudioQueue audio module (default disabled)])
 if test "${enable_audioqueue}" = "yes"
 then
@@ -3515,7 +3526,7 @@ PKG_ENABLE_MODULES_VLC([JACK], [jack access_jack], [jack], [JACK audio I/O modul
 dnl
 dnl  OpenSLES Android
 dnl
-AC_ARG_ENABLE(opensles,
+VLC_ARG_ENABLE(opensles,
   [  --enable-opensles       Android OpenSL ES audio module (default disabled)])
 if test "${HAVE_ANDROID}" = "1"; then
   if test "${enable_opensles}" = "yes"; then
@@ -3533,7 +3544,7 @@ PKG_ENABLE_MODULES_VLC([SAMPLERATE], [], [samplerate], [Resampler with libsample
 dnl
 dnl  OS/2 KAI plugin
 dnl
-AC_ARG_ENABLE(kai,
+VLC_ARG_ENABLE(kai,
   [AS_HELP_STRING([--enable-kai],
     [support the K Audio Interface KAI (default enabled on OS/2)])],, [
   AS_IF([test "$SYS" = "os2"], [
@@ -3562,7 +3573,7 @@ EXTEND_HELP_STRING([Interface plugins:])
 dnl
 dnl Hildon UI
 dnl
-AC_ARG_ENABLE(hildon,
+VLC_ARG_ENABLE(hildon,
   [  --enable-hildon         Hildon touchscreen UI (default disabled)])
 AS_IF([test "${enable_hildon}" = "yes"], [
   PKG_CHECK_MODULES(HILDON, [hildon-1], [
@@ -3588,7 +3599,7 @@ AM_CONDITIONAL(BUILD_HILDON, [test "${enable_hildon}" = "yes"])
 dnl
 dnl QT 4
 dnl
-AC_ARG_ENABLE(qt4,
+VLC_ARG_ENABLE(qt4,
   [  --enable-qt4            Qt 4 support (default enabled) ],, [
   AS_IF([test "${SYS}" = "darwin"], [enable_qt4=no])
 ])
@@ -3655,7 +3666,7 @@ AS_IF([test "x$enable_qt4" = "xno"], [
 dnl
 dnl Skins2 module
 dnl
-AC_ARG_ENABLE(skins2,
+VLC_ARG_ENABLE(skins2,
   [AS_HELP_STRING([--enable-skins2], [skins interface module (default auto)])])
 AS_IF([test "${enable_skins2}" != "no"], [
   have_skins_deps="yes"
@@ -3700,7 +3711,7 @@ AS_IF([test "${enable_skins2}" != "no"], [
 ])
 AM_CONDITIONAL(BUILD_SKINS, [test "${enable_skins2}" = "yes"])
 
-AC_ARG_ENABLE(libtar,
+VLC_ARG_ENABLE(libtar,
   [  --enable-libtar         libtar support for skins2 (default auto)])
 AS_IF([test "${enable_skins2}" = "yes" && test "${enable_libtar}" != "no"], [
   AC_CHECK_HEADERS(libtar.h, [
@@ -3711,7 +3722,7 @@ AS_IF([test "${enable_skins2}" = "yes" && test "${enable_libtar}" != "no"], [
 dnl
 dnl  MacOS X gui module
 dnl
-AC_ARG_ENABLE(macosx,
+VLC_ARG_ENABLE(macosx,
   [  --enable-macosx         Mac OS X gui support (default enabled on Mac OS X)])
 if test "x${enable_macosx}" != "xno" -a "${SYS}" = "darwin"
 then
@@ -3740,7 +3751,7 @@ fi
 dnl
 dnl  MacOS X dialor provider
 dnl
-AC_ARG_ENABLE(macosx-dialog-provider,
+VLC_ARG_ENABLE(macosx-dialog-provider,
   [  --enable-macosx-dialog-provider Mac OS X dialog module (default enabled on Mac OS X)])
 if test "x${enable_macosx_dialog_provider}" != "xno" &&
   (test "${SYS}" = "darwin" || test "${enable_macosx_dialog_provider}" = "yes")
@@ -3752,7 +3763,7 @@ fi
 dnl
 dnl  ncurses module
 dnl
-AC_ARG_ENABLE(ncurses,
+VLC_ARG_ENABLE(ncurses,
 [AS_HELP_STRING([--disable-ncurses],[ncurses text-based interface (default auto)])])
 if test "${enable_ncurses}" != "no"; then
   VLC_SAVE_FLAGS
@@ -3778,7 +3789,7 @@ fi
 dnl
 dnl  XOSD plugin
 dnl
-AC_ARG_ENABLE(xosd,
+VLC_ARG_ENABLE(xosd,
   [  --enable-xosd           xosd interface support (default disabled)])
 if test "${enable_xosd}" = "yes"
 then
@@ -3793,7 +3804,7 @@ fi
 dnl
 dnl Framebuffer (overlay) plugin
 dnl
-AC_ARG_ENABLE(fbosd,
+VLC_ARG_ENABLE(fbosd,
   [  --enable-fbosd          fbosd interface support (default disabled)])
 if test "${enable_fbosd}" = "yes"
 then
@@ -3805,7 +3816,7 @@ fi
 dnl
 dnl  Lirc plugin
 dnl
-AC_ARG_ENABLE(lirc,
+VLC_ARG_ENABLE(lirc,
   [  --enable-lirc           lirc support (default disabled)])
 if test "${enable_lirc}" = "yes"
 then
@@ -3821,7 +3832,7 @@ EXTEND_HELP_STRING([Visualisations and Video filter plugins:])
 dnl
 dnl Visualisation plugin
 dnl
-AC_ARG_ENABLE(visual,
+VLC_ARG_ENABLE(visual,
   [  --enable-visual         visualisation plugin (default enabled)])
 if test "${enable_visual}" != "no"
 then
@@ -3836,7 +3847,7 @@ PKG_ENABLE_MODULES_VLC([GOOM], [], [libgoom2], [goom visualization plugin], [aut
 dnl
 dnl libprojectM visualization plugin
 dnl
-AC_ARG_ENABLE(projectm,
+VLC_ARG_ENABLE(projectm,
   [  --enable-projectm       projectM visualization plugin (default enabled)])
 AS_IF([test "${enable_projectm}" != "no"],
   [
@@ -3857,7 +3868,7 @@ AS_IF([test "${enable_projectm}" != "no"],
 dnl
 dnl  AtmoLight (homemade Philips Ambilight clone)
 dnl
-AC_ARG_ENABLE(atmo,
+VLC_ARG_ENABLE(atmo,
 AS_HELP_STRING([--disable-atmo],[AtmoLight (homemade Philips Ambilight clone)
                 (default enabled)]),, [enable_atmo="yes"])
 AS_IF([test "${enable_atmo}" != no], [
@@ -3898,7 +3909,7 @@ PKG_ENABLE_MODULES_VLC([LIBXML2], [xml], [libxml-2.0 >= 2.5], [libxml2 support],
 dnl
 dnl libgcrypt
 dnl
-AC_ARG_ENABLE(libgcrypt,
+VLC_ARG_ENABLE(libgcrypt,
   [  --disable-libgcrypt     gcrypt support (default enabled)])
 AS_IF([test "${enable_libgcrypt}" != "no"], [
   AC_CHECK_DECL([GCRYCTL_SET_THREAD_CBS], [
@@ -3924,7 +3935,7 @@ AM_CONDITIONAL([HAVE_GCRYPT], [test "${have_libgcrypt}" = "yes"])
 dnl
 dnl TLS/SSL
 dnl
-AC_ARG_ENABLE(gnutls,
+VLC_ARG_ENABLE(gnutls,
   [  --enable-gnutls         GNU TLS TLS/SSL support (default enabled)])
 
 AS_IF([test "${have_libgcrypt}" != "yes"], [
@@ -3958,7 +3969,7 @@ AS_IF([test -f "/etc/maemo_version"], [
 dnl
 dnl Taglib plugin
 dnl
-AC_ARG_ENABLE(taglib,
+VLC_ARG_ENABLE(taglib,
   [AS_HELP_STRING([--disable-taglib],
     [do not use TagLib (default enabled)])])
 AS_IF([test "${enable_taglib}" != "no"], [
@@ -3974,7 +3985,7 @@ AS_IF([test "${enable_taglib}" != "no"], [
 dnl
 dnl update checking system
 dnl
-AC_ARG_ENABLE(update-check,
+VLC_ARG_ENABLE(update-check,
   [  --enable-update-check   update checking system (default disabled)])
 if test "${enable_update_check}" = "yes"
 then
@@ -3990,7 +4001,7 @@ fi
 dnl
 dnl Growl notification plugin
 dnl
-AC_ARG_ENABLE(growl,
+VLC_ARG_ENABLE(growl,
   [  --enable-growl          growl notification plugin (default disabled)],,
   [enable_growl=no])
 AS_IF([test "${enable_growl}" != "no"], [
@@ -4011,7 +4022,7 @@ PKG_ENABLE_MODULES_VLC([NOTIFY], [], [libnotify gtk+-2.0], [libnotify notificati
 dnl
 dnl media library
 dnl
-AC_ARG_ENABLE(media-library,
+VLC_ARG_ENABLE(media-library,
     [  --enable-media-library  media library (default disabled)])
 AS_IF([test "${enable_media_library}" = "yes"], [
   AC_DEFINE([MEDIA_LIBRARY], 1, [Define if you want to use the VLC media library])
@@ -4080,7 +4091,7 @@ dnl
 dnl  DLL loader copied from MPlayer copied from somewhere else (WINE ?)
 dnl
 loader=false
-AC_ARG_ENABLE(loader,
+VLC_ARG_ENABLE(loader,
   AS_HELP_STRING([--enable-loader],[build DLL loader for ELF i386 platforms
           (default disabled)]))
 AM_CONDITIONAL(LOADER, [test "${enable_loader}" = "yes"])
@@ -4097,11 +4108,11 @@ EXTEND_HELP_STRING([Components:])
 dnl
 dnl  the VLC binary
 dnl
-AC_ARG_ENABLE(vlc,
+VLC_ARG_ENABLE(vlc,
   [  --enable-vlc            build the VLC media player (default enabled)])
 AM_CONDITIONAL(BUILD_VLC, [test "${enable_vlc}" != "no"])
 
-AC_ARG_ENABLE(macosx-vlc-app,
+VLC_ARG_ENABLE(macosx-vlc-app,
   [  --enable-macosx-vlc-app build the VLC media player (default enabled on Mac OS X)])
 AM_CONDITIONAL(BUILD_MACOSX_VLC_APP, [test "${enable_macosx_vlc_app}" != "no" &&
     (test "${SYS}" = "darwin" || test "${enable_macosx_vlc_app}" = "yes") ])
diff --git a/m4/with_pkg.m4 b/m4/with_pkg.m4
index 340c695..5369d1c 100644
--- a/m4/with_pkg.m4
+++ b/m4/with_pkg.m4
@@ -60,7 +60,7 @@ m4_case(def_arg,
             [yes],[m4_pushdef([with_without], [--disable-]with_arg)],
             [m4_pushdef([with_without],[--enable-]with_arg)])
 
-AC_ARG_ENABLE(with_arg,
+VLC_ARG_ENABLE(with_arg,
      AS_HELP_STRING(with_without, description[ @<:@default=]def_arg[@:>@]),,
     [AS_TR_SH([enable_]with_arg)=def_arg])
 
-- 
1.7.5.1



From gf at unixsol.org  Thu Dec  1 17:43:18 2011
From: gf at unixsol.org (Georgi Chorbadzhiyski)
Date: Thu,  1 Dec 2011 18:43:18 +0200
Subject: [vlc-devel] [PATCH 3/3] configure: Show enabled plugins.
In-Reply-To: <1322757798-19213-1-git-send-email-gf@unixsol.org>
References: <1322757798-19213-1-git-send-email-gf@unixsol.org>
Message-ID: <1322757798-19213-4-git-send-email-gf@unixsol.org>

---
 configure.ac |    3 ++-
 1 files changed, 2 insertions(+), 1 deletions(-)

diff --git a/configure.ac b/configure.ac
index a44f447..5ea8e5e 100644
--- a/configure.ac
+++ b/configure.ac
@@ -4297,7 +4297,8 @@ libvlc configuration
 version               : ${VERSION}
 system                : ${SYS}
 architecture          : ${ARCH}
-optimizations         : ${enable_optimizations}"
+optimizations         : ${enable_optimizations}
+plugins               : ${PLUGINS}"
 if test "${enable_vlc}" != "no"; then
 echo "vlc aliases           :${ALIASES}"
 else
-- 
1.7.5.1



From alexey at alexeysokolov.co.cc  Thu Dec  1 17:47:35 2011
From: alexey at alexeysokolov.co.cc (Alexey Sokolov)
Date: Thu, 01 Dec 2011 23:47:35 +0700
Subject: [vlc-devel] [PATCH 3/3] configure: Show enabled plugins.
In-Reply-To: <1322757798-19213-4-git-send-email-gf@unixsol.org>
References: <1322757798-19213-1-git-send-email-gf@unixsol.org>
	<1322757798-19213-4-git-send-email-gf@unixsol.org>
Message-ID: <4ED7AFA7.5060007@alexeysokolov.co.cc>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

01.12.2011 23:43, Georgi Chorbadzhiyski ?????:
> --- configure.ac |    3 ++- 1 files changed, 2 insertions(+), 1
> deletions(-)
> 
> diff --git a/configure.ac b/configure.ac index a44f447..5ea8e5e
> 100644 --- a/configure.ac +++ b/configure.ac @@ -4297,7 +4297,8 @@
> libvlc configuration version               : ${VERSION} system
> : ${SYS} architecture          : ${ARCH} -optimizations         :
> ${enable_optimizations}" +optimizations         :
> ${enable_optimizations} +plugins               : ${PLUGINS}" if
> test "${enable_vlc}" != "no"; then echo "vlc aliases
> :${ALIASES}" else
Need to show both enabled and disabled plugins, so that user can see
what capabilities he will miss in his future VLC if he types "make".

- -- 
Best regards,
Alexey "DarthGandalf" Sokolov
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.17 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/
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=uQNu
-----END PGP SIGNATURE-----


From remi at remlab.net  Thu Dec  1 17:57:51 2011
From: remi at remlab.net (=?iso-8859-1?q?R=E9mi?= Denis-Courmont)
Date: Thu, 1 Dec 2011 18:57:51 +0200
Subject: [vlc-devel] [PATCH 0/3]i configure: Add --disable-all.
In-Reply-To: <1322757798-19213-1-git-send-email-gf@unixsol.org>
References: <1322757798-19213-1-git-send-email-gf@unixsol.org>
Message-ID: <201112011857.52214.remi@remlab.net>

Le jeudi 1 d?cembre 2011 18:43:15 Georgi Chorbadzhiyski, vous avez ?crit :
> The following patch series add --disable-all switch to configure and
> show enabled plugins when configure is over.
> 
> This allows me to compile minimal vlc with only:
> ./configure \
>   --disable-all \
>   --enable-vlc \
>   --enable-a52 \
>   --enable-x264 \
>   --enable-avcodec \
>   --enable-avformat \
>   --enable-swscale \
>   --enable-sout \
>   --enable-dvbpsi
> 
> There are some plugins in configure that are always enabled. For example
> ./configure --disable-all results in the following plugins:
> 
>  access_shm, dynamicoverlay and fb
> 
> I'll fix these (add configure options for them) if the current aproach is
> acceptable.

First, most builders are complaining that they are far too many options. There 
are about a hundred always or automatically enabled plugins. There is no way 
we are adding options for them.

Second, there is little to no point in adding options for plugins that don't 
add dependencies. If you don't want them, don't package them in your binary 
build rules.

Third, I have been trying to move as much logic away from configure into the 
Makefile as possible to reduce recompilation churn. I am totally against going 
backward to configure.

Forth, it's a feature, not a bug that you need to explicitly disable some 
important stuff. I don't see the point except to help people shoot themselves 
in the feet.

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From gf at unixsol.org  Thu Dec  1 17:58:36 2011
From: gf at unixsol.org (Georgi Chorbadzhiyski)
Date: Thu, 01 Dec 2011 18:58:36 +0200
Subject: [vlc-devel] [PATCH 3/3] configure: Show enabled plugins.
In-Reply-To: <4ED7AFA7.5060007@alexeysokolov.co.cc>
References: <1322757798-19213-1-git-send-email-gf@unixsol.org>
	<1322757798-19213-4-git-send-email-gf@unixsol.org>
	<4ED7AFA7.5060007@alexeysokolov.co.cc>
Message-ID: <4ED7B23C.2090503@unixsol.org>

Around 12/01/2011 06:47 PM, Alexey Sokolov scribbled:
> 01.12.2011 23:43, Georgi Chorbadzhiyski ?????:
>> --- configure.ac |    3 ++- 1 files changed, 2 insertions(+), 1
>> deletions(-)
>>
>> diff --git a/configure.ac b/configure.ac index a44f447..5ea8e5e
>> 100644 --- a/configure.ac +++ b/configure.ac @@ -4297,7 +4297,8 @@
>> libvlc configuration version               : ${VERSION} system
>> : ${SYS} architecture          : ${ARCH} -optimizations         :
>> ${enable_optimizations}" +optimizations         :
>> ${enable_optimizations} +plugins               : ${PLUGINS}" if
>> test "${enable_vlc}" != "no"; then echo "vlc aliases
>> :${ALIASES}" else
> Need to show both enabled and disabled plugins, so that user can see
> what capabilities he will miss in his future VLC if he types "make".

Something like the attached on top of previous patches.

-- 
Georgi Chorbadzhiyski
http://georgi.unixsol.org/
-------------- next part --------------
diff --git a/configure.ac b/configure.ac
index 5ea8e5e..de52cbf 100644
--- a/configure.ac
+++ b/configure.ac
@@ -4298,7 +4298,8 @@ version               : ${VERSION}
 system                : ${SYS}
 architecture          : ${ARCH}
 optimizations         : ${enable_optimizations}
-plugins               : ${PLUGINS}"
+disabled plugins      :${PLUGINS_DISABLED}
+enabled plugins       : ${PLUGINS}"
 if test "${enable_vlc}" != "no"; then
 echo "vlc aliases           :${ALIASES}"
 else
diff --git a/m4/with_pkg.m4 b/m4/with_pkg.m4
index 5369d1c..292ad4a 100644
--- a/m4/with_pkg.m4
+++ b/m4/with_pkg.m4
@@ -39,8 +39,10 @@ AC_DEFUN([VLC_ARG_ENABLE],
     AS_IF(
       [test x"${disable_all}" = "xyes" -a x"${xvar}" != "xyes" ],
       [eval ${var}=no
-       AC_MSG_NOTICE([$1 disabled. Enable it with --enable-$1])],
-      [AC_ARG_ENABLE([$1], [$2], [$3])]
+       PLUGINS_DISABLED="$PLUGINS_DISABLED $1"
+       AC_MSG_NOTICE([$1 disabled. Enable it with --enable-$1])
+      ],
+      [AC_ARG_ENABLE([$1],[$2],[$3])]
     )
 ])
 

From remi at remlab.net  Thu Dec  1 17:58:59 2011
From: remi at remlab.net (=?iso-8859-1?q?R=E9mi?= Denis-Courmont)
Date: Thu, 1 Dec 2011 18:58:59 +0200
Subject: [vlc-devel] [PATCH 3/3] configure: Show enabled plugins.
In-Reply-To: <1322757798-19213-4-git-send-email-gf@unixsol.org>
References: <1322757798-19213-1-git-send-email-gf@unixsol.org>
	<1322757798-19213-4-git-send-email-gf@unixsol.org>
Message-ID: <201112011858.59768.remi@remlab.net>

This was removed because it did not work.

People actually got confused into thinking that this list was authoritative, 
though it was not. Please don't do this.

Le jeudi 1 d?cembre 2011 18:43:18 Georgi Chorbadzhiyski, vous avez ?crit :
> ---
>  configure.ac |    3 ++-
>  1 files changed, 2 insertions(+), 1 deletions(-)
> 
> diff --git a/configure.ac b/configure.ac
> index a44f447..5ea8e5e 100644
> --- a/configure.ac
> +++ b/configure.ac
> @@ -4297,7 +4297,8 @@ libvlc configuration
>  version               : ${VERSION}
>  system                : ${SYS}
>  architecture          : ${ARCH}
> -optimizations         : ${enable_optimizations}"
> +optimizations         : ${enable_optimizations}
> +plugins               : ${PLUGINS}"
>  if test "${enable_vlc}" != "no"; then
>  echo "vlc aliases           :${ALIASES}"
>  else


-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From gf at unixsol.org  Thu Dec  1 18:00:15 2011
From: gf at unixsol.org (Georgi Chorbadzhiyski)
Date: Thu, 01 Dec 2011 19:00:15 +0200
Subject: [vlc-devel] [PATCH 0/3]i configure: Add --disable-all.
In-Reply-To: <201112011857.52214.remi@remlab.net>
References: <1322757798-19213-1-git-send-email-gf@unixsol.org>
	<201112011857.52214.remi@remlab.net>
Message-ID: <4ED7B29F.5030306@unixsol.org>

Around 12/01/2011 06:57 PM, R?mi Denis-Courmont scribbled:
> Le jeudi 1 d?cembre 2011 18:43:15 Georgi Chorbadzhiyski, vous avez ?crit :
>> The following patch series add --disable-all switch to configure and
>> show enabled plugins when configure is over.
>>
>> This allows me to compile minimal vlc with only:
>> ./configure \
>>   --disable-all \
>>   --enable-vlc \
>>   --enable-a52 \
>>   --enable-x264 \
>>   --enable-avcodec \
>>   --enable-avformat \
>>   --enable-swscale \
>>   --enable-sout \
>>   --enable-dvbpsi
>>
>> There are some plugins in configure that are always enabled. For example
>> ./configure --disable-all results in the following plugins:
>>
>>  access_shm, dynamicoverlay and fb
>>
>> I'll fix these (add configure options for them) if the current aproach is
>> acceptable.
> 
> First, most builders are complaining that they are far too many options. There 
> are about a hundred always or automatically enabled plugins. There is no way 
> we are adding options for them.
> 
> Second, there is little to no point in adding options for plugins that don't 
> add dependencies. If you don't want them, don't package them in your binary 
> build rules.
> 
> Third, I have been trying to move as much logic away from configure into the 
> Makefile as possible to reduce recompilation churn. I am totally against going 
> backward to configure.
> 
> Forth, it's a feature, not a bug that you need to explicitly disable some 
> important stuff. I don't see the point except to help people shoot themselves 
> in the feet.

Ok, ignore my comment about access_shm, dynamicoverlay and fb I don't care about
them a lot. What about --disable-all?

-- 
Georgi Chorbadzhiyski
http://georgi.unixsol.org/


From gf at unixsol.org  Thu Dec  1 18:01:19 2011
From: gf at unixsol.org (Georgi Chorbadzhiyski)
Date: Thu, 01 Dec 2011 19:01:19 +0200
Subject: [vlc-devel] [PATCH 3/3] configure: Show enabled plugins.
In-Reply-To: <201112011858.59768.remi@remlab.net>
References: <1322757798-19213-1-git-send-email-gf@unixsol.org>
	<1322757798-19213-4-git-send-email-gf@unixsol.org>
	<201112011858.59768.remi@remlab.net>
Message-ID: <4ED7B2DF.1070005@unixsol.org>

Around 12/01/2011 06:58 PM, R?mi Denis-Courmont scribbled:
> This was removed because it did not work.
> 
> People actually got confused into thinking that this list was authoritative, 
> though it was not. Please don't do this.

Affirmative. It doesn't look correct anyway. It was useful as debug info.

-- 
Georgi Chorbadzhiyski
http://georgi.unixsol.org/


From remi at remlab.net  Thu Dec  1 18:22:44 2011
From: remi at remlab.net (=?iso-8859-1?q?R=E9mi?= Denis-Courmont)
Date: Thu, 1 Dec 2011 19:22:44 +0200
Subject: [vlc-devel] [PATCH 0/3]i configure: Add --disable-all.
In-Reply-To: <4ED7B29F.5030306@unixsol.org>
References: <1322757798-19213-1-git-send-email-gf@unixsol.org>
	<201112011857.52214.remi@remlab.net> <4ED7B29F.5030306@unixsol.org>
Message-ID: <201112011922.44475.remi@remlab.net>

Le jeudi 1 d?cembre 2011 19:00:15 Georgi Chorbadzhiyski, vous avez ?crit :
> Ok, ignore my comment about access_shm, dynamicoverlay and fb I don't care
> about them a lot. What about --disable-all?

I don't understand what problem you're trying to solve with it, so I can't 
say. Neverthless I think it fundamentally cannot work. For instance, you 
cannot disable both shared and static simultaneously.

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From remi at remlab.net  Thu Dec  1 18:26:32 2011
From: remi at remlab.net (=?iso-8859-15?q?R=E9mi?= Denis-Courmont)
Date: Thu, 1 Dec 2011 19:26:32 +0200
Subject: [vlc-devel] [PATCH 6/9] Add missing getaddrinfo(),
	freeaddrinfo(), getnameinfo() and gai_strerror() for OS/2
In-Reply-To: <4ED77CD2.5090707@chollian.net>
References: <1322287529-7242-1-git-send-email-komh@chollian.net>
	<201111291655.40201.remi@remlab.net>
	<4ED77CD2.5090707@chollian.net>
Message-ID: <201112011926.33363.remi@remlab.net>

Le jeudi 1 d?cembre 2011 15:10:42 KO Myung-Hun, vous avez ?crit :
> R?mi Denis-Courmont wrote:
> > I'm neither allowed nor willing to merge GPL stuff into LibVLC as of now.
> 
> How about this ?

It looks OK but it does not apply on my tree.

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From funman at videolan.org  Thu Dec  1 18:56:28 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Thu, 1 Dec 2011 12:56:28 -0500
Subject: [vlc-devel] [PATCH] ActiveX,
 npapi win32: fixed mouse event handling in "stop" mode.
In-Reply-To: <CADOtZwgjEfeU0eoOVCp+N9WN3K6AEbZWJ=jYxyb-hUyqFGebSQ@mail.gmail.com>
References: <CADOtZwgjEfeU0eoOVCp+N9WN3K6AEbZWJ=jYxyb-hUyqFGebSQ@mail.gmail.com>
Message-ID: <20111201125628.283b102f@tok>

Le Thu, 1 Dec 2011 19:39:29 +0700,
Sergey Radionov <rsatom at gmail.com> a ?crit :

> From ce19fac387960df6389f9e16f49341fdaa0a9c67 Mon Sep 17 00:00:00 2001
> From: Sergey Radionov <RSATom at gmail.com>
> Date: Thu, 1 Dec 2011 19:31:55 +0700
> Subject: [PATCH] ActiveX, npapi win32: fixed mouse event handling in "stop"
>  mode.
> 
> ---
>  common/win32_fullscreen.cpp |    6 +++++-
>  1 files changed, 5 insertions(+), 1 deletions(-)
> 
> diff --git a/common/win32_fullscreen.cpp b/common/win32_fullscreen.cpp
> index 5967a1d..690d79c 100644
> --- a/common/win32_fullscreen.cpp
> +++ b/common/win32_fullscreen.cpp
> @@ -49,7 +49,7 @@ void VLCHolderWnd::RegisterWndClassName(HINSTANCE hInstance)
>  
>      if( ! GetClassInfo(_hinstance, getClassName(), &wClass) )
>      {
> -        wClass.style          = 0;//CS_NOCLOSE|CS_DBLCLKS;
> +        wClass.style          = CS_DBLCLKS;

It looks like we used flags CS_NOCLOSE|CS_DBLCKS and someone removed
them. Was that removal to fix something?

>          wClass.lpfnWndProc    = VLCHolderClassWndProc;
>          wClass.cbClsExtra     = 0;
>          wClass.cbWndExtra     = 0;
> @@ -130,6 +130,10 @@ LRESULT CALLBACK VLCHolderWnd::VLCHolderClassWndProc(HWND hWnd, UINT uMsg, WPARA
>              }
>              break;
>          }
> +        case WM_MOUSEMOVE:
> +        case WM_LBUTTONDBLCLK:
> +            h_data->_WindowsManager->OnMouseEvent(uMsg);
> +            break;
>          case WM_MOUSE_EVENT_NOTIFY:{
>              h_data->_WindowsManager->OnMouseEvent(wParam);
>              return WM_MOUSE_EVENT_NOTIFY_SUCCESS;



-- 
Rafa?l Carr?


From nkoriyama at gmail.com  Thu Dec  1 19:02:30 2011
From: nkoriyama at gmail.com (Naohiro KORIYAMA)
Date: Fri, 2 Dec 2011 03:02:30 +0900
Subject: [vlc-devel] UPnP module on Windows has been broken
Message-ID: <CAOa_K+u7x9axyUDSNB8dEidv6KvNpedw9yvnUmBTPDzAv8FGXw@mail.gmail.com>

UPnP module on Windows build has been broken from
vlc-1.2.0-git-20111104-0003-win32.

I built some version around this commit and found out.
5ceabdd8d3fa1c4641bb6d35a0f40d4cf4e5b650
Rewrite inet_pton() and inet_ntop() into compat
This issue happened after this commit.

On Mac build, tihs issue does not happen.
-- 
KORIYAMA, Naohiro
nkoriyama at gmail.com


From funman at videolan.org  Thu Dec  1 19:13:10 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Thu, 1 Dec 2011 13:13:10 -0500
Subject: [vlc-devel] [PATCH] contrib: --disable-all doesn't build
 anything
In-Reply-To: <1322679566-19328-1-git-send-email-funman@videolan.org>
References: <1322679566-19328-1-git-send-email-funman@videolan.org>
Message-ID: <20111201131310.60a5e77f@tok>

Le Wed, 30 Nov 2011 13:59:26 -0500,
Rafa?l Carr? <funman at videolan.org> a ?crit :

> ---
>  contrib/src/main.mak |    3 +++
>  1 files changed, 3 insertions(+), 0 deletions(-)
> 
> diff --git a/contrib/src/main.mak b/contrib/src/main.mak
> index eef0ac2..0ffacde 100644
> --- a/contrib/src/main.mak
> +++ b/contrib/src/main.mak
> @@ -264,6 +264,9 @@ CMAKE = cmake . -DCMAKE_TOOLCHAIN_FILE=$(abspath toolchain.cmake) \
>  PKGS_FOUND :=
>  include $(SRC)/*/rules.mak
>  
> +ifeq ($(PKGS_DISABLE), all)
> +PKGS :=
> +endif
>  #
>  # Targets
>  #

courmisch?

the goal is to use --disable-all --enable-ffmpeg --enable-a52 --enable-vpx (e.g. for embedded)

-- 
Rafa?l Carr?


From gf at unixsol.org  Thu Dec  1 19:31:19 2011
From: gf at unixsol.org (Georgi Chorbadzhiyski)
Date: Thu, 01 Dec 2011 20:31:19 +0200
Subject: [vlc-devel] [PATCH 0/3]i configure: Add --disable-all.
In-Reply-To: <201112011922.44475.remi@remlab.net>
References: <1322757798-19213-1-git-send-email-gf@unixsol.org>
	<201112011857.52214.remi@remlab.net> <4ED7B29F.5030306@unixsol.org>
	<201112011922.44475.remi@remlab.net>
Message-ID: <4ED7C7F7.7090804@unixsol.org>

On 12/1/11 7:22 PM, R?mi Denis-Courmont wrote:
> Le jeudi 1 d?cembre 2011 19:00:15 Georgi Chorbadzhiyski, vous avez ?crit :
>> Ok, ignore my comment about access_shm, dynamicoverlay and fb I don't care
>> about them a lot. What about --disable-all?
>
> I don't understand what problem you're trying to solve with it, so I can't
> say. Neverthless I think it fundamentally cannot work. For instance, you
> cannot disable both shared and static simultaneously.

I'm trying to have what libav have. I disable everything and then
enable only what I need. This option allows me to have white list
of enable parameters and to not care about new additions.

disable-all is only about modules not general options like shared
libs, opitmizaions, etc. I have provided sample use in my first
messaqe of the thread.

Again, I'm not trying to use VLC as desktop video player. I configure
vlc to transcode/encode streams that it reads from network or sdi/v4l.
Thats why I don't care about dbus, qt, xcb, lua, etc. :) I could use
avconv directly but libav's ts muxer is worse that vlc's and libav do not
have input module for decklink cards.

-- 
Georgi Chorbadzhiyski
http://georgi.unixsol.org/


From ptrtap at yahoo.com  Thu Dec  1 19:35:15 2011
From: ptrtap at yahoo.com (Peter Tap)
Date: Thu, 1 Dec 2011 10:35:15 -0800 (PST)
Subject: [vlc-devel] Production build settings for VLC
Message-ID: <1322764515.69318.YahooMailNeo@web125403.mail.ne1.yahoo.com>

Folks,

When I build vlc locally and run it, I see that mp4 vidoes have jitters in them. However, when I run vlc from the distribution image, there are no jitters. I am thinking I probably need additional options when I run configure script. Can someone please help me with these options? Perhaps these options are also used in nightly non-debug build of vlc.

Thank you in advance for your help.

Regards,
Peter
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111201/06c549c5/attachment.html>

From remi at remlab.net  Thu Dec  1 19:49:31 2011
From: remi at remlab.net (=?iso-8859-1?q?R=E9mi?= Denis-Courmont)
Date: Thu, 1 Dec 2011 20:49:31 +0200
Subject: [vlc-devel] [PATCH 1/3] m4: Add VLC_ARG_ENABLE function.
In-Reply-To: <1322757798-19213-2-git-send-email-gf@unixsol.org>
References: <1322757798-19213-1-git-send-email-gf@unixsol.org>
	<1322757798-19213-2-git-send-email-gf@unixsol.org>
Message-ID: <201112012049.32171.remi@remlab.net>

Le jeudi 1 d?cembre 2011 18:43:16 Georgi Chorbadzhiyski, vous avez ?crit :
> It does the same as AC_ARG_ENABLE but checks for if disable_all
> variable is set.
> ---
> 
> Somebody with better m4 knowledge should check this function. It
> probably could be written better.
> 
>  m4/with_pkg.m4 |   13 +++++++++++++
>  1 files changed, 13 insertions(+), 0 deletions(-)
> 
> diff --git a/m4/with_pkg.m4 b/m4/with_pkg.m4
> index 39439b7..340c695 100644
> --- a/m4/with_pkg.m4
> +++ b/m4/with_pkg.m4
> @@ -31,6 +31,19 @@ dnl Prepare a --with-variable-prefix triggered check for
> module, dnl disable by default.
>  dnl
> 
> +dnl VLC_ARG_ENABLE have same parameters as AC_ARG_ENABLE but checks for
> disable_all +AC_DEFUN([VLC_ARG_ENABLE],
> +[
> +    var=enable_$(echo $1 | tr .+- ___)
> +    eval xvar=\$${var}

This probably belongs in the default case of AC_ARG_ENABLE.

> +    AS_IF(
> +      [test x"${disable_all}" = "xyes" -a x"${xvar}" != "xyes" ],

The autoconf convention is that anything but no means yes. This is a 
consequence of how --enable-foo=BAR works.

> +      [eval ${var}=no
> +       AC_MSG_NOTICE([$1 disabled. Enable it with --enable-$1])],
> +      [AC_ARG_ENABLE([$1], [$2], [$3])]
> +    )
> +])
> +
>  AC_DEFUN([PKG_WITH_MODULES],
>  [
>  AC_REQUIRE([PKG_PROG_PKG_CONFIG])


-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From remi at remlab.net  Thu Dec  1 19:49:40 2011
From: remi at remlab.net (=?iso-8859-1?q?R=E9mi?= Denis-Courmont)
Date: Thu, 1 Dec 2011 20:49:40 +0200
Subject: [vlc-devel] [PATCH 0/3]i configure: Add --disable-all.
In-Reply-To: <4ED7C7F7.7090804@unixsol.org>
References: <1322757798-19213-1-git-send-email-gf@unixsol.org>
	<201112011922.44475.remi@remlab.net> <4ED7C7F7.7090804@unixsol.org>
Message-ID: <201112012049.40718.remi@remlab.net>

Le jeudi 1 d?cembre 2011 20:31:19 Georgi Chorbadzhiyski, vous avez ?crit :
> I'm trying to have what libav have. I disable everything and then
> enable only what I need. This option allows me to have white list
> of enable parameters and to not care about new additions.
> 
> disable-all is only about modules not general options like shared
> libs, opitmizaions, etc. I have provided sample use in my first
> messaqe of the thread.

Unfortunately, there is not always a clear cut. "vlc" is not a module.
sout, http and vlm are also not really modules. And then v4l2 is a module, but 
the equally named configure option does not turn it on/off.

And of course, many options enable more than one module at a time.

> Again, I'm not trying to use VLC as desktop video player. I configure
> vlc to transcode/encode streams that it reads from network or sdi/v4l.
> Thats why I don't care about dbus, qt, xcb, lua, etc. :)

While it could be nice to control the plugin set, I could not find a 
reasonable solution within autotools. Unfortunately, this means building from 
a clean chroot environment is currently the only way to tightly control the 
dependencies. I am afraid disable-all does not solve this.

> I could use
> avconv directly but libav's ts muxer is worse that vlc's and libav do not
> have input module for decklink cards.

This should be a libv4l2 plugin anyway. But that's another rant.

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From remi at remlab.net  Thu Dec  1 19:50:54 2011
From: remi at remlab.net (=?utf-8?q?R=C3=A9mi?= Denis-Courmont)
Date: Thu, 1 Dec 2011 20:50:54 +0200
Subject: [vlc-devel] [PATCH] contrib: --disable-all doesn't build
	anything
In-Reply-To: <20111201131310.60a5e77f@tok>
References: <1322679566-19328-1-git-send-email-funman@videolan.org>
	<20111201131310.60a5e77f@tok>
Message-ID: <201112012050.54682.remi@remlab.net>

Le jeudi 1 d?cembre 2011 20:13:10 Rafa?l Carr?, vous avez ?crit :
> the goal is to use --disable-all --enable-ffmpeg --enable-a52 --enable-vpx
> (e.g. for embedded)

We already have disable-disc and disable-encoder or whatever it was.

What's the problem?

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From remi at remlab.net  Thu Dec  1 20:15:35 2011
From: remi at remlab.net (=?iso-8859-1?q?R=E9mi?= Denis-Courmont)
Date: Thu, 1 Dec 2011 21:15:35 +0200
Subject: [vlc-devel] UPnP module on Windows has been broken
In-Reply-To: <CAOa_K+u7x9axyUDSNB8dEidv6KvNpedw9yvnUmBTPDzAv8FGXw@mail.gmail.com>
References: <CAOa_K+u7x9axyUDSNB8dEidv6KvNpedw9yvnUmBTPDzAv8FGXw@mail.gmail.com>
Message-ID: <201112012115.35895.remi@remlab.net>

Le jeudi 1 d?cembre 2011 20:02:30 Naohiro KORIYAMA, vous avez ?crit :
> UPnP module on Windows build has been broken from
> vlc-1.2.0-git-20111104-0003-win32.
> 
> I built some version around this commit and found out.
> 5ceabdd8d3fa1c4641bb6d35a0f40d4cf4e5b650
> Rewrite inet_pton() and inet_ntop() into compat
> This issue happened after this commit.

I guess libupnp.a is not linked properly in VLC contribs. It probably ends up 
linking against VLC libcompat instead of its own inet_pton/inet_ntop 
replacements. That is potentially a problem with any static library that 
provides functions with the exact same name as VLC does.

That being said, unless the VLC replacement functions are buggy, use a 
different calling convention or a different prototype, it should still work by 
pure luck.

> On Mac build, tihs issue does not happen.

I would expect MacOS to provide those functions from libc.

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From remi at remlab.net  Thu Dec  1 20:18:18 2011
From: remi at remlab.net (=?iso-8859-1?q?R=E9mi?= Denis-Courmont)
Date: Thu, 1 Dec 2011 21:18:18 +0200
Subject: [vlc-devel] UPnP module on Windows has been broken
In-Reply-To: <201112012115.35895.remi@remlab.net>
References: <CAOa_K+u7x9axyUDSNB8dEidv6KvNpedw9yvnUmBTPDzAv8FGXw@mail.gmail.com>
	<201112012115.35895.remi@remlab.net>
Message-ID: <201112012118.18678.remi@remlab.net>

Le jeudi 1 d?cembre 2011 21:15:35 R?mi Denis-Courmont, vous avez ?crit :
> That being said, unless the VLC replacement functions are buggy, use a
> different calling convention or a different prototype, it should still work
> by pure luck.

In this particular case, I suspect the bug lies not in VLC 
inet_pton()/inet_ntop() but in Windows sscanf()/printf().

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From funman at videolan.org  Thu Dec  1 20:27:58 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Thu, 1 Dec 2011 14:27:58 -0500
Subject: [vlc-devel] Production build settings for VLC
In-Reply-To: <1322764515.69318.YahooMailNeo@web125403.mail.ne1.yahoo.com>
References: <1322764515.69318.YahooMailNeo@web125403.mail.ne1.yahoo.com>
Message-ID: <20111201142758.508e41c7@tok>

Le Thu, 1 Dec 2011 10:35:15 -0800 (PST),
Peter Tap <ptrtap at yahoo.com> a ?crit :

> Folks,
> 
> When I build vlc locally and run it, I see that mp4 vidoes have jitters in them. However, when I run vlc from the distribution image, there are no jitters. I am thinking I probably need additional options when I run configure script. Can someone please help me with these options? Perhaps these options are also used in nightly non-debug build of vlc.
> 
> Thank you in advance for your help.

vlc win32 build is made with extras/package/win32/configure.sh
vlc macos build is made with extras/package/macosx/configure.sh

You can check the nightly builds option by running them with -vv (in console) or directly on buildbot.videolan.org (Configure step of each build)

-- 
Rafa?l Carr?


From funman at videolan.org  Thu Dec  1 20:31:02 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Thu, 1 Dec 2011 14:31:02 -0500
Subject: [vlc-devel] [PATCH] contrib: --disable-all doesn't build
 anything
In-Reply-To: <201112012050.54682.remi@remlab.net>
References: <1322679566-19328-1-git-send-email-funman@videolan.org>
	<20111201131310.60a5e77f@tok> <201112012050.54682.remi@remlab.net>
Message-ID: <20111201143102.1b0cfa15@tok>

Le Thu, 1 Dec 2011 20:50:54 +0200,
"R?mi Denis-Courmont" <remi at remlab.net> a ?crit :

> Le jeudi 1 d?cembre 2011 20:13:10 Rafa?l Carr?, vous avez ?crit :
> > the goal is to use --disable-all --enable-ffmpeg --enable-a52 --enable-vpx
> > (e.g. for embedded)
> 
> We already have disable-disc and disable-encoder or whatever it was.
> 
> What's the problem?

Everything that is not a disc or encoder related library.

Say, jpeg, subtitles, weird codecs ..

Here are the option i use for android:

    --disable-sout
    --disable-disc
    --disable-caca
    --disable-gecko
    --disable-goom
    --disable-SDL_image
    --disable-gnutls
    --disable-lua
    --disable-libxml2
    --disable-gcrypt
    --disable-gme
    --disable-mpcdec
    --disable-ass
    --disable-freetype2
    --disable-zvbi
    --disable-fontconfig
    --disable-fribidi
    --disable-kate
    --disable-fluid
    --disable-sidplay2
    --disable-gettext
    --disable-schroedinger
    --disable-libmpeg2

White listing would be more reliable

-- 
Rafa?l Carr?


From remi at remlab.net  Thu Dec  1 21:00:40 2011
From: remi at remlab.net (=?utf-8?q?R=C3=A9mi?= Denis-Courmont)
Date: Thu, 1 Dec 2011 22:00:40 +0200
Subject: [vlc-devel] [PATCH] contrib: --disable-all doesn't build
	anything
In-Reply-To: <20111201143102.1b0cfa15@tok>
References: <1322679566-19328-1-git-send-email-funman@videolan.org>
	<201112012050.54682.remi@remlab.net> <20111201143102.1b0cfa15@tok>
Message-ID: <201112012200.41417.remi@remlab.net>

Le jeudi 1 d?cembre 2011 21:31:02 Rafa?l Carr?, vous avez ?crit :
> Le Thu, 1 Dec 2011 20:50:54 +0200,
> 
> "R?mi Denis-Courmont" <remi at remlab.net> a ?crit :
> > Le jeudi 1 d?cembre 2011 20:13:10 Rafa?l Carr?, vous avez ?crit :
> > > the goal is to use --disable-all --enable-ffmpeg --enable-a52
> > > --enable-vpx (e.g. for embedded)
> > 
> > We already have disable-disc and disable-encoder or whatever it was.
> > 
> > What's the problem?
> 
> Everything that is not a disc or encoder related library.

Well you can add generic options for images, synthesis and text if you want to 
remove special codecs categories, and visualizations.

> Say, jpeg, subtitles, weird codecs ..
> 
> Here are the option i use for android:
> 
>     --disable-sout
>     --disable-disc
>     --disable-caca

I really don't understand why we need this in contrib.

>     --disable-gecko

VLC does not use this.

>     --disable-goom
>     --disable-SDL_image
>     --disable-gnutls
>     --disable-lua
>     --disable-libxml2
>     --disable-gcrypt
>     --disable-gme
>     --disable-mpcdec
>     --disable-ass
>     --disable-freetype2
>     --disable-zvbi
>     --disable-fontconfig
>     --disable-fribidi
>     --disable-kate
>     --disable-fluid
>     --disable-sidplay2
>     --disable-gettext
>     --disable-schroedinger
>     --disable-libmpeg2

But if you removed all of that, it's not really VLC anymore. I don't think we 
want to promote that sort of sub-standard builds.

And removing LUA and XML is not serious.

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From funman at videolan.org  Thu Dec  1 21:30:53 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Thu, 1 Dec 2011 15:30:53 -0500
Subject: [vlc-devel] [PATCH] contrib: --disable-all doesn't build
 anything
In-Reply-To: <201112012200.41417.remi@remlab.net>
References: <1322679566-19328-1-git-send-email-funman@videolan.org>
	<201112012050.54682.remi@remlab.net> <20111201143102.1b0cfa15@tok>
	<201112012200.41417.remi@remlab.net>
Message-ID: <20111201153053.08e8f0df@tok>

Le Thu, 1 Dec 2011 22:00:40 +0200,
"R?mi Denis-Courmont" <remi at remlab.net> a ?crit :

> Le jeudi 1 d?cembre 2011 21:31:02 Rafa?l Carr?, vous avez ?crit :
> > Le Thu, 1 Dec 2011 20:50:54 +0200,
> > 
> > "R?mi Denis-Courmont" <remi at remlab.net> a ?crit :
> > > Le jeudi 1 d?cembre 2011 20:13:10 Rafa?l Carr?, vous avez ?crit :
> > > > the goal is to use --disable-all --enable-ffmpeg --enable-a52
> > > > --enable-vpx (e.g. for embedded)
> > > 
> > > We already have disable-disc and disable-encoder or whatever it was.
> > > 
> > > What's the problem?
> > 
> > Everything that is not a disc or encoder related library.
> 
> Well you can add generic options for images, synthesis and text if you want to 
> remove special codecs categories, and visualizations.

possibly we could use more categories

everything is such a category

> > Say, jpeg, subtitles, weird codecs ..
> > 
> > Here are the option i use for android:
> > 
> >     --disable-sout
> >     --disable-disc
> >     --disable-caca
> 
> I really don't understand why we need this in contrib.

well people use it, so we need it

> >     --disable-gecko
> 
> VLC does not use this.

yes it needs to be moved to npapi.am i just didn't do it yet

> >     --disable-goom
> >     --disable-SDL_image
> >     --disable-gnutls
> >     --disable-lua
> >     --disable-libxml2
> >     --disable-gcrypt
> >     --disable-gme
> >     --disable-mpcdec
> >     --disable-ass
> >     --disable-freetype2
> >     --disable-zvbi
> >     --disable-fontconfig
> >     --disable-fribidi
> >     --disable-kate
> >     --disable-fluid
> >     --disable-sidplay2
> >     --disable-gettext
> >     --disable-schroedinger
> >     --disable-libmpeg2
> 
> But if you removed all of that, it's not really VLC anymore. I don't think we 
> want to promote that sort of sub-standard builds.

Well that's my problem. VLC is modular, I take advantage of that, and
still call it VLC.

Is VLC without sout still VLC? Despite the L for LAN ?

Is VLC without disc support still VLC? Despite that libdvdcss was
authored at VideoLAN?

> And removing LUA and XML is not serious.

I have no use for XML or lua and i'm not sure why it's not serious

-- 
Rafa?l Carr?


From remi at remlab.net  Thu Dec  1 21:46:18 2011
From: remi at remlab.net (=?utf-8?q?R=C3=A9mi?= Denis-Courmont)
Date: Thu, 1 Dec 2011 22:46:18 +0200
Subject: [vlc-devel] [PATCH] contrib: --disable-all doesn't build
	anything
In-Reply-To: <20111201153053.08e8f0df@tok>
References: <1322679566-19328-1-git-send-email-funman@videolan.org>
	<201112012200.41417.remi@remlab.net> <20111201153053.08e8f0df@tok>
Message-ID: <201112012246.18912.remi@remlab.net>

Le jeudi 1 d?cembre 2011 22:30:53 Rafa?l Carr?, vous avez ?crit :
> possibly we could use more categories
> 
> everything is such a category

Only if categories are nested.

Bluntly, I think disable-all is useless. If you want 'make all' not to do 
anything, then don't run it! If you don't want automatic detection, then you 
can simply run 'make .foo .bar'.

> > But if you removed all of that, it's not really VLC anymore. I don't
> > think we want to promote that sort of sub-standard builds.
> 
> Well that's my problem. VLC is modular, I take advantage of that, and
> still call it VLC.
> 
> Is VLC without sout still VLC? Despite the L for LAN ?
> 
> Is VLC without disc support still VLC? Despite that libdvdcss was
> authored at VideoLAN?

VLC without discs on a device that does not have a disc player is not 
something that is going to disappoint the user base.

VLC without synthesis or picture viewing will not disappoint many people 
because it is rarely used for this.

> > And removing LUA and XML is not serious.
> 
> I have no use for XML or lua and i'm not sure why it's not serious

They affect playlists and they don't really depend on anything. I don't see 
the point of removing them.

Certainly for deeply embedded applications, you will want to select every 
dependency one by all. But then, you will probably integrate VLC into a wider 
build system, so it is questionable whether contribs even make sense.

Is this really what you're trying to do, or is this just a facade to make it 
easier to make a substandard form of what an full Android port ought to be?

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From funman at videolan.org  Thu Dec  1 21:55:38 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Thu, 1 Dec 2011 15:55:38 -0500
Subject: [vlc-devel] [PATCH] contrib: --disable-all doesn't build
 anything
In-Reply-To: <201112012246.18912.remi@remlab.net>
References: <1322679566-19328-1-git-send-email-funman@videolan.org>
	<201112012200.41417.remi@remlab.net> <20111201153053.08e8f0df@tok>
	<201112012246.18912.remi@remlab.net>
Message-ID: <20111201155538.5755e897@tok>

Le Thu, 1 Dec 2011 22:46:18 +0200,
"R?mi Denis-Courmont" <remi at remlab.net> a ?crit :

> Le jeudi 1 d?cembre 2011 22:30:53 Rafa?l Carr?, vous avez ?crit :
> > possibly we could use more categories
> > 
> > everything is such a category
> 
> Only if categories are nested.
> 
> Bluntly, I think disable-all is useless. If you want 'make all' not to do 
> anything, then don't run it! If you don't want automatic detection, then you 
> can simply run 'make .foo .bar'.

true i could use that, doing the selection at boostrap only requires me to type 'make' though

> > > But if you removed all of that, it's not really VLC anymore. I don't
> > > think we want to promote that sort of sub-standard builds.
> > 
> > Well that's my problem. VLC is modular, I take advantage of that, and
> > still call it VLC.
> > 
> > Is VLC without sout still VLC? Despite the L for LAN ?
> > 
> > Is VLC without disc support still VLC? Despite that libdvdcss was
> > authored at VideoLAN?
> 
> VLC without discs on a device that does not have a disc player is not 
> something that is going to disappoint the user base.
> 
> VLC without synthesis or picture viewing will not disappoint many people 
> because it is rarely used for this.

VLC without contribs whitelisting is disappointing myself

> > > And removing LUA and XML is not serious.
> > 
> > I have no use for XML or lua and i'm not sure why it's not serious
> 
> They affect playlists and they don't really depend on anything. I don't see 
> the point of removing them.

i'm not going to open playlists for now

> Certainly for deeply embedded applications, you will want to select every 
> dependency one by all. But then, you will probably integrate VLC into a wider 
> build system, so it is questionable whether contribs even make sense.

That wider build system could use contrib as-is

> Is this really what you're trying to do, or is this just a facade to make it 
> easier to make a substandard form of what an full Android port ought to be?

it's a nice first step towards a full Android port

-- 
Rafa?l Carr?


From ptrtap at yahoo.com  Thu Dec  1 21:56:10 2011
From: ptrtap at yahoo.com (Peter Tap)
Date: Thu, 1 Dec 2011 12:56:10 -0800 (PST)
Subject: [vlc-devel] 1.2 requires speexdsp. Where is this?
Message-ID: <1322772970.7777.YahooMailNeo@web125403.mail.ne1.yahoo.com>

Folks,

So far I was working with VLC 1.1 stable fork. Today, I built a new machine and downloaded 1.2 fork. When I run configure, I now get the following error:

configure: error: speexdsp not found.

Looks like this is something new as I didn't have this problem before.

I am wondering where can I download this package. Or, is there a way to just disable this?

Thank you in advance for your help.

Regards,
Peter
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111201/29203174/attachment.html>

From ptrtap at yahoo.com  Thu Dec  1 22:00:19 2011
From: ptrtap at yahoo.com (Peter Tap)
Date: Thu, 1 Dec 2011 13:00:19 -0800 (PST)
Subject: [vlc-devel] 1.2 requires speexdsp. Where is this?
In-Reply-To: <1322772970.7777.YahooMailNeo@web125403.mail.ne1.yahoo.com>
References: <1322772970.7777.YahooMailNeo@web125403.mail.ne1.yahoo.com>
Message-ID: <1322773219.19179.YahooMailNeo@web125404.mail.ne1.yahoo.com>

Oh. I forgot to mention. I am building this on Ubuntu 11.10.

Thank you.
Peter



________________________________
 From: Peter Tap <ptrtap at yahoo.com>
To: "vlc-devel at videolan.org" <vlc-devel at videolan.org> 
Sent: Thursday, December 1, 2011 12:56 PM
Subject: [vlc-devel] 1.2 requires speexdsp. Where is this?
 

Folks,

So far I was working with VLC 1.1 stable fork. Today, I built a new machine and downloaded 1.2 fork. When I run configure, I now get the following error:

configure: error: speexdsp not found.

Looks like this is something new as I didn't have this problem before.

I am wondering where can I download this package. Or, is there a way to just disable this?

Thank you in advance for your help.

Regards,
Peter
_______________________________________________
vlc-devel mailing list
To unsubscribe or modify your subscription options:
http://mailman.videolan.org/listinfo/vlc-devel
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111201/a7675d4c/attachment.html>

From jb at videolan.org  Thu Dec  1 22:01:11 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Thu, 1 Dec 2011 22:01:11 +0100
Subject: [vlc-devel] 1.2 requires speexdsp. Where is this?
In-Reply-To: <1322773219.19179.YahooMailNeo@web125404.mail.ne1.yahoo.com>
References: <1322772970.7777.YahooMailNeo@web125403.mail.ne1.yahoo.com>
	<1322773219.19179.YahooMailNeo@web125404.mail.ne1.yahoo.com>
Message-ID: <20111201210111.GB28866@videolan.org>

On Thu, Dec 01, 2011 at 01:00:19PM -0800, Peter Tap wrote :
> Oh. I forgot to mention. I am building this on Ubuntu 11.10.

apt-get install libspeexdsp-dev

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From remi at remlab.net  Thu Dec  1 22:01:48 2011
From: remi at remlab.net (=?utf-8?q?R=C3=A9mi?= Denis-Courmont)
Date: Thu, 1 Dec 2011 23:01:48 +0200
Subject: [vlc-devel] [PATCH] contrib: --disable-all doesn't build
	anything
In-Reply-To: <20111201155538.5755e897@tok>
References: <1322679566-19328-1-git-send-email-funman@videolan.org>
	<201112012246.18912.remi@remlab.net> <20111201155538.5755e897@tok>
Message-ID: <201112012301.48707.remi@remlab.net>

Le jeudi 1 d?cembre 2011 22:55:38 Rafa?l Carr?, vous avez ?crit :
> true i could use that, doing the selection at boostrap only requires me to
> type 'make' though

Eh? My solution is definitely shorter:

# ../bootstrap --disable-all --enable-foo --enable-bar
# make

# ../bootstrap
# make .foo .bar

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From ptrtap at yahoo.com  Thu Dec  1 22:02:29 2011
From: ptrtap at yahoo.com (Peter Tap)
Date: Thu, 1 Dec 2011 13:02:29 -0800 (PST)
Subject: [vlc-devel] Production build settings for VLC
In-Reply-To: <20111201142758.508e41c7@tok>
References: <1322764515.69318.YahooMailNeo@web125403.mail.ne1.yahoo.com>
	<20111201142758.508e41c7@tok>
Message-ID: <1322773349.21700.YahooMailNeo@web125402.mail.ne1.yahoo.com>

Hi Rafa?l,

Thank you for your help.

I am building this on Ubuntu 11.10.

Looking at configure.ac in the root source directory, it appears by default "--enable_optimizations" is enabled. Is there anything else that I could use to optimize performance, especially for x264 files?

Appreciate your help.

Regards,
Peter




________________________________
 From: Rafa?l Carr? <funman at videolan.org>
To: vlc-devel at videolan.org 
Sent: Thursday, December 1, 2011 11:27 AM
Subject: Re: [vlc-devel] Production build settings for VLC
 
Le Thu, 1 Dec 2011 10:35:15 -0800 (PST),
Peter Tap <ptrtap at yahoo.com> a ?crit :

> Folks,
> 
> When I build vlc locally and run it, I see that mp4 vidoes have jitters in them. However, when I run vlc from the distribution image, there are no jitters. I am thinking I probably need additional options when I run configure script. Can someone please help me with these options? Perhaps these options are also used in nightly non-debug build of vlc.
> 
> Thank you in advance for your help.

vlc win32 build is made with extras/package/win32/configure.sh
vlc macos build is made with extras/package/macosx/configure.sh

You can check the nightly builds option by running them with -vv (in console) or directly on buildbot.videolan.org (Configure step of each build)

-- 
Rafa?l Carr?
_______________________________________________
vlc-devel mailing list
To unsubscribe or modify your subscription options:
http://mailman.videolan.org/listinfo/vlc-devel
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111201/e41abd21/attachment.html>

From remi at remlab.net  Thu Dec  1 22:02:49 2011
From: remi at remlab.net (=?iso-8859-15?q?R=E9mi?= Denis-Courmont)
Date: Thu, 1 Dec 2011 23:02:49 +0200
Subject: [vlc-devel] 1.2 requires speexdsp. Where is this?
In-Reply-To: <1322772970.7777.YahooMailNeo@web125403.mail.ne1.yahoo.com>
References: <1322772970.7777.YahooMailNeo@web125403.mail.ne1.yahoo.com>
Message-ID: <201112012302.49449.remi@remlab.net>

Le jeudi 1 d?cembre 2011 22:56:10 Peter Tap, vous avez ?crit :
> I am wondering where can I download this package.

It's part of speex, which VLC already depends on.

> Or, is there a way to just disable this?

Yeah, --disable-speex as before.


-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From funman at videolan.org  Thu Dec  1 22:10:11 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Thu, 1 Dec 2011 16:10:11 -0500
Subject: [vlc-devel] [PATCH] contrib: --disable-all doesn't build
 anything
In-Reply-To: <201112012301.48707.remi@remlab.net>
References: <1322679566-19328-1-git-send-email-funman@videolan.org>
	<201112012246.18912.remi@remlab.net> <20111201155538.5755e897@tok>
	<201112012301.48707.remi@remlab.net>
Message-ID: <20111201161011.6e21c30f@tok>

Le Thu, 1 Dec 2011 23:01:48 +0200,
"R?mi Denis-Courmont" <remi at remlab.net> a ?crit :

> Le jeudi 1 d?cembre 2011 22:55:38 Rafa?l Carr?, vous avez ?crit :
> > true i could use that, doing the selection at boostrap only requires me to
> > type 'make' though
> 
> Eh? My solution is definitely shorter:
> 
> # ../bootstrap --disable-all --enable-foo --enable-bar
> # make
> 
> # ../bootstrap
> # make .foo .bar

Nope, you don't get it:

# ../bootstrap --disable-all --enable-foo --enable-bar
# make
fix shit
# make
fix shit
# make
fix shit
# make
fix shit
# make
fix shit
# make
fix shit
# make
fix shit
# make

happens all the time with win32, osx, android ...

-- 
Rafa?l Carr?


From jb at videolan.org  Thu Dec  1 22:11:53 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Thu, 1 Dec 2011 22:11:53 +0100
Subject: [vlc-devel] [Patches] Dash
In-Reply-To: <CAK-C75QQttr4Mpb=B=YAVKgTpMrOeFpDVU9a4wf_6fRAHxFA9A@mail.gmail.com>
References: <CAK-C75QQttr4Mpb=B=YAVKgTpMrOeFpDVU9a4wf_6fRAHxFA9A@mail.gmail.com>
Message-ID: <20111201211153.GC28866@videolan.org>

On Thu, Dec 01, 2011 at 12:37:19PM +0100, Hugo Beauz?e-Luyssen wrote :
> As the patch has been submitted randomly in two other threads, I'll
> post all the dash patches that I have in local to ease up the
> follow-up.
> The patches that are still being discussed are patches 6 through 8

1, 3 applied.
2 doesn't compile here.
4 is deprecated (applied the next one)
5 doesn't apply
6-8 are still discussed.

Thanks a lot.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From remi at remlab.net  Thu Dec  1 22:13:40 2011
From: remi at remlab.net (=?iso-8859-15?q?R=E9mi?= Denis-Courmont)
Date: Thu, 1 Dec 2011 23:13:40 +0200
Subject: [vlc-devel] Production build settings for VLC
In-Reply-To: <1322773349.21700.YahooMailNeo@web125402.mail.ne1.yahoo.com>
References: <1322764515.69318.YahooMailNeo@web125403.mail.ne1.yahoo.com>
	<20111201142758.508e41c7@tok>
	<1322773349.21700.YahooMailNeo@web125402.mail.ne1.yahoo.com>
Message-ID: <201112012313.40548.remi@remlab.net>

Le jeudi 1 d?cembre 2011 23:02:29 Peter Tap, vous avez ?crit :
> Hi Rafa?l,
> 
> Thank you for your help.
> 
> I am building this on Ubuntu 11.10.

You can find the settings Ubuntu uses from the debian/rules files of the vlc 
source package (apt-get source vlc).

> Looking at configure.ac in the root source directory, it appears by default
> "--enable_optimizations" is enabled. Is there anything else that I could
> use to optimize performance, especially for x264 files?

There is no such a thing as x264 files. If you mean MPEG-4/AVC, then it 
depends on how libavcodec was compiled, not really VLC as such.

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From ptrtap at yahoo.com  Thu Dec  1 22:13:49 2011
From: ptrtap at yahoo.com (Peter Tap)
Date: Thu, 1 Dec 2011 13:13:49 -0800 (PST)
Subject: [vlc-devel] 1.2 requires speexdsp. Where is this?
In-Reply-To: <20111201210111.GB28866@videolan.org>
References: <1322772970.7777.YahooMailNeo@web125403.mail.ne1.yahoo.com>
	<1322773219.19179.YahooMailNeo@web125404.mail.ne1.yahoo.com>
	<20111201210111.GB28866@videolan.org>
Message-ID: <1322774029.31141.YahooMailNeo@web125402.mail.ne1.yahoo.com>

Mucho gracias.

Following your lead, I also had to run

apt-get install libxcb-composite0-dev

Both these packages were not required in 1.1 fork.

In addition, live555 requires some extra patches. But as I don't need live555, I just disabled it during configure.


Thank you once again for your help.

Peter



________________________________
 From: Jean-Baptiste Kempf <jb at videolan.org>
To: Mailing list for VLC media player developers <vlc-devel at videolan.org> 
Sent: Thursday, December 1, 2011 1:01 PM
Subject: Re: [vlc-devel] 1.2 requires speexdsp. Where is this?
 
On Thu, Dec 01, 2011 at 01:00:19PM -0800, Peter Tap wrote :
> Oh. I forgot to mention. I am building this on Ubuntu 11.10.

apt-get install libspeexdsp-dev

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device
_______________________________________________
vlc-devel mailing list
To unsubscribe or modify your subscription options:
http://mailman.videolan.org/listinfo/vlc-devel
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111201/1e070899/attachment.html>

From jb at videolan.org  Thu Dec  1 22:25:33 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Thu, 1 Dec 2011 22:25:33 +0100
Subject: [vlc-devel] [PATCH] npapi: zero unused members of NPClass
In-Reply-To: <CADOtZwhOwBei-Bd_naA0cgjRay4YQR4pRCujKFydocyYoucHtA@mail.gmail.com>
References: <CADOtZwhOwBei-Bd_naA0cgjRay4YQR4pRCujKFydocyYoucHtA@mail.gmail.com>
Message-ID: <20111201212533.GA2894@videolan.org>

On Thu, Dec 01, 2011 at 06:53:28PM +0700, Sergey Radionov wrote :
> 

Thanks a lot, and a lot more.

Did I say thanks? Thanks!

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Thu Dec  1 22:31:08 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Thu, 1 Dec 2011 22:31:08 +0100
Subject: [vlc-devel] [vlc-commits] Contribs: fix fluid compilation on
 Win32
In-Reply-To: <3196d82cf03d38e79df457bd03fc5aea@chewa.net>
References: <CADOtZwiVvvRkqjHwaJ6mY1DOQY=wepA=DXDcwRPnJcgjR8DN5A@mail.gmail.com>
	<20111129235549.05666eb5@tok> <20111130104935.GB19524@videolan.org>
	<3196d82cf03d38e79df457bd03fc5aea@chewa.net>
Message-ID: <20111201213108.GB2894@videolan.org>

On Wed, Nov 30, 2011 at 12:02:23PM +0100, R?mi Denis-Courmont wrote :
> Contrib fixes should be backported to stable.

Ok.

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Thu Dec  1 23:07:12 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Thu, 1 Dec 2011 23:07:12 +0100
Subject: [vlc-devel] [PATCH] ActiveX,
 npapi win32: fixed mouse event handling in "stop" mode.
In-Reply-To: <CADOtZwgjEfeU0eoOVCp+N9WN3K6AEbZWJ=jYxyb-hUyqFGebSQ@mail.gmail.com>
References: <CADOtZwgjEfeU0eoOVCp+N9WN3K6AEbZWJ=jYxyb-hUyqFGebSQ@mail.gmail.com>
Message-ID: <20111201220712.GA11163@videolan.org>

On Thu, Dec 01, 2011 at 07:39:29PM +0700, Sergey Radionov wrote :
> 

Applied.


Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From simon.hailes at screen.subtitling.com  Thu Dec  1 23:35:01 2011
From: simon.hailes at screen.subtitling.com (Simon Hailes)
Date: Thu, 1 Dec 2011 22:35:01 -0000
Subject: [vlc-devel] A small request
Message-ID: <4476B296B92A4741A49B0BD01759070001C5FBE0@sss-uk-ex-02.screen.subtitling.local>

Please can the DVB Subtitle decode be modified to include recognition of
the new 3D types in the PMT - 15 and 25.

Maybe it should actually accept any type, but just with a warning that
it does not know what the type is, rather than saying there are no subs
available?

 

Simon


Simon Hailes | Screen Subtitling Systems Ltd | Chief Technology Officer
Main Line : +44 (0)1473 831700 | Ext : 272 | Fax: +44 (0)1473 830078
simon.hailes at screen.subtitling.com | www.screen.subtitling.com


Visit us at 

BVE, Stand K56, Earls Court, London, 14-16 Feb 2012

Before Printing, think about the environment

This message may contain confidential and/or privileged information. If you are not the intended recipient you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation. Screen Subtitling Systems Ltd. Registered in England No. 2596832. Registered Office: The Old Rectory, Claydon Church Lane, Claydon, Ipswich, Suffolk, IP6 0EQ
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111201/00f761d4/attachment.html>

From jb at videolan.org  Fri Dec  2 00:02:12 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Fri, 2 Dec 2011 00:02:12 +0100
Subject: [vlc-devel] A small request
In-Reply-To: <4476B296B92A4741A49B0BD01759070001C5FBE0@sss-uk-ex-02.screen.subtitling.local>
References: <4476B296B92A4741A49B0BD01759070001C5FBE0@sss-uk-ex-02.screen.subtitling.local>
Message-ID: <20111201230212.GA21235@videolan.org>

Hello Simon,

On Thu, Dec 01, 2011 at 10:35:01PM -0000, Simon Hailes wrote :
> Please can the DVB Subtitle decode be modified to include recognition of
> the new 3D types in the PMT - 15 and 25.

Patches are welcome :D

> Maybe it should actually accept any type, but just with a warning that
> it does not know what the type is, rather than saying there are no subs
> available?

It sure should. Samples?

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From beauze.h at gmail.com  Fri Dec  2 00:31:15 2011
From: beauze.h at gmail.com (=?UTF-8?Q?Hugo_Beauz=C3=A9e=2DLuyssen?=)
Date: Fri, 2 Dec 2011 00:31:15 +0100
Subject: [vlc-devel] [Patches] Dash
In-Reply-To: <20111201211153.GC28866@videolan.org>
References: <CAK-C75QQttr4Mpb=B=YAVKgTpMrOeFpDVU9a4wf_6fRAHxFA9A@mail.gmail.com>
	<20111201211153.GC28866@videolan.org>
Message-ID: <CAK-C75TP_Wwk9A5=OEUCatK3XvkE_yR457Naim3t37+8PEXiPA@mail.gmail.com>

On Thu, Dec 1, 2011 at 10:11 PM, Jean-Baptiste Kempf <jb at videolan.org> wrote:
> On Thu, Dec 01, 2011 at 12:37:19PM +0100, Hugo Beauz?e-Luyssen wrote :
>> As the patch has been submitted randomly in two other threads, I'll
>> post all the dash patches that I have in local to ease up the
>> follow-up.
>> The patches that are still being discussed are patches 6 through 8
>
> 1, 3 applied.
> 2 doesn't compile here.
I have a patch to solve this, but on another workstation. Will send it tomorrow

> 4 is deprecated (applied the next one)
> 5 doesn't apply
Will check tomorrow.

> 6-8 are still discussed.

Still waiting for feedback!!

>
> Thanks a lot.
>
> Best regards,
>
> --
> Jean-Baptiste Kempf
> http://www.jbkempf.com/ - +33 672 704 734
> Sent from my Electronic Device
> _______________________________________________
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
> http://mailman.videolan.org/listinfo/vlc-devel



-- 
Hugo Beauz?e-Luyssen


From funman at videolan.org  Fri Dec  2 01:17:34 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Thu, 1 Dec 2011 19:17:34 -0500
Subject: [vlc-devel] qtsound: added audio capture functionality for MAC
	OS X
In-Reply-To: <20111104011815.B21A1144389@albiero.videolan.org>
References: <20111104011815.B21A1144389@albiero.videolan.org>
Message-ID: <20111201191734.17997cb5@tok>

Le Fri,  4 Nov 2011 02:18:15 +0100 (CET),
git at videolan.org (Michael Feurstein) a ?crit :

> vlc | branch: master | Michael Feurstein <michael.feurstein at gmail.com> | Sun Oct 30 22:27:46 2011 +0100| [3ddb9e51a7d11a69849ff427d38642d4f7d24f5e] | committer: Jean-Baptiste Kempf
> 
> qtsound: added audio capture functionality for MAC OS X
> 
> Tested with Built-in Input, Built-in Microphone and Griffin iMic USB system
> Usage: qtsound://"Built-In Input" or qtsound://"iMic USB audio system"
> 
> Modified-by: Jean-Baptiste Kempf <jb at videolan.org>
> Signed-off-by: Jean-Baptiste Kempf <jb at videolan.org>

breaks PowerPC build

is this the official end of VLC for OSX on PowerPC ?

-- 
Rafa?l Carr?


From jb at videolan.org  Fri Dec  2 01:21:42 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Fri, 2 Dec 2011 01:21:42 +0100
Subject: [vlc-devel] qtsound: added audio capture functionality for MAC
 OS X
In-Reply-To: <20111201191734.17997cb5@tok>
References: <20111104011815.B21A1144389@albiero.videolan.org>
	<20111201191734.17997cb5@tok>
Message-ID: <20111202002142.GA4184@videolan.org>

On Thu, Dec 01, 2011 at 07:17:34PM -0500, Rafa?l Carr? wrote :
> Le Fri,  4 Nov 2011 02:18:15 +0100 (CET),
> git at videolan.org (Michael Feurstein) a ?crit :
> 
> > vlc | branch: master | Michael Feurstein <michael.feurstein at gmail.com> | Sun Oct 30 22:27:46 2011 +0100| [3ddb9e51a7d11a69849ff427d38642d4f7d24f5e] | committer: Jean-Baptiste Kempf
> > 
> > qtsound: added audio capture functionality for MAC OS X
> > 
> > Tested with Built-in Input, Built-in Microphone and Griffin iMic USB system
> > Usage: qtsound://"Built-In Input" or qtsound://"iMic USB audio system"
> > 
> > Modified-by: Jean-Baptiste Kempf <jb at videolan.org>
> > Signed-off-by: Jean-Baptiste Kempf <jb at videolan.org>
> 
> breaks PowerPC build
> 
> is this the official end of VLC for OSX on PowerPC ?

I really doubt it does.

It probably breaks the X.5.sdk, like on jones, because of missing
header.
Taking the Header from the X.6.sdk and adding the import in QTKit.h
should be enough.

I fixed it on jones.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From funman at videolan.org  Fri Dec  2 01:22:58 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Thu, 1 Dec 2011 19:22:58 -0500
Subject: [vlc-devel] qtsound: added audio capture functionality for MAC
 OS X
In-Reply-To: <20111202002142.GA4184@videolan.org>
References: <20111104011815.B21A1144389@albiero.videolan.org>
	<20111201191734.17997cb5@tok> <20111202002142.GA4184@videolan.org>
Message-ID: <20111201192258.4fa63798@tok>

Le Fri, 2 Dec 2011 01:21:42 +0100,
Jean-Baptiste Kempf <jb at videolan.org> a ?crit :

> On Thu, Dec 01, 2011 at 07:17:34PM -0500, Rafa?l Carr? wrote :
> > Le Fri,  4 Nov 2011 02:18:15 +0100 (CET),
> > git at videolan.org (Michael Feurstein) a ?crit :
> > 
> > > vlc | branch: master | Michael Feurstein <michael.feurstein at gmail.com> | Sun Oct 30 22:27:46 2011 +0100| [3ddb9e51a7d11a69849ff427d38642d4f7d24f5e] | committer: Jean-Baptiste Kempf
> > > 
> > > qtsound: added audio capture functionality for MAC OS X
> > > 
> > > Tested with Built-in Input, Built-in Microphone and Griffin iMic USB system
> > > Usage: qtsound://"Built-In Input" or qtsound://"iMic USB audio system"
> > > 
> > > Modified-by: Jean-Baptiste Kempf <jb at videolan.org>
> > > Signed-off-by: Jean-Baptiste Kempf <jb at videolan.org>
> > 
> > breaks PowerPC build
> > 
> > is this the official end of VLC for OSX on PowerPC ?
> 
> I really doubt it does.
> 
> It probably breaks the X.5.sdk, like on jones, because of missing
> header.
> Taking the Header from the X.6.sdk and adding the import in QTKit.h
> should be enough.
> 
> I fixed it on jones.
> 
> Best regards,

Please document what extra steps are needed on a pristine PowerPC
installation, so we can reproduce it on veda for example.

-- 
Rafa?l Carr?


From funman at videolan.org  Fri Dec  2 01:48:26 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Thu, 1 Dec 2011 19:48:26 -0500
Subject: [vlc-devel] qtsound: added audio capture functionality for MAC
 OS X
In-Reply-To: <20111202002142.GA4184@videolan.org>
References: <20111104011815.B21A1144389@albiero.videolan.org>
	<20111201191734.17997cb5@tok> <20111202002142.GA4184@videolan.org>
Message-ID: <20111201194826.24111052@tok>

Le Fri, 2 Dec 2011 01:21:42 +0100,
Jean-Baptiste Kempf <jb at videolan.org> a ?crit :

> On Thu, Dec 01, 2011 at 07:17:34PM -0500, Rafa?l Carr? wrote :
> > Le Fri,  4 Nov 2011 02:18:15 +0100 (CET),
> > git at videolan.org (Michael Feurstein) a ?crit :
> > 
> > > vlc | branch: master | Michael Feurstein <michael.feurstein at gmail.com> | Sun Oct 30 22:27:46 2011 +0100| [3ddb9e51a7d11a69849ff427d38642d4f7d24f5e] | committer: Jean-Baptiste Kempf
> > > 
> > > qtsound: added audio capture functionality for MAC OS X
> > > 
> > > Tested with Built-in Input, Built-in Microphone and Griffin iMic USB system
> > > Usage: qtsound://"Built-In Input" or qtsound://"iMic USB audio system"
> > > 
> > > Modified-by: Jean-Baptiste Kempf <jb at videolan.org>
> > > Signed-off-by: Jean-Baptiste Kempf <jb at videolan.org>
> > 
> > breaks PowerPC build
> > 
> > is this the official end of VLC for OSX on PowerPC ?
> 
> I really doubt it does.
> 
> It probably breaks the X.5.sdk, like on jones, because of missing
> header.
> Taking the Header from the X.6.sdk and adding the import in QTKit.h
> should be enough.
> 
> I fixed it on jones.

Please put that in the wiki, GPL requires all scripts required to compile etcetera

-- 
Rafa?l Carr?


From jb at videolan.org  Fri Dec  2 01:49:41 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Fri, 2 Dec 2011 01:49:41 +0100
Subject: [vlc-devel] qtsound: added audio capture functionality for MAC
 OS X
In-Reply-To: <20111201192258.4fa63798@tok>
References: <20111104011815.B21A1144389@albiero.videolan.org>
	<20111201191734.17997cb5@tok> <20111202002142.GA4184@videolan.org>
	<20111201192258.4fa63798@tok>
Message-ID: <20111202004941.GA5039@videolan.org>

On Thu, Dec 01, 2011 at 07:22:58PM -0500, Rafa?l Carr? wrote :
> Please document what extra steps are needed on a pristine PowerPC
> installation, so we can reproduce it on veda for example.

IRC.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From funman at videolan.org  Fri Dec  2 02:08:54 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Thu, 1 Dec 2011 20:08:54 -0500
Subject: [vlc-devel] qtsound: added audio capture functionality for MAC
 OS X
In-Reply-To: <20111202004941.GA5039@videolan.org>
References: <20111104011815.B21A1144389@albiero.videolan.org>
	<20111201191734.17997cb5@tok> <20111202002142.GA4184@videolan.org>
	<20111201192258.4fa63798@tok> <20111202004941.GA5039@videolan.org>
Message-ID: <20111201200854.150ffbdd@tok>

Le Fri, 2 Dec 2011 01:49:41 +0100,
Jean-Baptiste Kempf <jb at videolan.org> a ?crit :

> On Thu, Dec 01, 2011 at 07:22:58PM -0500, Rafa?l Carr? wrote :
> > Please document what extra steps are needed on a pristine PowerPC
> > installation, so we can reproduce it on veda for example.
> 
> IRC.

The wiki now mentions getting a file (stolen straight from MacOSX 10.6
SDK) from our server.

License of that file doesn't explicitely permit redistribution though:

Copyright:  (c)2008-2010 by Apple Inc., all rights reserved.

Which is neither GPL neither BSD

-- 
Rafa?l Carr?


From jb at videolan.org  Fri Dec  2 02:52:31 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Fri, 2 Dec 2011 02:52:31 +0100
Subject: [vlc-devel] qtsound: added audio capture functionality for MAC
 OS X
In-Reply-To: <20111201200854.150ffbdd@tok>
References: <20111104011815.B21A1144389@albiero.videolan.org>
	<20111201191734.17997cb5@tok> <20111202002142.GA4184@videolan.org>
	<20111201192258.4fa63798@tok> <20111202004941.GA5039@videolan.org>
	<20111201200854.150ffbdd@tok>
Message-ID: <20111202015230.GA21629@videolan.org>

On Thu, Dec 01, 2011 at 08:08:54PM -0500, Rafa?l Carr? wrote :
> Le Fri, 2 Dec 2011 01:49:41 +0100,
> Jean-Baptiste Kempf <jb at videolan.org> a ?crit :
> 
> > On Thu, Dec 01, 2011 at 07:22:58PM -0500, Rafa?l Carr? wrote :
> > > Please document what extra steps are needed on a pristine PowerPC
> > > installation, so we can reproduce it on veda for example.
> > 
> > IRC.
> 
> The wiki now mentions getting a file (stolen straight from MacOSX 10.6
> SDK) from our server.

Get it from the SDK on the machine, as documented on the wiki.

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From funman at videolan.org  Fri Dec  2 03:10:36 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Thu, 1 Dec 2011 21:10:36 -0500
Subject: [vlc-devel] qtsound: added audio capture functionality for MAC
 OS X
In-Reply-To: <20111202015230.GA21629@videolan.org>
References: <20111104011815.B21A1144389@albiero.videolan.org>
	<20111201191734.17997cb5@tok> <20111202002142.GA4184@videolan.org>
	<20111201192258.4fa63798@tok> <20111202004941.GA5039@videolan.org>
	<20111201200854.150ffbdd@tok> <20111202015230.GA21629@videolan.org>
Message-ID: <20111201211036.3d21e25a@tok>

Le Fri, 2 Dec 2011 02:52:31 +0100,
Jean-Baptiste Kempf <jb at videolan.org> a ?crit :

> On Thu, Dec 01, 2011 at 08:08:54PM -0500, Rafa?l Carr? wrote :
> > Le Fri, 2 Dec 2011 01:49:41 +0100,
> > Jean-Baptiste Kempf <jb at videolan.org> a ?crit :
> > 
> > > On Thu, Dec 01, 2011 at 07:22:58PM -0500, Rafa?l Carr? wrote :
> > > > Please document what extra steps are needed on a pristine PowerPC
> > > > installation, so we can reproduce it on veda for example.
> > > 
> > > IRC.
> > 
> > The wiki now mentions getting a file (stolen straight from MacOSX 10.6
> > SDK) from our server.

When I wrote that, the wiki had a download link to the (copyrighted)
header, which pointed to download.videolan.org.

My answer obviously pointed that using copyrighted GPL-incompatible
material to build VLC is not possible.

> Get it from the SDK on the machine, as documented on the wiki.

This however looks fine to me.



I am not accusing JB of theft or whatever, merely pointing a (very
important IMO) irregularity in the development and build process.



Using copyrighted material is of course for VLC development is
of course possible.
We'll lie to nobody when we say the majority of video samples come from
peer to peer distribution.

It's fine for testing and research purposes, not for public consumption.

When you deal with such material and redistribute to other people,
please make that obvious that's it's for testing purposes, and that it
comes from dubious, if not blatently illegal, origins.

-- 
If you want illegal documents or software, ask me, I have plenty of
them.

Rafa?l Carr?


From jb at videolan.org  Fri Dec  2 03:12:42 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Fri, 2 Dec 2011 03:12:42 +0100
Subject: [vlc-devel] qtsound: added audio capture functionality for MAC
 OS X
In-Reply-To: <20111201211036.3d21e25a@tok>
References: <20111104011815.B21A1144389@albiero.videolan.org>
	<20111201191734.17997cb5@tok> <20111202002142.GA4184@videolan.org>
	<20111201192258.4fa63798@tok> <20111202004941.GA5039@videolan.org>
	<20111201200854.150ffbdd@tok> <20111202015230.GA21629@videolan.org>
	<20111201211036.3d21e25a@tok>
Message-ID: <20111202021242.GA25230@videolan.org>

On Thu, Dec 01, 2011 at 09:10:36PM -0500, Rafa?l Carr? wrote :
> We'll lie to nobody when we say the majority of video samples come from
> peer to peer distribution.

You would lie.
Did you even read the User Survey?


-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From funman at videolan.org  Fri Dec  2 03:18:41 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Thu, 1 Dec 2011 21:18:41 -0500
Subject: [vlc-devel] qtsound: added audio capture functionality for MAC
 OS X
In-Reply-To: <20111202021242.GA25230@videolan.org>
References: <20111104011815.B21A1144389@albiero.videolan.org>
	<20111201191734.17997cb5@tok> <20111202002142.GA4184@videolan.org>
	<20111201192258.4fa63798@tok> <20111202004941.GA5039@videolan.org>
	<20111201200854.150ffbdd@tok> <20111202015230.GA21629@videolan.org>
	<20111201211036.3d21e25a@tok> <20111202021242.GA25230@videolan.org>
Message-ID: <20111201211841.50711910@tok>

Le Fri, 2 Dec 2011 03:12:42 +0100,
Jean-Baptiste Kempf <jb at videolan.org> a ?crit :

> On Thu, Dec 01, 2011 at 09:10:36PM -0500, Rafa?l Carr? wrote :
> > We'll lie to nobody when we say the majority of video samples come from
> > peer to peer distribution.
> 
> You would lie.

Ok, I didn't analyze the samples to make a percentage

I have seen more than a few samples coming from copyrighted movies though

> Did you even read the User Survey?

What is that?

-- 
Rafa?l Carr?


From rsatom at gmail.com  Fri Dec  2 03:25:53 2011
From: rsatom at gmail.com (Sergey Radionov)
Date: Fri, 2 Dec 2011 09:25:53 +0700
Subject: [vlc-devel] [PATCH] ActiveX,
 npapi win32: fixed mouse event handling in "stop" mode.
In-Reply-To: <20111201125628.283b102f@tok>
References: <CADOtZwgjEfeU0eoOVCp+N9WN3K6AEbZWJ=jYxyb-hUyqFGebSQ@mail.gmail.com>
	<20111201125628.283b102f@tok>
Message-ID: <CADOtZwjA=EBFHNc1mXcNdKr6qnbtruhxpSro9tZdB-zT3kVwXQ@mail.gmail.com>

2011/12/2 Rafa?l Carr? <funman at videolan.org>

> Le Thu, 1 Dec 2011 19:39:29 +0700,
> Sergey Radionov <rsatom at gmail.com> a ?crit :
> >      if( ! GetClassInfo(_hinstance, getClassName(), &wClass) )
> >      {
> > -        wClass.style          = 0;//CS_NOCLOSE|CS_DBLCLKS;
> > +        wClass.style          = CS_DBLCLKS;
>
> It looks like we used flags CS_NOCLOSE|CS_DBLCKS and someone removed
> them. Was that removal to fix something?
>
no, it was so from initial commit:
http://git.videolan.org/?p=vlc/vlc-1.1.git;a=blob;f=projects/mozilla/vlc_win32_fullscreen.cpp;hb=3a3937f7d87fb3eaae8ea7503da45195be866b46#l39
because in that time VLCHolderWnd was disabled, and not get any user input,
but with
http://git.videolan.org/?p=npapi-vlc.git;a=commit;h=9eb201abd1444d3bb6cc6bd3b24112831b2752c0
all changed...


>
>
> --
> Rafa?l Carr?
> _______________________________________________
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
> http://mailman.videolan.org/listinfo/vlc-devel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111202/0cc67a79/attachment.html>

From funman at videolan.org  Fri Dec  2 04:04:55 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Thu, 1 Dec 2011 22:04:55 -0500
Subject: [vlc-devel] [PATCH] ActiveX,
 npapi win32: fixed mouse event handling in "stop" mode.
In-Reply-To: <CADOtZwjA=EBFHNc1mXcNdKr6qnbtruhxpSro9tZdB-zT3kVwXQ@mail.gmail.com>
References: <CADOtZwgjEfeU0eoOVCp+N9WN3K6AEbZWJ=jYxyb-hUyqFGebSQ@mail.gmail.com>
	<20111201125628.283b102f@tok>
	<CADOtZwjA=EBFHNc1mXcNdKr6qnbtruhxpSro9tZdB-zT3kVwXQ@mail.gmail.com>
Message-ID: <20111201220455.22ca01f7@tok>

Le Fri, 2 Dec 2011 09:25:53 +0700,
Sergey Radionov <rsatom at gmail.com> a ?crit :

> 2011/12/2 Rafa?l Carr? <funman at videolan.org>
> 
> > Le Thu, 1 Dec 2011 19:39:29 +0700,
> > Sergey Radionov <rsatom at gmail.com> a ?crit :
> > >      if( ! GetClassInfo(_hinstance, getClassName(), &wClass) )
> > >      {
> > > -        wClass.style          = 0;//CS_NOCLOSE|CS_DBLCLKS;
> > > +        wClass.style          = CS_DBLCLKS;
> >
> > It looks like we used flags CS_NOCLOSE|CS_DBLCKS and someone removed
> > them. Was that removal to fix something?
> >
> no, it was so from initial commit:
> http://git.videolan.org/?p=vlc/vlc-1.1.git;a=blob;f=projects/mozilla/vlc_win32_fullscreen.cpp;hb=3a3937f7d87fb3eaae8ea7503da45195be866b46#l39
> because in that time VLCHolderWnd was disabled, and not get any user input,
> but with
> http://git.videolan.org/?p=npapi-vlc.git;a=commit;h=9eb201abd1444d3bb6cc6bd3b24112831b2752c0
> all changed...

okay, i see jb already applied your patch

thanks,

-- 
Rafa?l Carr?


From jb at videolan.org  Fri Dec  2 04:14:03 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Fri, 2 Dec 2011 04:14:03 +0100
Subject: [vlc-devel] [PATCH] ActiveX,
 npapi win32: fixed mouse event handling in "stop" mode.
In-Reply-To: <20111201220455.22ca01f7@tok>
References: <CADOtZwgjEfeU0eoOVCp+N9WN3K6AEbZWJ=jYxyb-hUyqFGebSQ@mail.gmail.com>
	<20111201125628.283b102f@tok>
	<CADOtZwjA=EBFHNc1mXcNdKr6qnbtruhxpSro9tZdB-zT3kVwXQ@mail.gmail.com>
	<20111201220455.22ca01f7@tok>
Message-ID: <20111202031403.GA2904@videolan.org>

On Thu, Dec 01, 2011 at 10:04:55PM -0500, Rafa?l Carr? wrote :
> Le Fri, 2 Dec 2011 09:25:53 +0700,
> Sergey Radionov <rsatom at gmail.com> a ?crit :
> 
> > 2011/12/2 Rafa?l Carr? <funman at videolan.org>
> > 
> > > Le Thu, 1 Dec 2011 19:39:29 +0700,
> > > Sergey Radionov <rsatom at gmail.com> a ?crit :
> > > >      if( ! GetClassInfo(_hinstance, getClassName(), &wClass) )
> > > >      {
> > > > -        wClass.style          = 0;//CS_NOCLOSE|CS_DBLCLKS;
> > > > +        wClass.style          = CS_DBLCLKS;
> > >
> > > It looks like we used flags CS_NOCLOSE|CS_DBLCKS and someone removed
> > > them. Was that removal to fix something?
> > >
> > no, it was so from initial commit:
> > http://git.videolan.org/?p=vlc/vlc-1.1.git;a=blob;f=projects/mozilla/vlc_win32_fullscreen.cpp;hb=3a3937f7d87fb3eaae8ea7503da45195be866b46#l39
> > because in that time VLCHolderWnd was disabled, and not get any user input,
> > but with
> > http://git.videolan.org/?p=npapi-vlc.git;a=commit;h=9eb201abd1444d3bb6cc6bd3b24112831b2752c0
> > all changed...
> 
> okay, i see jb already applied your patch

Yep, for the reasons evoqued by Sergey, guessed by git blaming the
file...

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From nkoriyama at gmail.com  Fri Dec  2 06:25:16 2011
From: nkoriyama at gmail.com (Naohiro KORIYAMA)
Date: Fri, 2 Dec 2011 14:25:16 +0900
Subject: [vlc-devel] UPnP module on Windows has been broken
In-Reply-To: <201112012118.18678.remi@remlab.net>
References: <CAOa_K+u7x9axyUDSNB8dEidv6KvNpedw9yvnUmBTPDzAv8FGXw@mail.gmail.com>
	<201112012115.35895.remi@remlab.net>
	<201112012118.18678.remi@remlab.net>
Message-ID: <CAOa_K+u6OuTtt=WxeZ_b_Vc-6jozAxmP=zPj-RAtCpT5auQi1g@mail.gmail.com>

2011/12/2 R?mi Denis-Courmont <remi at remlab.net>:
> Le jeudi 1 d?cembre 2011 21:15:35 R?mi Denis-Courmont, vous avez ?crit :
>> That being said, unless the VLC replacement functions are buggy, use a
>> different calling convention or a different prototype, it should still work
>> by pure luck.
>
> In this particular case, I suspect the bug lies not in VLC
> inet_pton()/inet_ntop() but in Windows sscanf()/printf().

libupnp is built with ipv6 disabled, but in the library use
inet_ntop()/inet_ntop() with AF_INET6.
On initialize libupnp, always error occurs at initialize(UPNP_E_FAILED).
Adding implementation for AF_INET6 in compat/inet_pton.c, everything goes fine.

-- 
KORIYAMA, Naohiro
nkoriyama at gmail.com


From remi at remlab.net  Fri Dec  2 07:27:31 2011
From: remi at remlab.net (=?UTF-8?Q?R=C3=A9mi_Denis-Courmont?=)
Date: Fri, 02 Dec 2011 07:27:31 +0100
Subject: [vlc-devel] UPnP module on Windows has been broken
In-Reply-To: <CAOa_K+u6OuTtt=WxeZ_b_Vc-6jozAxmP=zPj-RAtCpT5auQi1g@mail.gmail.com>
References: <CAOa_K+u7x9axyUDSNB8dEidv6KvNpedw9yvnUmBTPDzAv8FGXw@mail.gmail.com>
	<201112012115.35895.remi@remlab.net>
	<201112012118.18678.remi@remlab.net>
	<CAOa_K+u6OuTtt=WxeZ_b_Vc-6jozAxmP=zPj-RAtCpT5auQi1g@mail.gmail.com>
Message-ID: <a7970263af9f698da8c79dc7f02923c8@chewa.net>

On Fri, 2 Dec 2011 14:25:16 +0900, Naohiro KORIYAMA <nkoriyama at gmail.com>
wrote:
> 2011/12/2 R?mi Denis-Courmont <remi at remlab.net>:
>> Le jeudi 1 d?cembre 2011 21:15:35 R?mi Denis-Courmont, vous avez ?crit
:
>>> That being said, unless the VLC replacement functions are buggy, use a
>>> different calling convention or a different prototype, it should still
>>> work
>>> by pure luck.
>>
>> In this particular case, I suspect the bug lies not in VLC
>> inet_pton()/inet_ntop() but in Windows sscanf()/printf().
> 
> libupnp is built with ipv6 disabled, but in the library use
> inet_ntop()/inet_ntop() with AF_INET6.
> On initialize libupnp, always error occurs at initialize(UPNP_E_FAILED).
> Adding implementation for AF_INET6 in compat/inet_pton.c, everything
goes
> fine.

That's a bug in libupnp then.

-- 
R?mi Denis-Courmont
http://www.remlab.net/


From nkoriyama at gmail.com  Fri Dec  2 07:32:24 2011
From: nkoriyama at gmail.com (Naohiro KORIYAMA)
Date: Fri, 2 Dec 2011 15:32:24 +0900
Subject: [vlc-devel] UPnP module on Windows has been broken
In-Reply-To: <a7970263af9f698da8c79dc7f02923c8@chewa.net>
References: <CAOa_K+u7x9axyUDSNB8dEidv6KvNpedw9yvnUmBTPDzAv8FGXw@mail.gmail.com>
	<201112012115.35895.remi@remlab.net>
	<201112012118.18678.remi@remlab.net>
	<CAOa_K+u6OuTtt=WxeZ_b_Vc-6jozAxmP=zPj-RAtCpT5auQi1g@mail.gmail.com>
	<a7970263af9f698da8c79dc7f02923c8@chewa.net>
Message-ID: <CAOa_K+v4hj-49=jsuROdpyhcF6PNwy3w9LaEOjbZ_VAWDPRy3w@mail.gmail.com>

2011/12/2 R?mi Denis-Courmont <remi at remlab.net>:
> That's a bug in libupnp then.

Absolutely yes!
Only a little inconvenience for me :-)


-- 
KORIYAMA, Naohiro
nkoriyama at gmail.com


From corno.roberto at gmail.com  Fri Dec  2 09:08:55 2011
From: corno.roberto at gmail.com (Roberto Corno)
Date: Fri, 2 Dec 2011 09:08:55 +0100
Subject: [vlc-devel] Fwd: [PATCH] Add 0x40 descriptor support
In-Reply-To: <1321969051-23927-1-git-send-email-corno.roberto@gmail.com>
References: <1321969051-23927-1-git-send-email-corno.roberto@gmail.com>
Message-ID: <CAO6YRm-mngoJWWtpRBxAJN8tiF+4xF5b3==04+LDcBg_FH=5aQ@mail.gmail.com>

ping for review

---------- Forwarded message ----------
From: Roberto Corno <corno.roberto at gmail.com>
Date: 22 November 2011 14:37
Subject: [PATCH] Add 0x40 descriptor support
To: vlc-devel at videolan.org
Cc: Roberto Corno <corno.roberto at gmail.com>


---
 src/descriptors/dr.h    |    1 +
 src/descriptors/dr_40.c |   90
+++++++++++++++++++++++++++++++++++++++++++++++
 src/descriptors/dr_40.h |   86 ++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 177 insertions(+), 0 deletions(-)
 create mode 100644 src/descriptors/dr_40.c
 create mode 100644 src/descriptors/dr_40.h

diff --git a/src/descriptors/dr.h b/src/descriptors/dr.h
index faf318a..93b27ee 100644
--- a/src/descriptors/dr.h
+++ b/src/descriptors/dr.h
@@ -46,6 +46,7 @@
 #include "dr_0d.h"
 #include "dr_0e.h"
 #include "dr_0f.h"
+#include "dr_40.h"
 #include "dr_42.h"
 #include "dr_43.h"
 #include "dr_44.h"
diff --git a/src/descriptors/dr_40.c b/src/descriptors/dr_40.c
new file mode 100644
index 0000000..0ae431e
--- /dev/null
+++ b/src/descriptors/dr_40.c
@@ -0,0 +1,90 @@
+
+
+
+#include "config.h"
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+
+#if defined(HAVE_INTTYPES_H)
+#include <inttypes.h>
+#elif defined(HAVE_STDINT_H)
+#include <stdint.h>
+#endif
+
+#include "../dvbpsi.h"
+#include "../dvbpsi_private.h"
+#include "../descriptor.h"
+
+#include "dr_40.h"
+
+/*****************************************************************************
+ * dvbpsi_DecodeNetworkNameDr
+
*****************************************************************************/
+dvbpsi_network_name_dr_t* dvbpsi_DecodeNetworkNameDr(
+                                        dvbpsi_descriptor_t *
p_descriptor) {
+       dvbpsi_network_name_dr_t * p_decoded;
+
+  /* Check the tag */
+  if(p_descriptor->i_tag != 0x40)
+  {
+       DVBPSI_ERROR_ARG("dr_40 decoder", "bad tag (0x%x)",
p_descriptor->i_tag);
+       return NULL;
+  }
+  /* Don't decode twice */
+  if(p_descriptor->p_decoded)
+       return p_descriptor->p_decoded;
+  /* Allocate memory */
+  p_decoded =
+
 (dvbpsi_network_name_dr_t*)malloc(sizeof(dvbpsi_network_name_dr_t));
+  if(!p_decoded)
+  {
+    DVBPSI_ERROR("dr_40 decoder", "out of memory");
+    return NULL;
+  }
+  /* Decode data */
+  p_decoded->i_name_length = p_descriptor->i_length;
+  if(p_decoded->i_name_length)
+    memcpy(p_decoded->i_name_byte,
+           p_descriptor->p_data,
+           p_decoded->i_name_length);
+
+  p_descriptor->p_decoded = (void*)p_decoded;
+
+  return p_decoded;
+}
+
+
+/*****************************************************************************
+ * dvbpsi_GenNetworkNameDr
+
*****************************************************************************/
+dvbpsi_descriptor_t * dvbpsi_GenNetworkNameDr(
+                                        dvbpsi_network_name_dr_t *
p_decoded,
+                                        int b_duplicate) {
+  /* Create the descriptor */
+  dvbpsi_descriptor_t * p_descriptor =
+               dvbpsi_NewDescriptor(0x40, p_decoded->i_name_length, NULL);
+
+  if(p_descriptor)
+  {
+    /* Encode data */
+    if(p_decoded->i_name_length)
+      memcpy(p_descriptor->p_data,
+             p_decoded->i_name_byte,
+             p_decoded->i_name_length);
+
+    if(b_duplicate)
+    {
+      /* Duplicate decoded data */
+       dvbpsi_network_name_dr_t * p_dup_decoded =
+
 (dvbpsi_network_name_dr_t*)malloc(sizeof(dvbpsi_network_name_dr_t));
+      if(p_dup_decoded)
+        memcpy(p_dup_decoded, p_decoded, sizeof(dvbpsi_network_name_dr_t));
+
+      p_descriptor->p_decoded = (void*)p_dup_decoded;
+    }
+  }
+
+  return p_descriptor;
+}
diff --git a/src/descriptors/dr_40.h b/src/descriptors/dr_40.h
new file mode 100644
index 0000000..4c9c8ee
--- /dev/null
+++ b/src/descriptors/dr_40.h
@@ -0,0 +1,86 @@
+/*
+ * dr_40.h
+ *
+ *  Created on: Nov 22, 2011
+ *      Author: rcorno
+ */
+
+/*!
+ * \file <dr_40.h>
+ * \author Corno Roberto <corno.roberto at gmail.com>
+ * \brief Application interface for the DVB "network name"
+ * descriptor decoder and generator.
+ *
+ * Application interface for the DVB "network name" descriptor
+ * decoder and generator. This descriptor's definition can be found in
+ * ETSI EN 300 468 section 6.2.27.
+ */
+
+#ifndef DR_40_H_
+#define DR_40_H_
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+/*****************************************************************************
+ * dvbpsi_network_name_dr_t
+
*****************************************************************************/
+/*!
+ * \struct dvbpsi_network_name_dr_t
+ * \brief "network name" descriptor structure.
+ *
+ * This structure is used to store a decoded "network name"
+ * descriptor. (ETSI EN 300 468 section 6.2.27).
+ */
+/*!
+ * \typedef struct dvbpsi_network_nameg_dr_s dvbpsi_network_name_dr_t
+ * \brief dvbpsi_network_name_dr_t type definition.
+ */
+typedef struct dvbpsi_network_name_dr_s
+{
+  uint8_t      i_name_length;            /*!< length of the i_name_byte
+                                                  array */
+  uint8_t      i_name_byte[255];         /*!< the name of the delivery
system */
+
+} dvbpsi_network_name_dr_t;
+
+/*****************************************************************************
+ * dvbpsi_DecodeNetworkNameDr
+
*****************************************************************************/
+/*!
+ * \fn dvbpsi_network_name_dr_t * dvbpsi_DecodeNetworkNameDr(
+                                        dvbpsi_descriptor_t * p_descriptor)
+ * \brief "network name" descriptor decoder.
+ * \param p_descriptor pointer to the descriptor structure
+ * \return a pointer to a new "network name" descriptor structure
+ * which contains the decoded data.
+ */
+dvbpsi_network_name_dr_t* dvbpsi_DecodeNetworkNameDr(
+                                        dvbpsi_descriptor_t *
p_descriptor);
+
+
+/*****************************************************************************
+ * dvbpsi_GenNetworkNameDr
+
*****************************************************************************/
+/*!
+ * \fn dvbpsi_descriptor_t * dvbpsi_GenNetworkNameDr(
+                        dvbpsi_network_name_dr_t * p_decoded, int
b_duplicate)
+ * \brief "network name" descriptor generator.
+ * \param p_decoded pointer to a decoded "network name" descriptor
+ * structure
+ * \param b_duplicate if non zero then duplicate the p_decoded structure
into
+ * the descriptor
+ * \return a pointer to a new descriptor structure which contains encoded
data.
+ */
+dvbpsi_descriptor_t * dvbpsi_GenNetworkNameDr(
+                                        dvbpsi_network_name_dr_t *
p_decoded,
+                                        int b_duplicate);
+
+#ifdef __cplusplus
+};
+#endif
+
+#else
+#error "Multiple inclusions of dr_40.h"
+#endif
--
1.7.0.4
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111202/cb61b4e3/attachment.html>

From ptrtap at yahoo.com  Fri Dec  2 09:27:47 2011
From: ptrtap at yahoo.com (Peter Tap)
Date: Fri, 2 Dec 2011 00:27:47 -0800 (PST)
Subject: [vlc-devel] How to get the list of .so files being used by vlc?
Message-ID: <1322814467.14339.YahooMailNeo@web125402.mail.ne1.yahoo.com>

Folks,

I need to create a zip file of my custom vlc build. The idea is to run it on a different test machine. Both the development machine and test machine are running Ubuntu. The test box runs an older version of Ubuntu than the dev box.


I first run the following commands to gather all the VLC files to be zipped.


$ ./configure --prefix /home/peter/MyVLC
$ make install

After running this, directory MyVLC contains most of the files that are needed. However, there are a bunch of files in /usr/lib and /usr/lib/i386-linux-gnu that need to be copied as well. I don't know why vlc won't just copy these files in the install directory as well.

Anyway, to build the list of required files, I run vlc on the test machine with -vv option, look at the output to see which .so file is missing and keep adding the missing library file. This process works but is very cumbersome.

I am wondering if there is a better way to get all the files required to run vlc under one directory hierarchy. Or, if there is an easier way to build the list of needed .so files, I can probably write a script to collect all the files.


Thank you in advance for your help.

Regards,
Peter
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111202/674881ba/attachment.html>

From remi at remlab.net  Fri Dec  2 10:52:24 2011
From: remi at remlab.net (=?UTF-8?Q?R=C3=A9mi_Denis-Courmont?=)
Date: Fri, 02 Dec 2011 10:52:24 +0100
Subject: [vlc-devel] Fwd: [PATCH] Add 0x40 descriptor support
In-Reply-To: <CAO6YRm-mngoJWWtpRBxAJN8tiF+4xF5b3==04+LDcBg_FH=5aQ@mail.gmail.com>
References: <1321969051-23927-1-git-send-email-corno.roberto@gmail.com>
	<CAO6YRm-mngoJWWtpRBxAJN8tiF+4xF5b3==04+LDcBg_FH=5aQ@mail.gmail.com>
Message-ID: <2e58609f6b0a46cf037297ca326afa4f@chewa.net>

On Fri, 2 Dec 2011 09:08:55 +0100, Roberto Corno <corno.roberto at gmail.com>
wrote:
> ping for review

This probably belonds on libdvbpsi-devel.

-- 
R?mi Denis-Courmont
http://www.remlab.net/


From mirsal at mirsal.fr  Fri Dec  2 11:00:44 2011
From: mirsal at mirsal.fr (mirsal)
Date: Fri, 02 Dec 2011 11:00:44 +0100
Subject: [vlc-devel] UPnP module on Windows has been broken
In-Reply-To: <CAOa_K+v4hj-49=jsuROdpyhcF6PNwy3w9LaEOjbZ_VAWDPRy3w@mail.gmail.com>
References: <CAOa_K+u7x9axyUDSNB8dEidv6KvNpedw9yvnUmBTPDzAv8FGXw@mail.gmail.com>
	<201112012115.35895.remi@remlab.net>
	<201112012118.18678.remi@remlab.net>
	<CAOa_K+u6OuTtt=WxeZ_b_Vc-6jozAxmP=zPj-RAtCpT5auQi1g@mail.gmail.com>
	<a7970263af9f698da8c79dc7f02923c8@chewa.net>
	<CAOa_K+v4hj-49=jsuROdpyhcF6PNwy3w9LaEOjbZ_VAWDPRy3w@mail.gmail.com>
Message-ID: <1322820044.2149.22.camel@mirsal-laptop1.mirsal.fr>

On Fri, 2011-12-02 at 15:32 +0900, Naohiro KORIYAMA wrote:
> 2011/12/2 R?mi Denis-Courmont <remi at remlab.net>:
> > That's a bug in libupnp then.
> 
> Absolutely yes!
> Only a little inconvenience for me :-)

I don't have access to any computer running windows so I can't test,
though I've written a patch during VDD which switches to the libupnp4
API (pupnp 1.8) and may fix that problem as it may not be present in
'recent' versions of libupnp.

It would be nice if someone was willing to build pupnp 1.8 for win32 and
test it.

See there:
http://git.videolan.org/?p=vlc/vlc-mirsal.git;a=commit;h=749c6015b733a224af0c98613ef316a0f389be33

Patch also attached for your convenience.

Best,

-- 
mirsal 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-UPnP-services-discovery-Add-support-for-IPv6-and-use.patch
Type: text/x-patch
Size: 7024 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111202/9c5de502/attachment.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 490 bytes
Desc: This is a digitally signed message part
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111202/9c5de502/attachment.sig>

From remi at remlab.net  Fri Dec  2 11:01:31 2011
From: remi at remlab.net (=?UTF-8?Q?R=C3=A9mi_Denis-Courmont?=)
Date: Fri, 02 Dec 2011 11:01:31 +0100
Subject: [vlc-devel]
 =?utf-8?q?How_to_get_the_list_of_=2Eso_files_being_us?=
 =?utf-8?q?ed_by_vlc=3F?=
In-Reply-To: <1322814467.14339.YahooMailNeo@web125402.mail.ne1.yahoo.com>
References: <1322814467.14339.YahooMailNeo@web125402.mail.ne1.yahoo.com>
Message-ID: <0ce4d21293049fead598ff612d55bec3@chewa.net>

On Fri, 2 Dec 2011 00:27:47 -0800 (PST), Peter Tap <ptrtap at yahoo.com>
wrote:
> After running this, directory MyVLC contains most of the files that are
> needed. However, there are a bunch of files in /usr/lib and
> /usr/lib/i386-linux-gnu that need to be copied as well. I don't know why
> vlc won't just copy these files in the install directory as well.

Just like VLC does not touch Windows DLLs, it does not touch Ubuntu shared
objects. These files are not part of VLC simply.

> Anyway, to build the list of required files, I run vlc on the test
machine
> with -vv option, look at the output to see which .so file is missing and
> keep adding the missing library file. This process works but is very
> cumbersome.

ldd gives the list. But overloading system libraries is a recipe for
failure.

> I am wondering if there is a better way to get all the files required to
> run vlc under one directory hierarchy. Or, if there is an easier way to
> build the list of needed .so files, I can probably write a script to
> collect all the files.

dpkg is your friend.

-- 
R?mi Denis-Courmont
http://www.remlab.net/_


From nkoriyama at gmail.com  Fri Dec  2 12:49:01 2011
From: nkoriyama at gmail.com (Naohiro KORIYAMA)
Date: Fri, 2 Dec 2011 20:49:01 +0900
Subject: [vlc-devel] UPnP module on Windows has been broken
In-Reply-To: <1322820044.2149.22.camel@mirsal-laptop1.mirsal.fr>
References: <CAOa_K+u7x9axyUDSNB8dEidv6KvNpedw9yvnUmBTPDzAv8FGXw@mail.gmail.com>
	<201112012115.35895.remi@remlab.net>
	<201112012118.18678.remi@remlab.net>
	<CAOa_K+u6OuTtt=WxeZ_b_Vc-6jozAxmP=zPj-RAtCpT5auQi1g@mail.gmail.com>
	<a7970263af9f698da8c79dc7f02923c8@chewa.net>
	<CAOa_K+v4hj-49=jsuROdpyhcF6PNwy3w9LaEOjbZ_VAWDPRy3w@mail.gmail.com>
	<1322820044.2149.22.camel@mirsal-laptop1.mirsal.fr>
Message-ID: <CAOa_K+u6fYukvQ+xb3EUQGmpCVkUegc9kc_hfb3W45RGqhUNFg@mail.gmail.com>

2011/12/2 mirsal <mirsal at mirsal.fr>:
> On Fri, 2011-12-02 at 15:32 +0900, Naohiro KORIYAMA wrote:
>> 2011/12/2 R?mi Denis-Courmont <remi at remlab.net>:
>> > That's a bug in libupnp then.
>>
>> Absolutely yes!
>> Only a little inconvenience for me :-)
>
> I don't have access to any computer running windows so I can't test,
> though I've written a patch during VDD which switches to the libupnp4
> API (pupnp 1.8) and may fix that problem as it may not be present in
> 'recent' versions of libupnp.
>
> It would be nice if someone was willing to build pupnp 1.8 for win32 and
> test it.

I don't build punp 1.8 yet. I'm using libupnp 1.6.14 but after
applying this patch,  upnp module works on Windows.

> See there:
> http://git.videolan.org/?p=vlc/vlc-mirsal.git;a=commit;h=749c6015b733a224af0c98613ef316a0f389be33
>
> Patch also attached for your convenience.

Thank you very much!

-- 
KORIYAMA, Naohiro
nkoriyama at gmail.com


From jpsaman at videolan.org  Fri Dec  2 13:04:41 2011
From: jpsaman at videolan.org (Jean-Paul Saman)
Date: Fri, 2 Dec 2011 13:04:41 +0100
Subject: [vlc-devel] Fwd: [PATCH] Add 0x40 descriptor support
In-Reply-To: <2e58609f6b0a46cf037297ca326afa4f@chewa.net>
References: <1321969051-23927-1-git-send-email-corno.roberto@gmail.com>
	<CAO6YRm-mngoJWWtpRBxAJN8tiF+4xF5b3==04+LDcBg_FH=5aQ@mail.gmail.com>
	<2e58609f6b0a46cf037297ca326afa4f@chewa.net>
Message-ID: <CAK9US3DBYjQmocM=yDcw0sKGQ1B6tTEB2x0Bnb-38EKwTJ6HGA@mail.gmail.com>

2011/12/2 R?mi Denis-Courmont <remi at remlab.net>:
> On Fri, 2 Dec 2011 09:08:55 +0100, Roberto Corno <corno.roberto at gmail.com>
> wrote:
>> ping for review
>
> This probably belonds on libdvbpsi-devel.

Indeed.

Kind regards,

Jean-Paul Saman

>
> --
> R?mi Denis-Courmont
> http://www.remlab.net/
> _______________________________________________
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
> http://mailman.videolan.org/listinfo/vlc-devel


From rsatom at gmail.com  Fri Dec  2 13:57:35 2011
From: rsatom at gmail.com (Sergey Radionov)
Date: Fri, 2 Dec 2011 19:57:35 +0700
Subject: [vlc-devel] [PATCH] ActiveX,
	npapi win32: setup parent hwnd on every "play".
In-Reply-To: <CADOtZwiKmfQ+KWyWSL5Y17=gbitHM5H0N-3byXtLhzJyrSZ4ZA@mail.gmail.com>
References: <CADOtZwiKmfQ+KWyWSL5Y17=gbitHM5H0N-3byXtLhzJyrSZ4ZA@mail.gmail.com>
Message-ID: <CADOtZwi0KWwseNhRHRXnZwwFJXTUgxz88uht2mhY=8tmaP4Zjg@mail.gmail.com>

Please ignore this patch, it will be included in more complex patch.

2011/12/1 Sergey Radionov <rsatom at gmail.com>

> But I don't shure, may be it is a bug in libvlc, when after "stop" it
> forget about previously seted parent hwnd (via
> libvlc_media_player_set_hwnd)?
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111202/1164b3e9/attachment.html>

From komh78 at gmail.com  Fri Dec  2 14:25:08 2011
From: komh78 at gmail.com (KO Myung-Hun)
Date: Fri, 02 Dec 2011 22:25:08 +0900
Subject: [vlc-devel] [PATCH 6/9] Add missing getaddrinfo(),
 freeaddrinfo(), getnameinfo() and gai_strerror() for OS/2
In-Reply-To: <201112011926.33363.remi@remlab.net>
References: <1322287529-7242-1-git-send-email-komh@chollian.net>	<201111291655.40201.remi@remlab.net>	<4ED77CD2.5090707@chollian.net>
	<201112011926.33363.remi@remlab.net>
Message-ID: <4ED8D1B4.2000507@chollian.net>



R?mi Denis-Courmont wrote:
> Le jeudi 1 d?cembre 2011 15:10:42 KO Myung-Hun, vous avez ?crit :
>> R?mi Denis-Courmont wrote:
>>> I'm neither allowed nor willing to merge GPL stuff into LibVLC as of now.
>>
>> How about this ?
> 
> It looks OK but it does not apply on my tree.
> 

As I said before, this patch depends on the following patch.

    [PATCH 5/9] Define IPv6 stuffs required in order to compile on OS/2.

If you cannot accept that patch, I'll make another version of this patch
not depending on it.

-- 
KO Myung-Hun

Using Mozilla SeaMonkey 2.0.14
Under OS/2 Warp 4 for Korean with FixPak #15
On AMD ThunderBird 1GHz with 512 MB RAM

Korean OS/2 User Community : http://www.ecomstation.co.kr



From nkoriyama at gmail.com  Fri Dec  2 14:25:52 2011
From: nkoriyama at gmail.com (Naohiro KORIYAMA)
Date: Fri, 2 Dec 2011 22:25:52 +0900
Subject: [vlc-devel] UPnP module on Windows has been broken
In-Reply-To: <CAOa_K+u6fYukvQ+xb3EUQGmpCVkUegc9kc_hfb3W45RGqhUNFg@mail.gmail.com>
References: <CAOa_K+u7x9axyUDSNB8dEidv6KvNpedw9yvnUmBTPDzAv8FGXw@mail.gmail.com>
	<201112012115.35895.remi@remlab.net>
	<201112012118.18678.remi@remlab.net>
	<CAOa_K+u6OuTtt=WxeZ_b_Vc-6jozAxmP=zPj-RAtCpT5auQi1g@mail.gmail.com>
	<a7970263af9f698da8c79dc7f02923c8@chewa.net>
	<CAOa_K+v4hj-49=jsuROdpyhcF6PNwy3w9LaEOjbZ_VAWDPRy3w@mail.gmail.com>
	<1322820044.2149.22.camel@mirsal-laptop1.mirsal.fr>
	<CAOa_K+u6fYukvQ+xb3EUQGmpCVkUegc9kc_hfb3W45RGqhUNFg@mail.gmail.com>
Message-ID: <CAOa_K+vCSK_a08w3+PC_099esbh-g3t74_rZadDoG83ew0jwng@mail.gmail.com>

2011/12/2 Naohiro KORIYAMA <nkoriyama at gmail.com>:
>> It would be nice if someone was willing to build pupnp 1.8 for win32 and
>> test it.
>
> I don't build punp 1.8 yet. I'm using libupnp 1.6.14 but after
> applying this patch, ?upnp module works on Windows.
>

Sorry, I was wrong. upnp module was not created, because of version
check (1.6.14).
So, working version module, I created for the test,  was not overritten :-)

I changed the libupnp version is 1.8.0 (from git), compile module is failed.
Because of "UpnpInit2 is not declared." This function maybe not
enabled when disable-ipv6.

Anyway, on Windows, if  inet_ntop()/inet_ntop() is called with
AF_INET6, the function always returns error value.
-- 
KORIYAMA, Naohiro
nkoriyama at gmail.com


From ee05154 at fe.up.pt  Fri Dec  2 16:20:29 2011
From: ee05154 at fe.up.pt (Nuno Mota)
Date: Fri, 02 Dec 2011 15:20:29 +0000
Subject: [vlc-devel] RTSP muxing
Message-ID: <b77e083e648ed932e98a07d22eb9d17b@fe.up.pt>

Hi,

I'm sorry if this seems out of the scope but there are a few questions 
I'm seeking an answer for. Currently I'm working on the VOD server and 
for some reason I'm having difficulties both in keeping the original TS 
pids and keeping the original TS audio and video sequence. I'm using 
ffmpeg for demuxing and for some reason --ts-es-id-pid --sout-es-id-pid 
fails because the sout pids are values from 0 to x, which may suggest 
that demuxing with ffmpeg does not translate the original pids into the 
es ids. This only happens when I switch on the demuxing using 
avformat(ffmpeg).

Other thing is that using the vod server I need to use mux mp2t in 
order to multiplex all the ESs into on RTP stream. The problem is that 
the original order is not kept and I can't seem to find a way to do so. 
Normally VLC seems to pick a random audio first, then a video, and then 
all the remaining audio.

type : vod
enabled : yes
mux : mp2t


PS: How can I use ffmpeg muxing in this situation?

Thanks for the attention,
Best regards,
Nuno




From funman at videolan.org  Fri Dec  2 17:23:49 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Fri, 2 Dec 2011 11:23:49 -0500
Subject: [vlc-devel] [PATCH] picture_BlendSubpicture() : ensure the
 chromas are compatible
In-Reply-To: <1321674623-5718-1-git-send-email-funman@videolan.org>
References: <1321674623-5718-1-git-send-email-funman@videolan.org>
Message-ID: <20111202112349.6f298e65@tok>

Le Fri, 18 Nov 2011 22:50:23 -0500,
Rafa?l Carr? <funman at videolan.org> a ?crit :

> In this testcase the picture being blended is J420 and source is I420
> 
> fourcc.c says J420 is YUV_PLANAR_420, like I420
> libavutil/pixdesc.c uses the same description for J420 & I420
> 552b93b46f79299a4b2d952bfd34e1d5eed7bbe0 says J420 & I420 are 'not exactly equal'
> vlc_fourcc.h says:
>     J420 /* Planar YUV 4:2:0 Y:U:V full scale */
>     I420 /* Planar YUV 4:2:0 Y:U:V 8-bit */
>   -> different pixel size? but video_format_Setup() says both have 12 bits per pixel
> VoutDisplayCreateRender() assumes they are identical if source is J4** and video surface is I4**
> libswscale assumes they are identical in its 'handle_jpeg' function
> ---
>  src/misc/picture.c |    6 +++++-
>  1 files changed, 5 insertions(+), 1 deletions(-)
> 
> diff --git a/src/misc/picture.c b/src/misc/picture.c
> index 87876b0..1904e98 100644
> --- a/src/misc/picture.c
> +++ b/src/misc/picture.c
> @@ -400,7 +400,11 @@ int picture_Export( vlc_object_t *p_obj,
>  void picture_BlendSubpicture(picture_t *dst,
>                               filter_t *blend, subpicture_t *src)
>  {
> -    assert(blend && dst && blend->fmt_out.video.i_chroma == dst->format.i_chroma);
> +    assert(blend && dst &&
> +        !memcmp(vlc_fourcc_GetChromaDescription(blend->fmt_out.video.i_chroma),
> +                vlc_fourcc_GetChromaDescription(dst->format.i_chroma),
> +                sizeof(vlc_chroma_description_t)));
> +
>      assert(src && !src->b_fade && src->b_absolute);
>  
>      for (subpicture_region_t *r = src->p_region; r != NULL; r = r->p_next) {


ping (fixes #5516)

Do we need a i420 <-> j420 filter instead?

-- 
Rafa?l Carr?


From funman at videolan.org  Fri Dec  2 17:30:44 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Fri, 2 Dec 2011 11:30:44 -0500
Subject: [vlc-devel] [PATCH] contrib: --disable-all doesn't build
 anything
In-Reply-To: <20111201161011.6e21c30f@tok>
References: <1322679566-19328-1-git-send-email-funman@videolan.org>
	<201112012246.18912.remi@remlab.net> <20111201155538.5755e897@tok>
	<201112012301.48707.remi@remlab.net> <20111201161011.6e21c30f@tok>
Message-ID: <20111202113044.0facc458@tok>

Le Thu, 1 Dec 2011 16:10:11 -0500,
Rafa?l Carr? <funman at videolan.org> a ?crit :

> Le Thu, 1 Dec 2011 23:01:48 +0200,
> "R?mi Denis-Courmont" <remi at remlab.net> a ?crit :
> 
> > Le jeudi 1 d?cembre 2011 22:55:38 Rafa?l Carr?, vous avez ?crit :
> > > true i could use that, doing the selection at boostrap only requires me to
> > > type 'make' though
> > 
> > Eh? My solution is definitely shorter:
> > 
> > # ../bootstrap --disable-all --enable-foo --enable-bar
> > # make
> > 
> > # ../bootstrap
> > # make .foo .bar
> 
> Nope, you don't get it:
> 
> # ../bootstrap --disable-all --enable-foo --enable-bar
> # make
> fix shit
> # make
> fix shit
> # make
> fix shit
> # make
> fix shit
> # make
> fix shit
> # make
> fix shit
> # make
> fix shit
> # make
> 
> happens all the time with win32, osx, android ...
> 

so can i push this?

-- 
Rafa?l Carr?


From remi at remlab.net  Fri Dec  2 18:36:29 2011
From: remi at remlab.net (=?utf-8?q?R=C3=A9mi?= Denis-Courmont)
Date: Fri, 2 Dec 2011 19:36:29 +0200
Subject: [vlc-devel] [PATCH] picture_BlendSubpicture() : ensure the
	chromas are compatible
In-Reply-To: <20111202112349.6f298e65@tok>
References: <1321674623-5718-1-git-send-email-funman@videolan.org>
	<20111202112349.6f298e65@tok>
Message-ID: <201112021936.30031.remi@remlab.net>

Le vendredi 2 d?cembre 2011 18:23:49 Rafa?l Carr?, vous avez ?crit :
> > -    assert(blend && dst && blend->fmt_out.video.i_chroma ==
> > dst->format.i_chroma); +    assert(blend && dst &&
> > +       
> > !memcmp(vlc_fourcc_GetChromaDescription(blend->fmt_out.video.i_chroma),
> > +                vlc_fourcc_GetChromaDescription(dst->format.i_chroma),
> > +                sizeof(vlc_chroma_description_t)));
> > +
> > 
> >      assert(src && !src->b_fade && src->b_absolute);
> >      
> >      for (subpicture_region_t *r = src->p_region; r != NULL; r =
> >      r->p_next) {
> 
> ping (fixes #5516)

It makes sense but it looks ugly as hell :P

> Do we need a i420 <-> j420 filter instead?

I wonder... Laurent?

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From remi at remlab.net  Fri Dec  2 18:37:26 2011
From: remi at remlab.net (=?utf-8?q?R=C3=A9mi?= Denis-Courmont)
Date: Fri, 2 Dec 2011 19:37:26 +0200
Subject: [vlc-devel] [PATCH] contrib: --disable-all doesn't build
	anything
In-Reply-To: <20111202113044.0facc458@tok>
References: <1322679566-19328-1-git-send-email-funman@videolan.org>
	<20111201161011.6e21c30f@tok> <20111202113044.0facc458@tok>
Message-ID: <201112021937.27007.remi@remlab.net>

Le vendredi 2 d?cembre 2011 18:30:44 Rafa?l Carr?, vous avez ?crit :
> so can i push this?

I don't like it but it's not so bad as to make me feel like revert.

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From fenrir at elivagar.org  Fri Dec  2 20:50:11 2011
From: fenrir at elivagar.org (Laurent Aimar)
Date: Fri, 2 Dec 2011 20:50:11 +0100
Subject: [vlc-devel] [PATCH] picture_BlendSubpicture() : ensure
	the	chromas are compatible
In-Reply-To: <201112021936.30031.remi@remlab.net>
References: <1321674623-5718-1-git-send-email-funman@videolan.org>
	<20111202112349.6f298e65@tok> <201112021936.30031.remi@remlab.net>
Message-ID: <20111202195011.GA28528@elivagar.org>

On Fri, Dec 02, 2011 at 07:36:29PM +0200, R?mi Denis-Courmont wrote:
> Le vendredi 2 d?cembre 2011 18:23:49 Rafa?l Carr?, vous avez ?crit :
> > > -    assert(blend && dst && blend->fmt_out.video.i_chroma ==
> > > dst->format.i_chroma); +    assert(blend && dst &&
> > > +       
> > > !memcmp(vlc_fourcc_GetChromaDescription(blend->fmt_out.video.i_chroma),
> > > +                vlc_fourcc_GetChromaDescription(dst->format.i_chroma),
> > > +                sizeof(vlc_chroma_description_t)));
> > > +
> > > 
> > >      assert(src && !src->b_fade && src->b_absolute);
> > >      
> > >      for (subpicture_region_t *r = src->p_region; r != NULL; r =
> > >      r->p_next) {
> > 
> > ping (fixes #5516)
> 
> It makes sense but it looks ugly as hell :P
 yes :(
> 
> > Do we need a i420 <-> j420 filter instead?
 Not really. We could but it's way better done at the vout level.

Anyway, about the assert, I think we could also simply remove it.

-- 
fenrir


From jb at videolan.org  Fri Dec  2 20:52:58 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Fri, 2 Dec 2011 20:52:58 +0100
Subject: [vlc-devel] [PATCH] picture_BlendSubpicture() : ensure the
 chromas are compatible
In-Reply-To: <20111202195011.GA28528@elivagar.org>
References: <1321674623-5718-1-git-send-email-funman@videolan.org>
	<20111202112349.6f298e65@tok> <201112021936.30031.remi@remlab.net>
	<20111202195011.GA28528@elivagar.org>
Message-ID: <20111202195258.GA32159@videolan.org>

On Fri, Dec 02, 2011 at 08:50:11PM +0100, Laurent Aimar wrote :
> Anyway, about the assert, I think we could also simply remove it.

Agreed. But I think a msg_Dbg might be good.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Fri Dec  2 21:20:44 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Fri,  2 Dec 2011 21:20:44 +0100
Subject: [vlc-devel] [PATCH] MacOS: default to 10.6.sdk
Message-ID: <1322857244-8184-1-git-send-email-jb@videolan.org>

---
 configure.ac |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/configure.ac b/configure.ac
index 09766ac..e6b30ed 100644
--- a/configure.ac
+++ b/configure.ac
@@ -158,7 +158,7 @@ case "${host_os}" in
     AC_ARG_WITH(macosx-sdk,
       [AS_HELP_STRING([--with-macosx-sdk=DIR],
         [compile using the SDK in DIR])])
-    test "${with_macosx_sdk}" = "" && with_macosx_sdk=/Developer/SDKs/MacOSX10.5.sdk
+    test "${with_macosx_sdk}" = "" && with_macosx_sdk=/Developer/SDKs/MacOSX10.6.sdk
     ! test -d "${with_macosx_sdk}" && AC_MSG_ERROR([SDK "${with_macosx_sdk}" not found])
     AC_ARG_WITH(macosx-version-min,
       [AS_HELP_STRING([--with-macosx-version-min=VERSION],
-- 
1.7.8.rc3



From funman at videolan.org  Fri Dec  2 21:52:39 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Fri, 2 Dec 2011 15:52:39 -0500
Subject: [vlc-devel] [PATCH] MacOS: default to 10.6.sdk
In-Reply-To: <1322857244-8184-1-git-send-email-jb@videolan.org>
References: <1322857244-8184-1-git-send-email-jb@videolan.org>
Message-ID: <20111202155239.1eac4067@tok>

Le Fri,  2 Dec 2011 21:20:44 +0100,
Jean-Baptiste Kempf <jb at videolan.org> a ?crit :

> ---
>  configure.ac |    2 +-
>  1 files changed, 1 insertions(+), 1 deletions(-)
> 
> diff --git a/configure.ac b/configure.ac
> index 09766ac..e6b30ed 100644
> --- a/configure.ac
> +++ b/configure.ac
> @@ -158,7 +158,7 @@ case "${host_os}" in
>      AC_ARG_WITH(macosx-sdk,
>        [AS_HELP_STRING([--with-macosx-sdk=DIR],
>          [compile using the SDK in DIR])])
> -    test "${with_macosx_sdk}" = "" && with_macosx_sdk=/Developer/SDKs/MacOSX10.5.sdk
> +    test "${with_macosx_sdk}" = "" && with_macosx_sdk=/Developer/SDKs/MacOSX10.6.sdk
>      ! test -d "${with_macosx_sdk}" && AC_MSG_ERROR([SDK "${with_macosx_sdk}" not found])
>      AC_ARG_WITH(macosx-version-min,
>        [AS_HELP_STRING([--with-macosx-version-min=VERSION],

if binary runs on 10.5 ok

-- 
Rafa?l Carr?


From jb at videolan.org  Fri Dec  2 21:55:09 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Fri, 2 Dec 2011 21:55:09 +0100
Subject: [vlc-devel] [PATCH] MacOS: default to 10.6.sdk
In-Reply-To: <20111202155239.1eac4067@tok>
References: <1322857244-8184-1-git-send-email-jb@videolan.org>
	<20111202155239.1eac4067@tok>
Message-ID: <20111202205509.GA5028@videolan.org>

On Fri, Dec 02, 2011 at 03:52:39PM -0500, Rafa?l Carr? wrote :
> if binary runs on 10.5 ok

It should.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From rogerdpack2 at gmail.com  Sat Dec  3 00:35:24 2011
From: rogerdpack2 at gmail.com (Roger Pack)
Date: Fri, 2 Dec 2011 16:35:24 -0700
Subject: [vlc-devel] error cross compiling
Message-ID: <CAL1QdWeZ3dhBnN+rF=MjjjLeSnWQ5RmNk5=ME2e8F2EAtfXLRQ@mail.gmail.com>

When trying to carefully cross compile vlc-1.1.12

./configure --with-contrib=/usr/win32 --disable-mad --disable-speex
--enable-dvdread --disable-a52 --host=i586-mingw32msvc
CPPFLAGS="-I/usr/win32/include -L/usr/win32/lib" --disable-libgcrypt
--disable-remoteosd LDFLAGS=-L/usr/win32/lib --disable-libva
--disable-alsa && make
...

I get this:

  CCLD   libaccess_avio_plugin.la
i586-mingw32msvc-gcc: unrecognized option '-pthread'
i586-mingw32msvc-gcc: unrecognized option '-pthread'
/usr/lib/gcc/i586-mingw32msvc/4.4.4/../../../../i586-mingw32msvc/bin/ld:
cannot find -lva

Any ideas here?  I already have --disable-libva so that's not it I
don't think...
-roger-


From xtophe at chewa.net  Sat Dec  3 00:38:51 2011
From: xtophe at chewa.net (Christophe Mutricy)
Date: Sat, 3 Dec 2011 00:38:51 +0100
Subject: [vlc-devel] error cross compiling
In-Reply-To: <CAL1QdWeZ3dhBnN+rF=MjjjLeSnWQ5RmNk5=ME2e8F2EAtfXLRQ@mail.gmail.com>
References: <CAL1QdWeZ3dhBnN+rF=MjjjLeSnWQ5RmNk5=ME2e8F2EAtfXLRQ@mail.gmail.com>
Message-ID: <20111202233851.GL29860@chewa.net>

On Fri, Dec 02, 11 at 16:35 -0700, Roger Pack wrote:
>   CCLD   libaccess_avio_plugin.la
> i586-mingw32msvc-gcc: unrecognized option '-pthread'
> i586-mingw32msvc-gcc: unrecognized option '-pthread'
> /usr/lib/gcc/i586-mingw32msvc/4.4.4/../../../../i586-mingw32msvc/bin/ld:
> cannot find -lva
> 
> Any ideas here?  I already have --disable-libva so that's not it I
> don't think...


Check that ./configure detect your win32 libav{codec,format,util,io} and
not your linux ones

-- 
Xtophe


From rogerdpack2 at gmail.com  Sat Dec  3 02:45:45 2011
From: rogerdpack2 at gmail.com (Roger Pack)
Date: Fri, 2 Dec 2011 18:45:45 -0700
Subject: [vlc-devel] --enable-libdvdcss and static dll linking?
Message-ID: <CAL1QdWdit4ZjnFAsZjygC5NvF++yvhZqeJJyurdZ9UM9ajPqTg@mail.gmail.com>

Hello.
I tried the libdvdread patch to --enable-libdvdcss
with cross compiling:

$ ./configure --host=i586-mingw32msvc --prefix=/usr/win32 --enable-libdvdcss
...
checking for dvdcss_interface_2 in -ldvdcss... no
configure: error: You need libdvdcss.so.2 or later

Maybe it needs some cross compile love?

Also next question:
how does vlc create "static" dll's, like
libdvdnav_plugin.dll
that doesn't need libdvdnav-4.dll to function?  I was forced to copy
all the various .dll's to my vlc directory in order to get vlc to work
with my own cross compiled version of libdvdnav.  Any ideas?

Thanks!
-roger-


From jb at videolan.org  Sat Dec  3 03:15:09 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Sat, 3 Dec 2011 03:15:09 +0100
Subject: [vlc-devel] --enable-libdvdcss and static dll linking?
In-Reply-To: <CAL1QdWdit4ZjnFAsZjygC5NvF++yvhZqeJJyurdZ9UM9ajPqTg@mail.gmail.com>
References: <CAL1QdWdit4ZjnFAsZjygC5NvF++yvhZqeJJyurdZ9UM9ajPqTg@mail.gmail.com>
Message-ID: <20111203021509.GA5770@videolan.org>

On Fri, Dec 02, 2011 at 06:45:45PM -0700, Roger Pack wrote :
> Also next question:
> how does vlc create "static" dll's, like
> libdvdnav_plugin.dll
> that doesn't need libdvdnav-4.dll to function?  I was forced to copy
> all the various .dll's to my vlc directory in order to get vlc to work
> with my own cross compiled version of libdvdnav.  Any ideas?

You didn't compile libdvdnav as static.

Best regards,


-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From nkoriyama at gmail.com  Sat Dec  3 03:48:31 2011
From: nkoriyama at gmail.com (Naohiro KORIYAMA)
Date: Sat, 3 Dec 2011 11:48:31 +0900
Subject: [vlc-devel] UPnP module on Windows has been broken
In-Reply-To: <CAOa_K+vCSK_a08w3+PC_099esbh-g3t74_rZadDoG83ew0jwng@mail.gmail.com>
References: <CAOa_K+u7x9axyUDSNB8dEidv6KvNpedw9yvnUmBTPDzAv8FGXw@mail.gmail.com>
	<201112012115.35895.remi@remlab.net>
	<201112012118.18678.remi@remlab.net>
	<CAOa_K+u6OuTtt=WxeZ_b_Vc-6jozAxmP=zPj-RAtCpT5auQi1g@mail.gmail.com>
	<a7970263af9f698da8c79dc7f02923c8@chewa.net>
	<CAOa_K+v4hj-49=jsuROdpyhcF6PNwy3w9LaEOjbZ_VAWDPRy3w@mail.gmail.com>
	<1322820044.2149.22.camel@mirsal-laptop1.mirsal.fr>
	<CAOa_K+u6fYukvQ+xb3EUQGmpCVkUegc9kc_hfb3W45RGqhUNFg@mail.gmail.com>
	<CAOa_K+vCSK_a08w3+PC_099esbh-g3t74_rZadDoG83ew0jwng@mail.gmail.com>
Message-ID: <CAOa_K+tKrL=pDnzmp3HDAm_OJoswm28rsjCPtStnnzf4f5XCdg@mail.gmail.com>

2011/12/2 Naohiro KORIYAMA <nkoriyama at gmail.com>:
> Anyway, on Windows, if ?inet_ntop()/inet_ntop() is called with
> AF_INET6, the function always returns error value.

This is still true, but this is not real cause of the UPnP problem.

inet_ntop() in compat/inet_pton.c has a bug.

iconst char *inet_ntop (int af, const void *src, char *dst, int len)
{
    const unsigned char *b = src;

    switch (af)
    {
        case AF_INET:
            if (snprintf (dst, len, "%hhu.%hhu.%hhu.%hhu",
                          b[0], b[1], b[2], b[3]) <= len)
            {
                errno = ENOSPC;
                return NULL;
            }
            return dst;
    }
    errno = EAFNOSUPPORT;
    return NULL;
}

If af is AF_INET, len is 16(INET_ADDRSTRLEN).
snprintf's result(stored length of dst) can be from 7 to 15. (dst:
0.0.0.0, 255.255.255.255).
snprintf's result is always smaller than size.
So, this funciton always returns NULL and set ENOSPC to errno.

I attached the patch for this problem.
I'm very very sorry for libupnp.

BTW, Should we add implementation to inet_ntop/inet_pton for
AF_INET6(IPv6) in the future?
-- 
KORIYAMA, Naohiro
nkoriyama at gmail.com
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-inet_ntop-always-returns-NULL-on-Windows.patch
Type: application/octet-stream
Size: 816 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111203/d9944c5c/attachment.obj>

From nkoriyama at gmail.com  Sat Dec  3 04:28:55 2011
From: nkoriyama at gmail.com (Naohiro KORIYAMA)
Date: Sat, 3 Dec 2011 12:28:55 +0900
Subject: [vlc-devel] --enable-libdvdcss and static dll linking?
In-Reply-To: <20111203021509.GA5770@videolan.org>
References: <CAL1QdWdit4ZjnFAsZjygC5NvF++yvhZqeJJyurdZ9UM9ajPqTg@mail.gmail.com>
	<20111203021509.GA5770@videolan.org>
Message-ID: <CAOa_K+tVeccH4Uq3Ta72ui5a=xxxpooFOcZhQg4nQd7XPb6BXQ@mail.gmail.com>

2011/12/3 Jean-Baptiste Kempf <jb at videolan.org>:
> On Fri, Dec 02, 2011 at 06:45:45PM -0700, Roger Pack wrote :
>> Also next question:
>> how does vlc create "static" dll's, like
>> libdvdnav_plugin.dll
>> that doesn't need libdvdnav-4.dll to function? ?I was forced to copy
>> all the various .dll's to my vlc directory in order to get vlc to work
>> with my own cross compiled version of libdvdnav. ?Any ideas?
>
> You didn't compile libdvdnav as static.
>

Making contrib with Makefile generated by contrib/bootstrap,
contrib libraries must be compiled as static.

I compiled contrib libraries on my mingw32 on my Mac OS X by the
following steps.
you may change host and build for your environment.

% cd vlc_topdir/contrib
% mkdir win
% cd win
% ../bootstrap --host=i686-pc-mingw32 --build=x86_64-apple-darwin11
% make

And contrib are installed to vlc_topdir/contrib/i686-pc-mingw32 (or
your setting value to host).
It's the easiest and preferred way, I think.

-- 
KORIYAMA, Naohiro
nkoriyama at gmail.com


From nkoriyama at gmail.com  Sat Dec  3 05:39:20 2011
From: nkoriyama at gmail.com (Naohiro KORIYAMA)
Date: Sat, 3 Dec 2011 13:39:20 +0900
Subject: [vlc-devel] Windows 64bit build
Message-ID: <CAOa_K+t6Xsyuzts28iEycQ3Aed5HFrS-qCdP7TcFPOjBjAN2NQ@mail.gmail.com>

Does anyone know about the status of windows 64bit build?

I cannot run win64 build since vlc-1.3.0-git-20111129-0403-win32.
Previous build vlc-1.2.0-pre1-20111122-0104-win32 works fine.
I want to check git but checking changes for a week is a little hard :-(

P.S. I have a little question. Why is Windows 64bit build's name
"-win32" instead of "-win64"?
-- 
KORIYAMA, Naohiro
nkoriyama at gmail.com


From funman at videolan.org  Sat Dec  3 10:11:15 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Sat, 3 Dec 2011 04:11:15 -0500
Subject: [vlc-devel] Windows 64bit build
In-Reply-To: <CAOa_K+t6Xsyuzts28iEycQ3Aed5HFrS-qCdP7TcFPOjBjAN2NQ@mail.gmail.com>
References: <CAOa_K+t6Xsyuzts28iEycQ3Aed5HFrS-qCdP7TcFPOjBjAN2NQ@mail.gmail.com>
Message-ID: <20111203041115.434ac8ff@tok>

Le Sat, 3 Dec 2011 13:39:20 +0900,
Naohiro KORIYAMA <nkoriyama at gmail.com> a ?crit :

> Does anyone know about the status of windows 64bit build?

Yes, I tried various builds until the end of movember and they seemed to
work with quick testing. I don't think I tried builds after the end of
november. We changed the compiler and runtime, so builds after one week
interruption use the new runtime (mingw-w64 2.0.1)

> I cannot run win64 build since vlc-1.3.0-git-20111129-0403-win32.
> Previous build vlc-1.2.0-pre1-20111122-0104-win32 works fine.
> I want to check git but checking changes for a week is a little hard :-(

I'll try to test newer builds tomorrow

> P.S. I have a little question. Why is Windows 64bit build's name
> "-win32" instead of "-win64"?

Nobody (including me) cared to spend time on changing the name

If you have a patch to change the name to win64 that would be nice.

-- 
Rafa?l Carr?


From remi at remlab.net  Sat Dec  3 10:25:42 2011
From: remi at remlab.net (=?iso-8859-15?q?R=E9mi?= Denis-Courmont)
Date: Sat, 3 Dec 2011 11:25:42 +0200
Subject: [vlc-devel] UPnP module on Windows has been broken
In-Reply-To: <CAOa_K+tKrL=pDnzmp3HDAm_OJoswm28rsjCPtStnnzf4f5XCdg@mail.gmail.com>
References: <CAOa_K+u7x9axyUDSNB8dEidv6KvNpedw9yvnUmBTPDzAv8FGXw@mail.gmail.com>
	<CAOa_K+vCSK_a08w3+PC_099esbh-g3t74_rZadDoG83ew0jwng@mail.gmail.com>
	<CAOa_K+tKrL=pDnzmp3HDAm_OJoswm28rsjCPtStnnzf4f5XCdg@mail.gmail.com>
Message-ID: <201112031125.42773.remi@remlab.net>

Le samedi 3 d?cembre 2011 04:48:31 Naohiro KORIYAMA, vous avez ?crit :
> I attached the patch for this problem.
> I'm very very sorry for libupnp.

Merged, thanks.

> BTW, Should we add implementation to inet_ntop/inet_pton for
> AF_INET6(IPv6) in the future?

I don't see the point. It is quite much code. Yet practically any operating 
system version with IPv6 support has inet_ntop() and inet_pton() (including 
Windows).

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From nkoriyama at gmail.com  Sat Dec  3 11:51:14 2011
From: nkoriyama at gmail.com (Naohiro KORIYAMA)
Date: Sat, 3 Dec 2011 19:51:14 +0900
Subject: [vlc-devel] Windows 64bit build
In-Reply-To: <20111203041115.434ac8ff@tok>
References: <CAOa_K+t6Xsyuzts28iEycQ3Aed5HFrS-qCdP7TcFPOjBjAN2NQ@mail.gmail.com>
	<20111203041115.434ac8ff@tok>
Message-ID: <CAOa_K+tf_TCYNiE2+CCkLMp8uEaRLELdqFwSQ-b1LePgdF51ZA@mail.gmail.com>

2011/12/3 Rafa?l Carr? <funman at videolan.org>:
> Le Sat, 3 Dec 2011 13:39:20 +0900,
> Naohiro KORIYAMA <nkoriyama at gmail.com> a ?crit :
>
>> Does anyone know about the status of windows 64bit build?
>
> Yes, I tried various builds until the end of movember and they seemed to
> work with quick testing. I don't think I tried builds after the end of
> november. We changed the compiler and runtime, so builds after one week
> interruption use the new runtime (mingw-w64 2.0.1)

I recently built  mingw-w64 (from git) on Mac OSX,. I can build contrib and VLC.
But it doesn't work on my desktop pc, Windows 7 64bit, Phenom II x4.
If I run VLC, process instantly finishes, no window appears.
This behavior is same as nightly build since the end of November.

> I'll try to test newer builds tomorrow

Thank you!

-- 
KORIYAMA, Naohiro
nkoriyama at gmail.com


From rsatom at gmail.com  Sat Dec  3 11:53:34 2011
From: rsatom at gmail.com (Sergey Radionov)
Date: Sat, 3 Dec 2011 17:53:34 +0700
Subject: [vlc-devel] [PATCH] 6 patches for compatability with libvlc 1.2
Message-ID: <CADOtZwjrr4BmqTTOmW3iXcVUUxg2t3jKS-coCK7ZB-gsaWSWvQ@mail.gmail.com>


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111203/cd17bdba/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0004-common-win32_fullscreen-Added-automatic-Detach-from-.patch
Type: application/octet-stream
Size: 862 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111203/cd17bdba/attachment.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0005-npapi-Added-detaching-of-WindowsManager-from-libvlc-.patch
Type: application/octet-stream
Size: 845 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111203/cd17bdba/attachment-0001.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0006-ActiveX-npapi-fixed-handling-mouse-events-when-vout-.patch
Type: application/octet-stream
Size: 3174 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111203/cd17bdba/attachment-0002.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-common-win32_fullscreen-SetVideoPosScrollPosByVideoP.patch
Type: application/octet-stream
Size: 1526 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111203/cd17bdba/attachment-0003.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0002-common-win32_fullscreen-Attaching-to-libvlc-events-m.patch
Type: application/octet-stream
Size: 19157 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111203/cd17bdba/attachment-0004.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0003-common-win32_fullscreen-hook-operations-moved-to-fun.patch
Type: application/octet-stream
Size: 3093 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111203/cd17bdba/attachment-0005.obj>

From rsatom at gmail.com  Sat Dec  3 13:01:07 2011
From: rsatom at gmail.com (Sergey Radionov)
Date: Sat, 3 Dec 2011 19:01:07 +0700
Subject: [vlc-devel] dependency_libs in libvlc.la from nightly builds for
	windows
Message-ID: <CADOtZwhv34OMdbNm0Jn80Uubvnrn6+tKoQm0EbS0zPknuqy+-w@mail.gmail.com>

dependency_libs='
-L/home/buildslave/slave/nb-win32-trunk/build/contrib/i586-mingw32msvc/lib
/home/buildslave/slave/nb-win32-trunk/build/_win32/lib/libvlccore.la/home/buildslave/slave/nb-win32-trunk/build/contrib/i586-mingw32msvc/lib/
libgpg-error.la -lnetapi32 -lwinmm
/home/buildslave/slave/nb-win32-trunk/build/contrib/i586-mingw32msvc/lib/
libiconv.la -lws2_32 -lmingw32'

on my work system I have no
/home/buildslave/slave/nb-win32-trunk/build/_win32, and there for when I
build under MSys I get:

*libtool: link: warning: library `/E/vlc-bin/vlc/sdk/lib/libvlc.la' was
moved.
grep: /home/buildslave/slave/nb-win32-trunk/build/_win32/lib/libvlccore.la:
No such file or directory
/bin/sed: can't read /home/buildslave/slave/nb-win32-trunk/build/_win32/lib/
libvlccore.la: No such file or directory
libtool: link: `/home/buildslave/slave/nb-win32-trunk/build/_win32/lib/
libvlccore.la' is not a valid libtool archive*

not a big problem, ofcourse, still I can edit it manually, but it
unpleasantly..
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111203/a9b28e76/attachment.html>

From edward.c.wang at compdigitec.com  Sat Dec  3 15:38:21 2011
From: edward.c.wang at compdigitec.com (Edward Wang)
Date: Sat, 3 Dec 2011 09:38:21 -0500
Subject: [vlc-devel] [PATCH] Qt4: add play/pause options under Playback menu
	(#5522)
Message-ID: <CAHqHa+JnsejXT_6iuPujL8mZV-FPsOcebxO4tYCBXjgqFejvCQ@mail.gmail.com>

Hi all:

This patch will add the new feature in bug #5522 (
http://trac.videolan.org/vlc/ticket/5522) and the GCI task. Attached is the
patch file.

Thanks,
Edward Wang
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111203/feb462cb/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: patch5522_v2.patch
Type: application/octet-stream
Size: 2376 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111203/feb462cb/attachment.obj>

From jb at videolan.org  Sat Dec  3 16:13:34 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Sat, 3 Dec 2011 16:13:34 +0100
Subject: [vlc-devel] [PATCH] Qt4: add play/pause options under Playback
 menu (#5522)
In-Reply-To: <CAHqHa+JnsejXT_6iuPujL8mZV-FPsOcebxO4tYCBXjgqFejvCQ@mail.gmail.com>
References: <CAHqHa+JnsejXT_6iuPujL8mZV-FPsOcebxO4tYCBXjgqFejvCQ@mail.gmail.com>
Message-ID: <20111203151334.GA22614@videolan.org>

On Sat, Dec 03, 2011 at 09:38:21AM -0500, Edward Wang wrote :
> This patch will add the new feature in bug #5522 (
> http://trac.videolan.org/vlc/ticket/5522) and the GCI task. Attached is the
> patch file.


> diff --git a/modules/gui/qt4/menus.cpp b/modules/gui/qt4/menus.cpp
> index f85e515..440f76c 100644
> --- a/modules/gui/qt4/menus.cpp
> +++ b/modules/gui/qt4/menus.cpp
> @@ -39,8 +39,6 @@
>  #include <vlc_vout.h>                             /* vout_thread_t */
>  #include <vlc_aout.h>                             /* audio_output_t */
>  
> -#include "menus.hpp"
> -
>  #include "main_interface.hpp"                     /* View modifications */
>  #include "dialogs_provider.hpp"                   /* Dialogs display */
>  #include "input_manager.hpp"                      /* Input Management */
> @@ -59,6 +57,8 @@
>  #include <QStatusBar>
>  #include <QFontMetrics>
>  
> +#include "menus.hpp"
> +

Unneeded change.

>  /*
>    This file defines the main menus and the pop-up menu (right-click menu)
>    and the systray menu (in that order in the file)
> @@ -688,6 +688,7 @@ QMenu *QVLCMenu::NavigMenu( intf_thread_t *p_intf, QMenu *menu )
>  {
>      QAction *action;
>      QMenu *submenu;
> +	input_thread_t* p_input = THEMIM->getInput();

No tabs in sourcecode, please.

> -            case 3: QVLCMenu::RebuildNavigMenu( p_intf, menu ); break;
> +            case 3:
> +                menu->clear();
> +                QVLCMenu::NavigMenu( p_intf, menu );
> +                break;

This is probably wrong and leaking, to do it everytime.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From funman at videolan.org  Sat Dec  3 18:17:20 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Sat, 3 Dec 2011 12:17:20 -0500
Subject: [vlc-devel] Windows 64bit build
In-Reply-To: <CAOa_K+tf_TCYNiE2+CCkLMp8uEaRLELdqFwSQ-b1LePgdF51ZA@mail.gmail.com>
References: <CAOa_K+t6Xsyuzts28iEycQ3Aed5HFrS-qCdP7TcFPOjBjAN2NQ@mail.gmail.com>
	<20111203041115.434ac8ff@tok>
	<CAOa_K+tf_TCYNiE2+CCkLMp8uEaRLELdqFwSQ-b1LePgdF51ZA@mail.gmail.com>
Message-ID: <20111203121720.5dafa870@tok>

Le Sat, 3 Dec 2011 19:51:14 +0900,
Naohiro KORIYAMA <nkoriyama at gmail.com> a ?crit :

> 2011/12/3 Rafa?l Carr? <funman at videolan.org>:
> > Le Sat, 3 Dec 2011 13:39:20 +0900,
> > Naohiro KORIYAMA <nkoriyama at gmail.com> a ?crit :
> >
> >> Does anyone know about the status of windows 64bit build?
> >
> > Yes, I tried various builds until the end of movember and they seemed to
> > work with quick testing. I don't think I tried builds after the end of
> > november. We changed the compiler and runtime, so builds after one week
> > interruption use the new runtime (mingw-w64 2.0.1)
> 
> I recently built  mingw-w64 (from git) on Mac OSX,. I can build contrib and VLC.
> But it doesn't work on my desktop pc, Windows 7 64bit, Phenom II x4.
> If I run VLC, process instantly finishes, no window appears.
> This behavior is same as nightly build since the end of November.
> 
> > I'll try to test newer builds tomorrow
> 
> Thank you!

So the binary effectively crashes for me.

VLC can't save the crashdump, if ran in console i see:
"unhandled vlc exception" => our exception handler is running
"error opening file" => crashdump can't be saved

Here is windbg backtrace (i added the instructions)

0:000> k
Child-SP          RetAddr           Call Site
00000000`0022f200 00000000`72dd066d libvlccore!vlm_MessageAdd+0x6f711  ||| mov byte ptr [rax], 0
00000000`0022f260 00000000`72dd0b96 libvlccore!vlc_GetActionId+0x8d    ||| add rsp, 38h
00000000`0022f2a0 00000000`72d61677 libvlccore!vlc_GetActionId+0x5b6   ||| inc eax
*** ERROR: Symbol file could not be found.  Defaulted to export symbols for C:\Users\Rafa?l Carr?\Downloads\vlc-1.3.0-git-20111203-0102\libvlc.dll - 
00000000`0022fb80 00000000`732c1632 libvlccore!libvlc_InternalInit+0x37
*** ERROR: Module load completed but symbols could not be loaded for image00000000`00400000
00000000`0022fd40 00000000`00401eab libvlc!libvlc_new+0x1c2
00000000`0022fdc0 00000000`004013ce image00000000_00400000+0x1eab



no idea how to debug on/for windows, courmisch suggested using wine but we need wine64 at least

i am not sure why but i don't see all the symbols in the disassembler (i'm using IDA but objdump works too)
it's strange because we don't run strip in the -debug packages afaik


-- 
Rafa?l Carr?


From rsatom at gmail.com  Sun Dec  4 04:51:17 2011
From: rsatom at gmail.com (Sergey Radionov)
Date: Sun, 4 Dec 2011 10:51:17 +0700
Subject: [vlc-devel] [PATCH] common/win32_fullscreen: relicensed as LGPL
Message-ID: <CADOtZwg3spBRFw28iz-+gG++KQSoeWxw3-4VWP_pQpGNmvbtqA@mail.gmail.com>


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111204/088c0665/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-common-win32_fullscreen-relicensed-as-LGPL.patch
Type: application/octet-stream
Size: 3917 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111204/088c0665/attachment.obj>

From jb at videolan.org  Sun Dec  4 04:53:11 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Sun, 4 Dec 2011 04:53:11 +0100
Subject: [vlc-devel] [PATCH] common/win32_fullscreen: relicensed as LGPL
In-Reply-To: <CADOtZwg3spBRFw28iz-+gG++KQSoeWxw3-4VWP_pQpGNmvbtqA@mail.gmail.com>
References: <CADOtZwg3spBRFw28iz-+gG++KQSoeWxw3-4VWP_pQpGNmvbtqA@mail.gmail.com>
Message-ID: <20111204035311.GA4649@videolan.org>

On Sun, Dec 04, 2011 at 10:51:17AM +0700, Sergey Radionov wrote :
> 

OK.

I'll do it.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From rsatom at gmail.com  Sun Dec  4 04:54:11 2011
From: rsatom at gmail.com (Sergey Radionov)
Date: Sun, 4 Dec 2011 10:54:11 +0700
Subject: [vlc-devel] [PATCH] common/win32_fullscreen: relicensed as LGPL
In-Reply-To: <20111204035311.GA4649@videolan.org>
References: <CADOtZwg3spBRFw28iz-+gG++KQSoeWxw3-4VWP_pQpGNmvbtqA@mail.gmail.com>
	<20111204035311.GA4649@videolan.org>
Message-ID: <CADOtZwjmkecwYuvEbwFMkTkXJkPfLwMj+waiHJ1VxoZzbxVUGw@mail.gmail.com>

I hope it is really not problem?

2011/12/4 Jean-Baptiste Kempf <jb at videolan.org>

> On Sun, Dec 04, 2011 at 10:51:17AM +0700, Sergey Radionov wrote :
> >
>
> OK.
>
> I'll do it.
>
> Best regards,
>
> --
> Jean-Baptiste Kempf
> http://www.jbkempf.com/ - +33 672 704 734
> Sent from my Electronic Device
> _______________________________________________
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
> http://mailman.videolan.org/listinfo/vlc-devel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111204/b86f2093/attachment.html>

From jb at videolan.org  Sun Dec  4 04:55:01 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Sun, 4 Dec 2011 04:55:01 +0100
Subject: [vlc-devel] [PATCH] common/win32_fullscreen: relicensed as LGPL
In-Reply-To: <CADOtZwjmkecwYuvEbwFMkTkXJkPfLwMj+waiHJ1VxoZzbxVUGw@mail.gmail.com>
References: <CADOtZwg3spBRFw28iz-+gG++KQSoeWxw3-4VWP_pQpGNmvbtqA@mail.gmail.com>
	<20111204035311.GA4649@videolan.org>
	<CADOtZwjmkecwYuvEbwFMkTkXJkPfLwMj+waiHJ1VxoZzbxVUGw@mail.gmail.com>
Message-ID: <20111204035501.GA4993@videolan.org>

On Sun, Dec 04, 2011 at 10:54:11AM +0700, Sergey Radionov wrote :
> I hope it is really not problem?

As you are the only contributor on that file, no issue.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From rsatom at gmail.com  Sun Dec  4 07:40:07 2011
From: rsatom at gmail.com (Sergey Radionov)
Date: Sun, 4 Dec 2011 13:40:07 +0700
Subject: [vlc-devel] [PATCH] npapi win32: removed useless stuff from mozilla
	window proc.
Message-ID: <CADOtZwgHz1FeOO42KpNLCrQxySSxq342Xu92eNsfQxrGfzo+FA@mail.gmail.com>


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111204/e9d1b613/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-npapi-win32-removed-useless-stuff-from-mozilla-windo.patch
Type: application/octet-stream
Size: 1875 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111204/e9d1b613/attachment.obj>

From jb at videolan.org  Sun Dec  4 14:45:47 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Sun, 4 Dec 2011 14:45:47 +0100
Subject: [vlc-devel] [PATCH] npapi win32: removed useless stuff from
 mozilla window proc.
In-Reply-To: <CADOtZwgHz1FeOO42KpNLCrQxySSxq342Xu92eNsfQxrGfzo+FA@mail.gmail.com>
References: <CADOtZwgHz1FeOO42KpNLCrQxySSxq342Xu92eNsfQxrGfzo+FA@mail.gmail.com>
Message-ID: <20111204134547.GA15512@videolan.org>

On Sun, Dec 04, 2011 at 01:40:07PM +0700, Sergey Radionov wrote :

How are they useless, sorry?

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-npapi-win32-removed-useless-stuff-from-mozilla-windo.patch
Type: application/octet-stream
Size: 1875 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111204/1c31024d/attachment.obj>

From Christoph.Miebach at web.de  Sat Dec  3 15:41:30 2011
From: Christoph.Miebach at web.de (Christoph Miebach)
Date: Sat, 3 Dec 2011 15:41:30 +0100 (CET)
Subject: [vlc-devel] Possible typos in vlcs pot
Message-ID: <1464271236.1119154.1322923290802.JavaMail.fmail@mwmweb041>

Hello!

On translating into german (still a lot of work left) I found some possible typos in the english version:

#: modules/codec/schroedinger.c:152
msgid "Low Pass Ffilter"
msgstr ""

Is it "Ffilter"?



#: modules/control/netsync.c:58
msgid ""
"When set then This VLC instance shall dictate its clock for synchronization "

"This" or "this"?


#: modules/control/netsync.c:63
msgid ""
"The IP address of The network master clock to use for clock synchronization."

"of The" or "of the"?


#: src/libvlc-module.c:204
msgid ""
"This is a ',' separated string, each objects should be prefixed by a '+' or "

"each objects" or "each object"?



#: modules/audio_filter/channel_mixer/mono.c:89
msgid ""
"This option silences all other channels except the selected channel. Choose "
"one from (0=left, 1=right, 2=rear left, 3=rear right, 4=center, 5=left front)"
msgstr ""

Is "5=left front" correct? There is no similar for right front for example...


Final question: When must the translated file arrive to be shipped with the next release?
Not sure, if we get it translated fully, soon.

Regards

Christoph


From jb at videolan.org  Sun Dec  4 15:19:21 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Sun, 4 Dec 2011 15:19:21 +0100
Subject: [vlc-devel] Possible typos in vlcs pot
In-Reply-To: <1464271236.1119154.1322923290802.JavaMail.fmail@mwmweb041>
References: <1464271236.1119154.1322923290802.JavaMail.fmail@mwmweb041>
Message-ID: <20111204141921.GA21628@videolan.org>

On Sat, Dec 03, 2011 at 03:41:30PM +0100, Christoph Miebach wrote :
> On translating into german (still a lot of work left) I found some possible typos in the english version:
> 
> #: modules/codec/schroedinger.c:152
> msgid "Low Pass Ffilter"
> msgstr ""
> 
> Is it "Ffilter"?

Fixed.

> msgid ""
> "When set then This VLC instance shall dictate its clock for synchronization "
> 
> "This" or "this"?

Fixed, differently.

> #: modules/control/netsync.c:63
> msgid ""
> "The IP address of The network master clock to use for clock synchronization."
> 
> "of The" or "of the"?

Fixed.

> #: src/libvlc-module.c:204
> msgid ""
> "This is a ',' separated string, each objects should be prefixed by a '+' or "
> 
> "each objects" or "each object"?

Fixed.

> #: modules/audio_filter/channel_mixer/mono.c:89
> msgid ""
> "This option silences all other channels except the selected channel. Choose "
> "one from (0=left, 1=right, 2=rear left, 3=rear right, 4=center, 5=left front)"
> msgstr ""
> 
> Is "5=left front" correct? There is no similar for right front for example...

I don't see that being wrong, to be honest.
6 might be needed for right front, though...

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From remi at remlab.net  Sun Dec  4 15:25:55 2011
From: remi at remlab.net (=?iso-8859-1?q?R=E9mi?= Denis-Courmont)
Date: Sun, 4 Dec 2011 16:25:55 +0200
Subject: [vlc-devel] Possible typos in vlcs pot
In-Reply-To: <20111204141921.GA21628@videolan.org>
References: <1464271236.1119154.1322923290802.JavaMail.fmail@mwmweb041>
	<20111204141921.GA21628@videolan.org>
Message-ID: <201112041625.57706.remi@remlab.net>

Le dimanche 4 d?cembre 2011 16:19:21 Jean-Baptiste Kempf, vous avez ?crit :
> On Sat, Dec 03, 2011 at 03:41:30PM +0100, Christoph Miebach wrote :
> > On translating into german (still a lot of work left) I found some
> > possible typos in the english version:
> > 
> > #: modules/codec/schroedinger.c:152
> > msgid "Low Pass Ffilter"
> > msgstr ""
> > 
> > Is it "Ffilter"?
> 
> Fixed.
> 
> > msgid ""
> > "When set then This VLC instance shall dictate its clock for
> > synchronization "
> > 
> > "This" or "this"?
> 
> Fixed, differently.
> 
> > #: modules/control/netsync.c:63
> > msgid ""
> > "The IP address of The network master clock to use for clock
> > synchronization."
> > 
> > "of The" or "of the"?
> 
> Fixed.
> 
> > #: src/libvlc-module.c:204
> > msgid ""
> > "This is a ',' separated string, each objects should be prefixed by a '+'
> > or "
> > 
> > "each objects" or "each object"?
> 
> Fixed.
> 
> > #: modules/audio_filter/channel_mixer/mono.c:89
> > msgid ""
> > "This option silences all other channels except the selected channel.
> > Choose " "one from (0=left, 1=right, 2=rear left, 3=rear right,
> > 4=center, 5=left front)" msgstr ""
> > 
> > Is "5=left front" correct? There is no similar for right front for
> > example...
> 
> I don't see that being wrong, to be honest.
> 6 might be needed for right front, though...

In any case, this parenthesis is useless. It duplicates the choice lists. The 
command line already shows the list automatically and the preferences UI does 
not need it.

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From rsatom at gmail.com  Sun Dec  4 15:34:42 2011
From: rsatom at gmail.com (Sergey Radionov)
Date: Sun, 4 Dec 2011 21:34:42 +0700
Subject: [vlc-devel] [PATCH] npapi win32: removed useless stuff from
 mozilla window proc.
In-Reply-To: <20111204134547.GA15512@videolan.org>
References: <CADOtZwgHz1FeOO42KpNLCrQxySSxq342Xu92eNsfQxrGfzo+FA@mail.gmail.com>
	<20111204134547.GA15512@videolan.org>
Message-ID: <CADOtZwgccckdod0zKvx2E=ZewaLUWr7DgU+G_v4nN9yM0vaUxw@mail.gmail.com>

now this window under VLCHolderWnd, and never visible....

2011/12/4 Jean-Baptiste Kempf <jb at videolan.org>

> On Sun, Dec 04, 2011 at 01:40:07PM +0700, Sergey Radionov wrote :
>
> How are they useless, sorry?
>
> Best regards,
>
> --
> Jean-Baptiste Kempf
> http://www.jbkempf.com/ - +33 672 704 734
> Sent from my Electronic Device
>
> _______________________________________________
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
> http://mailman.videolan.org/listinfo/vlc-devel
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111204/d34a6fcf/attachment.html>

From rsatom at gmail.com  Sun Dec  4 15:35:59 2011
From: rsatom at gmail.com (Sergey Radionov)
Date: Sun, 4 Dec 2011 21:35:59 +0700
Subject: [vlc-devel] [PATCH] npapi win32: removed useless stuff from
 mozilla window proc.
In-Reply-To: <CADOtZwgccckdod0zKvx2E=ZewaLUWr7DgU+G_v4nN9yM0vaUxw@mail.gmail.com>
References: <CADOtZwgHz1FeOO42KpNLCrQxySSxq342Xu92eNsfQxrGfzo+FA@mail.gmail.com>
	<20111204134547.GA15512@videolan.org>
	<CADOtZwgccckdod0zKvx2E=ZewaLUWr7DgU+G_v4nN9yM0vaUxw@mail.gmail.com>
Message-ID: <CADOtZwg0Wvd4yXJ4PizUws5=RP-Rrg5t2xgLZ0BZogWydbW7jQ@mail.gmail.com>

2011/12/4 Sergey Radionov <rsatom at gmail.com>

> now this window under VLCHolderWnd, and never visible....
>
and it *content* never visible....


> 2011/12/4 Jean-Baptiste Kempf <jb at videolan.org>
>
>> On Sun, Dec 04, 2011 at 01:40:07PM +0700, Sergey Radionov wrote :
>>
>> How are they useless, sorry?
>>
>> Best regards,
>>
>> --
>> Jean-Baptiste Kempf
>> http://www.jbkempf.com/ - +33 672 704 734
>> Sent from my Electronic Device
>>
>> _______________________________________________
>> vlc-devel mailing list
>> To unsubscribe or modify your subscription options:
>> http://mailman.videolan.org/listinfo/vlc-devel
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111204/61ccdff3/attachment.html>

From jb at videolan.org  Sun Dec  4 15:53:45 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Sun, 4 Dec 2011 15:53:45 +0100
Subject: [vlc-devel] Possible typos in vlcs pot
In-Reply-To: <201112041625.57706.remi@remlab.net>
References: <1464271236.1119154.1322923290802.JavaMail.fmail@mwmweb041>
	<20111204141921.GA21628@videolan.org>
	<201112041625.57706.remi@remlab.net>
Message-ID: <20111204145345.GA30295@videolan.org>

On Sun, Dec 04, 2011 at 04:25:55PM +0200, R?mi Denis-Courmont wrote :
> > I don't see that being wrong, to be honest.
> > 6 might be needed for right front, though...
> 
> In any case, this parenthesis is useless.

Ok, removed.

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From mrdvlpr at gmail.com  Sun Dec  4 16:03:42 2011
From: mrdvlpr at gmail.com (mrdvlpr)
Date: Sun, 4 Dec 2011 17:03:42 +0200
Subject: [vlc-devel] VLCKit.framework and VLCVideoLayer output with VLC 1.11
	problem
Message-ID: <CADk0TOUtUAYDR7UP4NHLMhUx7kiAk0-3uygw0j9azd5VELcARQ@mail.gmail.com>

Hello everyone,

I am trying to play video with VLCVideoLayer instead of VLCVideoView
without any success. (I can hear the movie's audio stream, however)

I have the VLCKit framework compiled correctly, and its test application
running movies without a problem (with some files not playing, but that's a
different story).

My question is:

Did anyone tried doing this or I am wasting my time because VLCVideoLayer
is not working correctly/not supported?

Thanks in advance,

Omer.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111204/fbb6550c/attachment.html>

From jb at videolan.org  Sun Dec  4 16:18:34 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Sun, 4 Dec 2011 16:18:34 +0100
Subject: [vlc-devel] VLCKit.framework and VLCVideoLayer output with VLC
 1.11 problem
In-Reply-To: <CADk0TOUtUAYDR7UP4NHLMhUx7kiAk0-3uygw0j9azd5VELcARQ@mail.gmail.com>
References: <CADk0TOUtUAYDR7UP4NHLMhUx7kiAk0-3uygw0j9azd5VELcARQ@mail.gmail.com>
Message-ID: <20111204151834.GA2699@videolan.org>

On Sun, Dec 04, 2011 at 05:03:42PM +0200, mrdvlpr wrote :
> Did anyone tried doing this or I am wasting my time because VLCVideoLayer
> is not working correctly/not supported?

I am not sure anyone ever tried. But fixing it would be welcome, since
this is the right way for CA.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From gci at paulk.fr  Sun Dec  4 19:01:55 2011
From: gci at paulk.fr (PaulK)
Date: Sun, 04 Dec 2011 19:01:55 +0100
Subject: [vlc-devel] [PATCH] add a way to enable/disable video effects
 via libvlc
In-Reply-To: <e966df51def904cccca19876dcf46aef@chewa.net>
References: <1322429874.1554.13.camel@aldrin>
	<201111282223.13713.remi@remlab.net> <1322587662.1588.11.camel@aldrin>
	<1322690764.1554.3.camel@aldrin>
	<e966df51def904cccca19876dcf46aef@chewa.net>
Message-ID: <1323021715.1607.1.camel@aldrin>

Le jeudi 01 d?cembre 2011 ? 08:06 +0100, R?mi Denis-Courmont a ?crit :
> Please use a correct function prefix. libvlc_set_video_filter() sounds
> like it applies to a whole Libvlc instance.

What about libvlc_video_et_filter() ?
Some other functions are name like "libvlc_video_set_key_input" but
there is also "libvlc_toggle_fullscreen"?  



From gci at paulk.fr  Sun Dec  4 19:04:42 2011
From: gci at paulk.fr (PaulK)
Date: Sun, 04 Dec 2011 19:04:42 +0100
Subject: [vlc-devel] [PATCH] Added libvlc_video_take_snapshot_addr to
 libvlc and related functions to libvlccore
In-Reply-To: <a8a9b166865d290d4beb39c0e809cadc@chewa.net>
References: <1322508588.1586.0.camel@aldrin>
	<20111128203428.GA31114@elivagar.org> <1322513160.1586.1.camel@aldrin>
	<1322515374.1586.2.camel@aldrin> <20111128213008.GA1747@elivagar.org>
	<1322586972.1588.3.camel@aldrin> <4ED5164B.9010103@alexeysokolov.co.cc>
	<1322689414.1554.0.camel@aldrin>
	<a8a9b166865d290d4beb39c0e809cadc@chewa.net>
Message-ID: <1323021882.1607.3.camel@aldrin>

Le jeudi 01 d?cembre 2011 ? 08:07 +0100, R?mi Denis-Courmont a ?crit :
> On Wed, 30 Nov 2011 22:43:34 +0100, PaulK <gci at paulk.fr> wrote:
> > diff --git a/include/vlc/libvlc_media_player.h
> > b/include/vlc/libvlc_media_player.h
> > index a2b5748..b2a5c54 100644
> > --- a/include/vlc/libvlc_media_player.h
> > +++ b/include/vlc/libvlc_media_player.h
> > @@ -35,6 +35,7 @@
> >  extern "C" {
> >  # else
> >  #  include <stdbool.h>
> > +#  include <stdlib.h>
> >  # endif
> >  
> > 
> /*****************************************************************************
> > @@ -1151,6 +1152,25 @@ int
> > libvlc_video_take_snapshot( libvlc_media_player_t *p_mi, unsigned num,
> >                                  unsigned int i_height );
> >  
> >  /**
> > + * Take a snapshot of the current video window and return the snapshot
> > address.
> > + *
> > + * If i_width AND i_height is 0, original size is used.
> > + * If i_width XOR i_height is 0, original aspect-ratio is preserved.
> > + *
> > + * \param p_mi media player instance
> > + * \param num number of video output (typically 0 for the first/only
> > one)
> > + * \param psz_addr place to return the snapshot address (you MUST free
> > it after use)
> > + * \param i_width the snapshot's width
> > + * \param i_height the snapshot's height
> > + * \return the snapshot memory area size on success, -1 if there was a
> > problem
> > + */
> > +LIBVLC_API
> > +ssize_t
> > +libvlc_video_snapshot_get_data( libvlc_media_player_t *p_mi, unsigned
> > num,
> > +                                 void **psz_addr, unsigned int
> > i_width, 
> > +                                 unsigned int i_height );
> > +
> > +/**
> >   * Enable or disable deinterlace filter
> >   *
> >   * \param p_mi libvlc media player
> > diff --git a/lib/libvlc.sym b/lib/libvlc.sym
> > index 6582a96..099cbb8 100644
> > --- a/lib/libvlc.sym
> > +++ b/lib/libvlc.sym
> > @@ -215,6 +215,7 @@ libvlc_video_set_spu
> >  libvlc_video_set_subtitle_file
> >  libvlc_video_set_teletext
> >  libvlc_video_set_track
> > +libvlc_video_snapshot_get_data
> >  libvlc_video_take_snapshot
> >  libvlc_vlm_add_broadcast
> >  libvlc_vlm_add_vod
> > diff --git a/lib/video.c b/lib/video.c
> > index 6e1ea06..3071646 100644
> > --- a/lib/video.c
> > +++ b/lib/video.c
> > @@ -34,6 +34,7 @@
> >  #include <vlc/libvlc_media_player.h>
> >  
> >  #include <vlc_common.h>
> > +#include <vlc_block.h>
> >  #include <vlc_input.h>
> >  #include <vlc_vout.h>
> >  
> > @@ -158,6 +159,63 @@ libvlc_video_take_snapshot( libvlc_media_player_t
> > *p_mi, unsigned num,
> >      return 0;
> >  }
> >  
> > +ssize_t
> > +libvlc_video_snapshot_get_data( libvlc_media_player_t *p_mi, unsigned
> > num,
> > +                                 void **psz_addr, unsigned int
> > i_width, 
> > +                                 unsigned int i_height )
> > +{
> > +    block_t *block_buffer;
> > +    size_t buffer_len = 0;
> > +    char *buffer = NULL;
> > +    ssize_t ret_len;
> > +    video_format_t fmt;
> > +
> > +    vout_thread_t *p_vout = GetVout( p_mi, num );
> > +    if (p_vout == NULL)
> > +    {
> > +        libvlc_printerr ("No usable vout");
> > +        return -1;
> > +    }
> > +
> > +    var_SetInteger( p_vout, "snapshot-width", i_width);
> > +    var_SetInteger( p_vout, "snapshot-height", i_height );
> 
> Not thread-safe.

I know this is not thread safe but this is how it's done in the other
snapshot function "libvlc_video_take_snapshot" (which is actually
worse). How can I fix this? I don't know how we're supposed to handle
this in VLC (mutex?)? 

> > +
> > +    if( vout_GetSnapshot( p_vout, &block_buffer, NULL, &fmt, "png",
> > 500*1000 ) )
> > +    {
> > +        libvlc_printerr ("Can't get the snapshot");
> > +        ret_len = -1;
> > +        goto exit;
> > +    }
> > +
> > +
> > +    buffer_len = block_buffer->i_buffer;
> > +    if( buffer_len <= 0 )
> > +    {
> > +        libvlc_printerr ("Snapshot buffer length is negative or zero");
> > +        ret_len = -1;
> > +        goto exit;
> > +    }
> > +
> > +    buffer = malloc( buffer_len );
> > +    if( buffer == NULL )
> > +    {
> > +        libvlc_printerr ("buffer memory allocation failed");
> > +        ret_len = -1;
> > +        goto exit;
> > +    }
> > +
> > +    psz_addr[0] = buffer;
> > +    memcpy( buffer, block_buffer->p_buffer, buffer_len );
> 
> For the second time, please don't do this. We have buffer management.
> 
> > +
> > +    ret_len = buffer_len;
> > +
> > +exit:
> > +    if( block_buffer != NULL )
> > +        block_Release( block_buffer );
> > +
> > +    return ret_len;
> > +}
> > +
> >  int libvlc_video_get_size( libvlc_media_player_t *p_mi, unsigned num,
> >                             unsigned *restrict px, unsigned *restrict py
> > )
> >  {
> > 
> > 
> > _______________________________________________
> > vlc-devel mailing list
> > To unsubscribe or modify your subscription options:
> > http://mailman.videolan.org/listinfo/vlc-devel
> 




From Christoph.Miebach at web.de  Sun Dec  4 17:12:51 2011
From: Christoph.Miebach at web.de (Christoph Miebach)
Date: Sun, 4 Dec 2011 17:12:51 +0100 (CET)
Subject: [vlc-devel] Next round on typos
Message-ID: <1238298459.762753.1323015171603.JavaMail.fmail@mwmweb024>

Hello!

Version from Nov, 24
http://git.videolan.org/?p=vlc.git;a=blob_plain;f=po/vlc.pot;hb=HEAD

Possible typos (my own english isn't perfect, either):

#: modules/audio_filter/resampler/sr.c:34
msgid ""
"Different resampling algorithm are supported. The best one is slower, while "
"the fast one exhibits low quality."
algorithm
=>
algorithms

#: modules/codec/schroedinger.c:130
msgid ""
"Number of pictures between successive sequenence headers i.e. length of the "
"group of pictures"
sequenence
=>
sequence


#: modules/demux/image.c:62
msgid ""
"Duration in second before simulating an end of file. A negative value means "
"an unlimited play time."

second
=>
seconds



#: modules/demux/ts.c:143
msgid ""
"Seek and position based on a percent byte position, not a PCR generated time "
"position. If seeking doesn't work propery, turn on this option."
propery
=>
properly


#: modules/gui/qt4/qt4.cpp:165
msgid ""
"Display background cone or current album art when not playing.Can be "
"disabled to prevent burning screen."

Missing space after .   playing.Can
Regards

Christoph


From jb at videolan.org  Sun Dec  4 23:43:07 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Sun, 4 Dec 2011 23:43:07 +0100
Subject: [vlc-devel] Next round on typos
In-Reply-To: <1238298459.762753.1323015171603.JavaMail.fmail@mwmweb024>
References: <1238298459.762753.1323015171603.JavaMail.fmail@mwmweb024>
Message-ID: <20111204224307.GA21534@videolan.org>

> Possible typos (my own english isn't perfect, either):

Fixed, thanks a lot.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From edward.c.wang at compdigitec.com  Mon Dec  5 03:07:09 2011
From: edward.c.wang at compdigitec.com (Edward Wang)
Date: Sun, 4 Dec 2011 21:07:09 -0500
Subject: [vlc-devel] [PATCH] Qt4: Add media opened in the command line to
	MRL (#5404)
Message-ID: <CAHqHa++JvuDMK4ykQNZ4HU=pYi8JR67gvmpfGv-z7QNq2oVOiA@mail.gmail.com>

Hi all:

This patch will add the new feature in bug #5404 (
http://trac.videolan.org/vlc/ticket/5404) and the GCI task - Qt4: Add media
opened in the command line to MRL. Attached is the patch file.

Thanks,
Edward Wang
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111204/affa0118/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: patch5404.diff
Type: application/octet-stream
Size: 747 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111204/affa0118/attachment.obj>

From linkfanel at yahoo.fr  Mon Dec  5 03:31:16 2011
From: linkfanel at yahoo.fr (Pierre Ynard)
Date: Mon, 5 Dec 2011 03:31:16 +0100
Subject: [vlc-devel] [PATCH] NEWS about luaintf
Message-ID: <20111205023115.GA26079@via.ecp.fr>

There was a bit of confusion about the future of lua interface
scripts... Is this what we want?


diff --git a/NEWS b/NEWS
index 8bb1c4e..a01e947 100644
--- a/NEWS
+++ b/NEWS
@@ -120,6 +120,8 @@ Interfaces:
  * Mac OS X: support for the native fullscreen mode on OS X Lion
  * Mac OS X: enhanced AppleScript support
  * The rc and telnet lua interfaces were merged into a new "cli" interface
+ * lua: the recommended way to run custom interface scripts is now to pass
+   -I luaintf --lua-intf myscript
  * ncurses: heavy refactor of the complete interface
  * dbus: Rewrite of the main loop to use a more efficient poll-based model
  * dbus: Upgrade to an mpris2 compliant interface, see http://www.mpris.org


-- 
Pierre Ynard
"Une ?me dans un corps, c'est comme un dessin sur une feuille de papier."


From rogerdpack2 at gmail.com  Mon Dec  5 06:00:20 2011
From: rogerdpack2 at gmail.com (Roger Pack)
Date: Sun, 4 Dec 2011 22:00:20 -0700
Subject: [vlc-devel] --enable-libdvdcss and static dll linking?
In-Reply-To: <20111203021509.GA5770@videolan.org>
References: <CAL1QdWdit4ZjnFAsZjygC5NvF++yvhZqeJJyurdZ9UM9ajPqTg@mail.gmail.com>
	<20111203021509.GA5770@videolan.org>
Message-ID: <CAL1QdWcbv7efiNA=WRs34c7BUykRMb2cWO1iCwXgkgFD6F5YNQ@mail.gmail.com>

> You didn't compile libdvdnav as static.

I assume "compile as static" means --enable-static ?
In this particular instance, I did, however, I left --enable-shared on
(I think so anyway).
Anybody know what it will do if it has both static and shared to choose from?

-roger-


From funman at videolan.org  Mon Dec  5 06:14:28 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Mon, 5 Dec 2011 00:14:28 -0500
Subject: [vlc-devel] --enable-libdvdcss and static dll linking?
In-Reply-To: <CAL1QdWcbv7efiNA=WRs34c7BUykRMb2cWO1iCwXgkgFD6F5YNQ@mail.gmail.com>
References: <CAL1QdWdit4ZjnFAsZjygC5NvF++yvhZqeJJyurdZ9UM9ajPqTg@mail.gmail.com>
	<20111203021509.GA5770@videolan.org>
	<CAL1QdWcbv7efiNA=WRs34c7BUykRMb2cWO1iCwXgkgFD6F5YNQ@mail.gmail.com>
Message-ID: <20111205001428.0c2385b0@tok>

Le Sun, 4 Dec 2011 22:00:20 -0700,
Roger Pack <rogerdpack2 at gmail.com> a ?crit :

> > You didn't compile libdvdnav as static.
> 
> I assume "compile as static" means --enable-static ?
> In this particular instance, I did, however, I left --enable-shared on
> (I think so anyway).
> Anybody know what it will do if it has both static and shared to choose from?

I guess it depends of this gcc option:

`-static'
     On systems that support dynamic linking, this prevents linking
     with the shared libraries.  On other systems, this option has no
     effect.

So I assume if there is no "-static" on the gcc command line, the
dynamic version is chosen first.

-- 
Rafa?l Carr?


From remi at remlab.net  Mon Dec  5 07:48:41 2011
From: remi at remlab.net (=?UTF-8?Q?R=C3=A9mi_Denis-Courmont?=)
Date: Mon, 05 Dec 2011 07:48:41 +0100
Subject: [vlc-devel] [PATCH] Qt4: Add media opened in the command line
 to MRL (#5404)
In-Reply-To: <CAHqHa++JvuDMK4ykQNZ4HU=pYi8JR67gvmpfGv-z7QNq2oVOiA@mail.gmail.com>
References: <CAHqHa++JvuDMK4ykQNZ4HU=pYi8JR67gvmpfGv-z7QNq2oVOiA@mail.gmail.com>
Message-ID: <6e288e974450fb6a3a10ef5355aa56ac@chewa.net>

On Sun, 4 Dec 2011 21:07:09 -0500, Edward Wang
<edward.c.wang at compdigitec.com> wrote:
> This patch will add the new feature in bug #5404 (
> http://trac.videolan.org/vlc/ticket/5404) and the GCI task - Qt4: Add
media
> opened in the command line to MRL. Attached is the patch file.

This cannot work. There is no warranty that the playlist is stable and
frozen at that point in the code. For instance, if the command line had a
playlist item in it, the input thread may already be half-way through
parsing the playlist.

You cannot do this.

-- 
R?mi Denis-Courmont
http://www.remlab.net/


From edward.c.wang at compdigitec.com  Mon Dec  5 12:04:08 2011
From: edward.c.wang at compdigitec.com (Edward Wang)
Date: Mon, 5 Dec 2011 06:04:08 -0500
Subject: [vlc-devel] vlc-devel Digest, Vol 55, Issue 24
In-Reply-To: <mailman.17.1323082804.24312.vlc-devel@videolan.org>
References: <mailman.17.1323082804.24312.vlc-devel@videolan.org>
Message-ID: <CAHqHa++jw+eojiEfvbyQnP8cZXfVDMAV0vjRSCWOtZw+u5WV5A@mail.gmail.com>

On Mon, Dec 5, 2011 at 6:00 AM, <vlc-devel-request at videolan.org> wrote:

> Send vlc-devel mailing list submissions to
>        vlc-devel at videolan.org
>
> To subscribe or unsubscribe via the World Wide Web, visit
>        http://mailman.videolan.org/listinfo/vlc-devel
> or, via email, send a message with subject or body 'help' to
>        vlc-devel-request at videolan.org
>
> You can reach the person managing the list at
>        vlc-devel-owner at videolan.org
>
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of vlc-devel digest..."
>
>
> Today's Topics:
>
>   1. Re: [PATCH] Qt4: Add media opened in the command line to MRL
>      (#5404) (R?mi Denis-Courmont)
>
>
> ----------------------------------------------------------------------
>
> Message: 1
> Date: Mon, 05 Dec 2011 07:48:41 +0100
> From: R?mi Denis-Courmont <remi at remlab.net>
> To: Mailing list for VLC media player developers
>        <vlc-devel at videolan.org>
> Subject: Re: [vlc-devel] [PATCH] Qt4: Add media opened in the command
>        line to MRL (#5404)
> Message-ID: <6e288e974450fb6a3a10ef5355aa56ac at chewa.net>
> Content-Type: text/plain; charset=UTF-8
>
> On Sun, 4 Dec 2011 21:07:09 -0500, Edward Wang
> <edward.c.wang at compdigitec.com> wrote:
> > This patch will add the new feature in bug #5404 (
> > http://trac.videolan.org/vlc/ticket/5404) and the GCI task - Qt4: Add
> media
> > opened in the command line to MRL. Attached is the patch file.
>
> This cannot work. There is no warranty that the playlist is stable and
> frozen at that point in the code. For instance, if the command line had a
> playlist item in it, the input thread may already be half-way through
> parsing the playlist.
>
> You cannot do this.
>
> --
> R?mi Denis-Courmont
> http://www.remlab.net/
>
>
> ------------------------------
>
> _______________________________________________
> vlc-devel mailing list
> vlc-devel at videolan.org
> http://mailman.videolan.org/listinfo/vlc-devel
>
>
> End of vlc-devel Digest, Vol 55, Issue 24
> *****************************************
>

Then how else might I do it? Perhaps with a mutex or locking mechanism?
Please help me.

Thanks,
Edward Wang
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111205/b5638eae/attachment.html>

From Christoph.Miebach at web.de  Mon Dec  5 18:27:03 2011
From: Christoph.Miebach at web.de (Christoph Miebach)
Date: Mon, 5 Dec 2011 18:27:03 +0100 (CET)
Subject: [vlc-devel] Use Launchpad for our german translation of vlc?
Message-ID: <55709484.221683.1323106023237.JavaMail.fmail@mwmweb043>

Hello!

We want to set up an online version of the german translation on launchpad.
It shall be used mostly to simplify our work on the translation.

Are there any (license) problems with that?

Or, better:

May we do so?
What will we have to keep in mind?

Regards

Christoph Miebach & Mario Siegmann


From jb at videolan.org  Mon Dec  5 22:11:27 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Mon, 5 Dec 2011 22:11:27 +0100
Subject: [vlc-devel] [vlc-commits] Qt: rename QVLCMenu to
	QVLCMenuManager.
In-Reply-To: <20111205210856.21E1F142F3E@albiero.videolan.org>
References: <20111205210856.21E1F142F3E@albiero.videolan.org>
Message-ID: <20111205211127.GA19761@videolan.org>

On Mon, Dec 05, 2011 at 10:08:56PM +0100, Francois Cartegnie wrote :
> vlc | branch: master | Francois Cartegnie <fcvlcdev at free.fr> | Fri Dec  2 14:58:17 2011 +0100| [76f155bda6f6f3b60f97587228a589874ac47227] | committer: Francois Cartegnie
> 
> Qt: rename QVLCMenu to QVLCMenuManager.

Please don't.

*Manager is a bad name for a class. Because we've done that in the past,
but that is wrong.

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From fcvlcdev at free.fr  Mon Dec  5 22:15:20 2011
From: fcvlcdev at free.fr (Francois Cartegnie)
Date: Mon, 05 Dec 2011 22:15:20 +0100
Subject: [vlc-devel] [vlc-commits] Qt: rename QVLCMenu
	to	QVLCMenuManager.
In-Reply-To: <20111205211127.GA19761@videolan.org>
References: <20111205210856.21E1F142F3E@albiero.videolan.org>
	<20111205211127.GA19761@videolan.org>
Message-ID: <4EDD3468.8020109@free.fr>

On 05/12/2011 22:11, Jean-Baptiste Kempf wrote:
> *Manager is a bad name for a class. Because we've done that in the past,
> but that is wrong.

Can't match *factory, or *builder. rename to Toolkit ?

Francois


From jb at videolan.org  Mon Dec  5 22:42:42 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Mon, 5 Dec 2011 22:42:42 +0100
Subject: [vlc-devel] [PATCH] yadif : Add SSSE3 and SSE2 support. porting
 from FFmpeg.
In-Reply-To: <CAOa_K+vkZK2uO4jrbqUXDkdW4B75v+Dzgg4Joc0089hj4Kf3tg@mail.gmail.com>
References: <CAOa_K+vdRUzBj4cE9qoLyU19+NYnVuPx03UqdOEXgkT6fw60Gg@mail.gmail.com>
	<201111260919.16263.remi@remlab.net>
	<CAOa_K+uZt76_CgJ9DZHVa5-a0CD-wNJFy=scd1Q4xVr8DCx_xQ@mail.gmail.com>
	<20111126024120.2709a235@tok> <20111126112504.GA32214@videolan.org>
	<CAOa_K+svS3RvmmdpwuvgAvYJ=svjTf2CQqe93yDSO7TEwR23Mg@mail.gmail.com>
	<CAOa_K+vkZK2uO4jrbqUXDkdW4B75v+Dzgg4Joc0089hj4Kf3tg@mail.gmail.com>
Message-ID: <20111205214242.GA27741@videolan.org>

On Sun, Nov 27, 2011 at 09:19:17AM +0900, Naohiro KORIYAMA wrote :
> 2011/11/27 Naohiro KORIYAMA <nkoriyama at gmail.com>:
> > 2011/11/26 Jean-Baptiste Kempf <jb at videolan.org>:
> >> However, we have a yadif filter that is largely tested, and seeing how
> >> costly it is in terms of CPU, any speed improvement, notably with newer
> >> CPUs is very very welcome. Especially if it is usable at 1080i.
> >>
> >> So, please fix the clobber list and apply.
> >
> > I changed the clobber list.
> > But I'm not sure how to treat with pb_1, pw_1 at yadif_complete.h.
> > I added these variables to the clobber list and can compile it. Also
> > Filtering using yadif deinterlacer is no problem.
> > But I'm not so familiar with the gcc assemblar, so I want to hear my
> > modification is OK or NG.
> 
> I completely misunderstood about the clobber list. I don't have to
> care about pb_1 and pw_1.
> So, this patch is maybe OK for the clobber list issue.

http://pastebin.com/4uMWVbRV

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From cyzicusmedia at gmail.com  Tue Dec  6 05:10:13 2011
From: cyzicusmedia at gmail.com (Derek Scott)
Date: Tue, 06 Dec 2011 11:10:13 +0700
Subject: [vlc-devel]  Compiling VLC in MacOS under Lion 10.7.2
Message-ID: <4EDD95A5.2020905@gmail.com>


Hello everyone,

I've been trying to compile VLC from the git repository using the 
instructions at http://wiki.videolan.org/OSXCompile.

1.  I successfully get the development tools by bootstrapping in 
vlc/extras/tools.

2.  I select 10.6 SDK by $ export OSX_VERSION=10.6.

3.  Prepare 3rd Party Libraries.  After creating the osx directory in 
contrib, I do

  $ ../bootstrap --host=x86_64-apple-darwin11 --build=x86_64-apple-darwin11

Since I'm running in 64-bit in OS 10.7.2, I thought my host should be 
"--host=x86_64-apple-darwin11" and my build should be 
"--build=x86_64-apple-darwin11."

4.  Prebuilt libraries.

I try make prebuilt, but

ftp://ftp.videolan.org/pub/videolan/contrib/x86_64-apple-darwin11/vlc-contrib-x86_64-apple-darwin11-latest.tar.bz2"

fails because there's no file there.

5.  Build your own libraries.

I try

make -j4 .gettext

This errors out with the following:

In file included from ./bindtextdom.c:24:
/Developer/SDKs/MacOSX10.6.sdk/usr/include/stddef.h:74: error: two or 
more data types in declaration specifiers
In file included from ./gettextP.h:23,
                  from ./dgettext.c:23:
/Developer/SDKs/MacOSX10.6.sdk/usr/include/stddef.h:74: error: two or 
more data types in declaration specifiers
In file included from ./gettextP.h:23,
                  from ./dcgettext.c:23:
/Developer/SDKs/MacOSX10.6.sdk/usr/include/stddef.h:74: error: two or 
more data types in declaration specifiers
In file included from ./gettextP.h:23,
                  from ./gettext.c:30:
/Developer/SDKs/MacOSX10.6.sdk/usr/include/stddef.h:74: error: two or 
more data types in declaration specifiers
make[4]: *** [dgettext.lo] Error 1
make[4]: *** Waiting for unfinished jobs....
make[4]: *** [gettext.lo] Error 1
make[4]: *** [dcgettext.lo] Error 1
make[4]: *** [bindtextdom.lo] Error 1
make[3]: *** [all-recursive] Error 1
make[2]: *** [all] Error 2
make[1]: *** [all-recursive] Error 1
make: *** [.gettext] Error 2

--I'd appreciate your suggestions.

Best regards,

Derek Scott






From beauze.h at gmail.com  Tue Dec  6 13:09:55 2011
From: beauze.h at gmail.com (=?UTF-8?Q?Hugo_Beauz=C3=A9e=2DLuyssen?=)
Date: Tue, 6 Dec 2011 13:09:55 +0100
Subject: [vlc-devel] [Patches] Dash
In-Reply-To: <CAK-C75TP_Wwk9A5=OEUCatK3XvkE_yR457Naim3t37+8PEXiPA@mail.gmail.com>
References: <CAK-C75QQttr4Mpb=B=YAVKgTpMrOeFpDVU9a4wf_6fRAHxFA9A@mail.gmail.com>
	<20111201211153.GC28866@videolan.org>
	<CAK-C75TP_Wwk9A5=OEUCatK3XvkE_yR457Naim3t37+8PEXiPA@mail.gmail.com>
Message-ID: <CAK-C75RMTbvADqo50U0xSzKo+z37MdHs1U_xA6dca--Xgt2x5Q@mail.gmail.com>

On Fri, Dec 2, 2011 at 12:31 AM, Hugo Beauz?e-Luyssen
<beauze.h at gmail.com> wrote:
> On Thu, Dec 1, 2011 at 10:11 PM, Jean-Baptiste Kempf <jb at videolan.org> wrote:
>> On Thu, Dec 01, 2011 at 12:37:19PM +0100, Hugo Beauz?e-Luyssen wrote :
>>> As the patch has been submitted randomly in two other threads, I'll
>>> post all the dash patches that I have in local to ease up the
>>> follow-up.
>>> The patches that are still being discussed are patches 6 through 8
>>
>> 1, 3 applied.
>> 2 doesn't compile here.
> I have a patch to solve this, but on another workstation. Will send it tomorrow
>

Woops, a bit late. These patches should fix this problem.

>> 4 is deprecated (applied the next one)
>> 5 doesn't apply

Should apply now.

> Will check tomorrow.
>
>> 6-8 are still discussed.
>

Patch 6 has been split (attached patches 1 & 5)

> Still waiting for feedback!!
>
>>
>> Thanks a lot.
>>
>> Best regards,
>>

Best regards,

-- 
Hugo Beauz?e-Luyssen
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0003-dash-Fixing-NullManager-interface-implementation.patch
Type: text/x-patch
Size: 1615 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111206/ac75a888/attachment.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-dash-DashManager-Adding-a-getMpdManager-method.patch
Type: text/x-patch
Size: 1461 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111206/ac75a888/attachment-0001.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0005-dash-Don-t-try-to-do-anything-if-we-can-t-get-a-vali.patch
Type: text/x-patch
Size: 9275 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111206/ac75a888/attachment-0002.bin>

From beauze.h at gmail.com  Tue Dec  6 13:11:10 2011
From: beauze.h at gmail.com (=?UTF-8?Q?Hugo_Beauz=C3=A9e=2DLuyssen?=)
Date: Tue, 6 Dec 2011 13:11:10 +0100
Subject: [vlc-devel] [Patch] Adding an helper to convert iso8601
 durations to seconds
In-Reply-To: <CAK-C75TPsc1Lh07NNXFnmzW5ZQgSQbkfm1S2Vfb=e2Nc11Z_Tw@mail.gmail.com>
References: <CAK-C75R5BPUud6z2QxrFNNnh1wMFktu48O59oXi2bAUqZ0mYMw@mail.gmail.com>
	<4ED62AB1.4080208@web.de>
	<CAK-C75RzQhAOxdHZfqijb2G_p27GdVCSYu4nhcakwwW1dD+kuQ@mail.gmail.com>
	<CAK-C75TPsc1Lh07NNXFnmzW5ZQgSQbkfm1S2Vfb=e2Nc11Z_Tw@mail.gmail.com>
Message-ID: <CAK-C75SA9p6Sfv4YoGe6=V8d8WG00goDi9ACOmJzzkJDFM=PVw@mail.gmail.com>

On Wed, Nov 30, 2011 at 4:41 PM, Hugo Beauz?e-Luyssen
<beauze.h at gmail.com> wrote:
> On Wed, Nov 30, 2011 at 2:45 PM, Hugo Beauz?e-Luyssen
> <beauze.h at gmail.com> wrote:
>> On Wed, Nov 30, 2011 at 2:08 PM, Tobias G?ntner <fatbull at web.de> wrote:
>>> Am 30.11.2011 11:05, schrieb Hugo Beauz?e-Luyssen:
>>>> +/*
>>>> + ?Decodes a duration as defined by ISO 8601
>>>> + ?http://en.wikipedia.org/wiki/ISO_8601#Durations
>>>> + ?@param str A null-terminated string to convert
>>>
>>> Perhaps a short example of a valid input string would be useful, so
>>> readers can see at first glance what kind of data this function expects,
>>> without looking it up.
>>>
>>
>> Yep I'm going to write some tests in a this afternoon or tomorrow, but
>> I could add a link to the format description and an exemple in the
>> comments.
>>
>>>> + ?@return: The duration in seconds. 0 if an error occured.
>>>
>>> This function usually does not return 0 if the input is invalid.
>>>
>>
>> Yep i'm going to change the return type as a signed int 64, so -1 is an error.
>>
>>>> + */
>>>> +uint64_t str_iso8601_duration_to_seconds( const char *psz_duration )
>>>> +{
>>>> + ? ?bool ? ? ? ?timeDesignatorReached = false;
>>>> + ? ?double ? ? ?mul = 0;
>>>
>>> I think an integer should be fine.
>>>
>>
>> Well in the exemple file I've got
>> (http://www-itec.uni-klu.ac.at/ftp/datasets/mmsys12/Selected_VLC_Plugin_MPDs/BigBuckBunny_15s_DASH_VLC_Compatible_4-Representations_720p.mpd)
>> there are some decimal values, so I try to reduce the loss of
>> precision when converting by using doubles.
>>
>>>> + ? ?mtime_t ? ? res = 0;
>>>
>>> Shouldn't this type and the return type be the same?
>>>
>>
>> I was using mtime_t at first, it seems I forgot to change this one, my bad.
>>
>> Since the line below is a direct pointer dereferencement without any
>> prior check, this was intended, but a check could be added.
>>
>>>> +
>>>> + ? ?if ( *psz_duration++ != 'P' )
>>>> + ? ? ? ?return 0;
>>>> + ? ?for ( const char* nptr = psz_duration; *psz_duration; ++psz_duration )
>>>> + ? ?{
>>>> + ? ? ? ?switch( *psz_duration )
>>>> + ? ? ? ?{
>>>> + ? ? ? ? ? ?case 'M':
>>>> + ? ? ? ? ? ?{
>>>
>>> I find this a little confusing because letters normally follow numbers
>>> in the input, yet you somehow parse letters first. ;)
>>>
>>
>> That is absolutely right...
>>
>>>> + ? ? ? ? ? ? ? ?//M can mean month or minutes, if the 'T' flag has been reached.
>>>> + ? ? ? ? ? ? ? ?//We don't handle months though.
>>>> + ? ? ? ? ? ? ? ?if ( timeDesignatorReached == true )
>>>> + ? ? ? ? ? ? ? ? ? ?mul = 60.0;
>>>> + ? ? ? ? ? ? ? ?break ;
>>>> + ? ? ? ? ? ?}
>>>> + ? ? ? ? ? ?case 'Y':
>>>> + ? ? ? ? ? ?case 'W':
>>>> + ? ? ? ? ? ? ? ?break ; //Don't handle this duration.
>>>
>>> Maybe return an error here?
>>>
>>
>> I don't really know... warning should be mandatory, an error may be to
>> agressive...
>>
>>>> + ? ? ? ? ? ?case 'D':
>>>> + ? ? ? ? ? ? ? ?mul = 86400.0;
>>>> + ? ? ? ? ? ? ? ?break ;
>>>> + ? ? ? ? ? ?case 'T':
>>>> + ? ? ? ? ? ? ? ?timeDesignatorReached = true;
>>>> + ? ? ? ? ? ? ? ?break ;
>>>> + ? ? ? ? ? ?case 'H':
>>>> + ? ? ? ? ? ? ? ?mul = 3600.0;
>>>> + ? ? ? ? ? ? ? ?break ;
>>>> + ? ? ? ? ? ?case 'S':
>>>> + ? ? ? ? ? ? ? ?mul = 1.0;
>>>> + ? ? ? ? ? ? ? ?break ;
>>>> + ? ? ? ? ? ?default:
>>>> + ? ? ? ? ? ? ? ?continue ;
>>>> + ? ? ? ?}
>>>> + ? ? ? ?//It means we went trough some digits. So we must compute a number
>>>> + ? ? ? ?if ( nptr != psz_duration )
>>>> + ? ? ? ?{
>>>> + ? ? ? ? ? ?double tmp = strtold( nptr, NULL );
>>>
>>> I think you can achieve a slightly cleaner design if you utilize the
>>> second parameter of that function. The strtoX functions store the
>>> address of the first invalid character in there, so all you have to do
>>> is basically:
>>>
>>
>> Yep since I'm doing it in the wrong way, using the second parameter
>> makes a lot of sense now :D
>>
>>> 1. Call strtoul(psz_duration, &endptr, 10)
>>> 2. Check the character at *endptr
>>> 3. Repeat with psz_duration = endptr + 1
>>>
>>>> + ? ? ? ? ? ?res += (uint64_t)(tmp * mul);
>>>> + ? ? ? ? ? ?nptr = psz_duration;
>>>> + ? ? ? ?}
>>>> + ? ? ? ?nptr++;
>>>> + ? ?}
>>>> + ? ?return res;
>>>> +}
>>>
>>> Regards,
>>> Tobias
>>>
>>
>> Thanks for the review!
>> Regards,
>>
>
> Here is a new version, along with some unit tests.
>
> Regards,
>
>

Hi,

Could somebody review these patches?

Thanks in advance,
Best regards,


-- 
Hugo Beauz?e-Luyssen


From fatbull at web.de  Tue Dec  6 13:17:51 2011
From: fatbull at web.de (=?ISO-8859-15?Q?Tobias_G=FCntner?=)
Date: Tue, 06 Dec 2011 13:17:51 +0100
Subject: [vlc-devel] [Patches] Dash: mostly cosmetics
In-Reply-To: <CAK-C75TCOTO2dUUHz8HDPk4X395ifWNnO5bdzS2=JBJtOtkL_Q@mail.gmail.com>
References: <CAK-C75SN=k0WL==GiO89Fm1VmiqapCCZ7k1uOefyg9brc=Aafw@mail.gmail.com>
	<CAK-C75TGwHkm5iTZpydXUe3JGeW5CASrxrsu7xe3CYc5veb9Eg@mail.gmail.com>
	<CAK-C75S4oXDKecXAxwZyP6qfFRh0Q+85iPgYtpt23Yx0SWF7mA@mail.gmail.com>
	<201111241914.49264.remi@remlab.net>
	<CAK-C75TcveJQ3LeKUxA4JzupLiN5JAyUMi5=tGC+QXiTrqFhPQ@mail.gmail.com>
	<CAK-C75QRRLSqe6EKSsNEfnJGwOnjA8b80_t-aa245=bq167Rrg@mail.gmail.com>
	<4ECFD86A.3050305@web.de>
	<CAK-C75QujbKk9EP2M4aDSw26ttq8+H2ALU=RhUiraUB-BzAu4g@mail.gmail.com>
	<4ED47565.1090602@web.de>
	<CAK-C75QcXCDoJF7cpjZQRDHf30soO4bK5-90K57r98ycNb11Kg@mail.gmail.com>
	<4ED4CC73.6010008@web.de>
	<CAK-C75SiVvUN_tnh+pvo+H5xCCNN+_Afi3qG+ZzsSMthAf8QaQ@mail.gmail.com>
	<4ED71FE0.5070003@web.de>
	<CAK-C75TCOTO2dUUHz8HDPk4X395ifWNnO5bdzS2=JBJtOtkL_Q@mail.gmail.com>
Message-ID: <4EDE07EF.2030803@web.de>

Am 01.12.2011 12:15, schrieb Hugo Beauz?e-Luyssen:
> That's exactly what I'm trying to do. However I guess your point here
> is STL is "our" code, while Qt is not. However the point stays, to me.

I was referring to Dash only. Qt and STL are both out of scope. Whether 
Qt does or does not handle exceptions is IMHO irrelevant to our discussion.

> We should check for bad_allow, bad_cast if we use references,
> bad_typeid and ios_base::failure.
> Saying that I've never encountered a code catching these exceptions
> would be falacious, but true nevertheless.

I'm not saying those exceptions must be caught explicitly. IMHO a 
generic catch(std::exception&) at the point of entry is often 
sufficient, and I'm sure you have encountered *that* before. ;)

> I'm not saying we should get read of the STL exceptions, just most of
> those in Dash, that are useless most of the time.

Oh, I see. I assumed you wanted to remove all exceptions, including 
non-Dash exceptions.

> Let's take a code sample :
>
> Representation.cpp :
>
> SegmentInfo*        Representation::getSegmentInfo          () //...
> {
>      if(this->segmentInfo == NULL)
>          throw ElementNotPresentException();
>
>      return this->trickModeType;
> }
>
> BasicCMManager.cpp:
>
> try
>      {
>          SegmentInfo* info = rep->getSegmentInfo();
>          //....
>      }
>      catch(ElementNotPresentException&e)
>      {
>          /*TODO Debug */
>      }
>
> The error is handled twice. One so we can raise, and one to catch the error.
> It looks clearer to me just to return the pointer, regardless of it's
> value, and check it just once.

I agree exceptions seem to be overkill in this case. But before 
rewriting it, please consider:
* Is BasicCMManager::getSegments really the best place to handle this error?
* Is returning an empty/incomplete result vector really the best way to 
handle this error?
* Is segmentInfo really optional?
* Should the error be propagated further?

> Again, I'm not saying anything about STL's exceptions. They're not
> handled for now anyway.

IMHO this must definitely be fixed. Not catching bad_alloc (or a base 
class thereof) is like ignoring the return value of malloc, i.e, a bug.

Regards,
Tobias



From beauze.h at gmail.com  Tue Dec  6 13:43:48 2011
From: beauze.h at gmail.com (=?UTF-8?Q?Hugo_Beauz=C3=A9e=2DLuyssen?=)
Date: Tue, 6 Dec 2011 13:43:48 +0100
Subject: [vlc-devel] [Patches] Dash: mostly cosmetics
In-Reply-To: <4EDE07EF.2030803@web.de>
References: <CAK-C75SN=k0WL==GiO89Fm1VmiqapCCZ7k1uOefyg9brc=Aafw@mail.gmail.com>
	<CAK-C75TGwHkm5iTZpydXUe3JGeW5CASrxrsu7xe3CYc5veb9Eg@mail.gmail.com>
	<CAK-C75S4oXDKecXAxwZyP6qfFRh0Q+85iPgYtpt23Yx0SWF7mA@mail.gmail.com>
	<201111241914.49264.remi@remlab.net>
	<CAK-C75TcveJQ3LeKUxA4JzupLiN5JAyUMi5=tGC+QXiTrqFhPQ@mail.gmail.com>
	<CAK-C75QRRLSqe6EKSsNEfnJGwOnjA8b80_t-aa245=bq167Rrg@mail.gmail.com>
	<4ECFD86A.3050305@web.de>
	<CAK-C75QujbKk9EP2M4aDSw26ttq8+H2ALU=RhUiraUB-BzAu4g@mail.gmail.com>
	<4ED47565.1090602@web.de>
	<CAK-C75QcXCDoJF7cpjZQRDHf30soO4bK5-90K57r98ycNb11Kg@mail.gmail.com>
	<4ED4CC73.6010008@web.de>
	<CAK-C75SiVvUN_tnh+pvo+H5xCCNN+_Afi3qG+ZzsSMthAf8QaQ@mail.gmail.com>
	<4ED71FE0.5070003@web.de>
	<CAK-C75TCOTO2dUUHz8HDPk4X395ifWNnO5bdzS2=JBJtOtkL_Q@mail.gmail.com>
	<4EDE07EF.2030803@web.de>
Message-ID: <CAK-C75SV21xePRH-8pnJ-P4HMK=USxbOQtMVhRvwMGZd2_fG9w@mail.gmail.com>

Hi,

On Tue, Dec 6, 2011 at 1:17 PM, Tobias G?ntner <fatbull at web.de> wrote:
> Am 01.12.2011 12:15, schrieb Hugo Beauz?e-Luyssen:
>
>> That's exactly what I'm trying to do. However I guess your point here
>> is STL is "our" code, while Qt is not. However the point stays, to me.
>
>
> I was referring to Dash only. Qt and STL are both out of scope. Whether Qt
> does or does not handle exceptions is IMHO irrelevant to our discussion.
>
>
>> We should check for bad_allow, bad_cast if we use references,
>> bad_typeid and ios_base::failure.
>> Saying that I've never encountered a code catching these exceptions
>> would be falacious, but true nevertheless.
>
>
> I'm not saying those exceptions must be caught explicitly. IMHO a generic
> catch(std::exception&) at the point of entry is often sufficient, and I'm
> sure you have encountered *that* before. ;)
>

I can live with that indeed :)

>
>> I'm not saying we should get read of the STL exceptions, just most of
>> those in Dash, that are useless most of the time.
>
>
> Oh, I see. I assumed you wanted to remove all exceptions, including non-Dash
> exceptions.
>

At last we understand each other \o/

>
>> Let's take a code sample :
>>
>> Representation.cpp :
>>
>> SegmentInfo* ? ? ? ?Representation::getSegmentInfo ? ? ? ? ?() //...
>> {
>> ? ? if(this->segmentInfo == NULL)
>> ? ? ? ? throw ElementNotPresentException();
>>
>> ? ? return this->trickModeType;
>> }
>>
>> BasicCMManager.cpp:
>>
>> try
>> ? ? {
>> ? ? ? ? SegmentInfo* info = rep->getSegmentInfo();
>> ? ? ? ? //....
>> ? ? }
>> ? ? catch(ElementNotPresentException&e)
>> ? ? {
>> ? ? ? ? /*TODO Debug */
>> ? ? }
>>
>> The error is handled twice. One so we can raise, and one to catch the
>> error.
>> It looks clearer to me just to return the pointer, regardless of it's
>> value, and check it just once.
>
>
> I agree exceptions seem to be overkill in this case. But before rewriting
> it, please consider:

That was exactly my point :)

> * Is BasicCMManager::getSegments really the best place to handle this error?
Most of the error handling could be done once and for all after the
parsing occurred.

> * Is returning an empty/incomplete result vector really the best way to
> handle this error?

If you mean returning a newly constructed empty vector, then no, I
don't like that very much. However, if we were to return a pre-built
vector, stored in the appropriate instance, I don't have any issue
returning an empty vector.
As said before, once the file was parsed, we can know if there are
some segments or not, and if there's none, that's an error, so we can
either ignore the representation, of the file if no valid
representation may be found, so if an empty vector is an error,
processing will stop before.

> * Is segmentInfo really optional?

If I understand the standard correctly (?5.4.4), if the Representation
tag doesn't contain any SegmentInfo, the Group.SegmentInfoDefault tag
must be used (and it must be combined with every present SegmentInfo
to compute available information about the segment).

> * Should the error be propagated further?
>

Well it doesn't seem like an error anymore.
I'll have to read the code again for that specific case.

>
>> Again, I'm not saying anything about STL's exceptions. They're not
>> handled for now anyway.
>
>
> IMHO this must definitely be fixed. Not catching bad_alloc (or a base class
> thereof) is like ignoring the return value of malloc, i.e, a bug.
>

I agree.
There's quite a good chance many of the dash errors currently handled
by the code may be better addressed by reading the standard again,
using specified default values.
For the "unrecoverable" errors, I'm more in favor of returning a good
old NULL pointer in the getter (or even an empty vector), but having a
prior check, just after the parse operation, to check the file can be
used.

>
> Regards,
> Tobias
>

Regards,

-- 
Hugo Beauz?e-Luyssen


From fcvlcdev at free.fr  Tue Dec  6 14:13:08 2011
From: fcvlcdev at free.fr (Francois Cartegnie)
Date: Tue,  6 Dec 2011 14:13:08 +0100
Subject: [vlc-devel] [PATCH] vaapi: fix profiles detection.
Message-ID: <1323177188-14117-1-git-send-email-fcvlcdev@free.fr>

The selected profile wasn't checked against card's ones.
(ex: resulting in sending MPEG2 to a VC1/h264 only card)
---
 modules/codec/avcodec/vaapi.c |   22 +++++++++++++++++++++-
 1 files changed, 21 insertions(+), 1 deletions(-)

diff --git a/modules/codec/avcodec/vaapi.c b/modules/codec/avcodec/vaapi.c
index 710fddd..2705025 100644
--- a/modules/codec/avcodec/vaapi.c
+++ b/modules/codec/avcodec/vaapi.c
@@ -95,8 +95,10 @@ static vlc_va_vaapi_t *vlc_va_vaapi_Get( void *p_va )
 /* */
 static int Open( vlc_va_vaapi_t *p_va, int i_codec_id )
 {
-    VAProfile i_profile;
+    VAProfile i_profile, *p_profiles_list;
     int i_surface_count;
+    bool b_supported_profile = false;
+    int i_profiles_nb = 0;
 
     /* */
     switch( i_codec_id )
@@ -144,6 +146,24 @@ static int Open( vlc_va_vaapi_t *p_va, int i_codec_id )
     if( vaInitialize( p_va->p_display, &p_va->i_version_major, &p_va->i_version_minor ) )
         goto error;
 
+    /* Check if the selected profile is supported */
+    i_profiles_nb = vaMaxNumEntrypoints( p_va->p_display );
+    p_profiles_list = malloc( i_profiles_nb * sizeof( VAProfile ) );
+    if ( p_profiles_list &&
+         vaQueryConfigProfiles( p_va->p_display, p_profiles_list, &i_profiles_nb ) == VA_STATUS_SUCCESS )
+    {
+        while( --i_profiles_nb >= 0 )
+        {
+            if ( p_profiles_list[i_profiles_nb] == i_profile )
+            {
+                b_supported_profile = true;
+                break;
+            }
+        }
+    }
+    free( p_profiles_list );
+    if ( !b_supported_profile ) goto error;
+
     /* Create a VA configuration */
     VAConfigAttrib attrib;
     memset( &attrib, 0, sizeof(attrib) );
-- 
1.7.3.4



From cyzicusmedia at gmail.com  Tue Dec  6 14:47:29 2011
From: cyzicusmedia at gmail.com (Derek Scott)
Date: Tue, 06 Dec 2011 20:47:29 +0700
Subject: [vlc-devel]  Compiling VLC in MacOS under Lion 10.7.2
Message-ID: <4EDE1CF1.5030300@gmail.com>


Hello everyone,

I've been trying to compile VLC from the git repository using the 
instructions at http://wiki.videolan.org/OSXCompile.

1.  I successfully get the development tools by bootstrapping in 
vlc/extras/tools.

2.  I select 10.6 SDK by $ export OSX_VERSION=10.6.

3.  Prepare 3rd Party Libraries.  After creating the osx directory in 
contrib, I do

  $ ../bootstrap --host=x86_64-apple-darwin11 --build=x86_64-apple-darwin11

Since I'm running in 64-bit in OS 10.7.2, I thought my host should be 
"--host=x86_64-apple-darwin11" and my build should be 
"--build=x86_64-apple-darwin11."

4.  Prebuilt libraries.

I try make prebuilt, but

ftp://ftp.videolan.org/pub/videolan/contrib/x86_64-apple-darwin11/vlc-contrib-x86_64-apple-darwin11-latest.tar.bz2"

fails because there's no file there.

5.  Build your own libraries.

I try

make -j4 .gettext

This errors out with the following:

In file included from ./bindtextdom.c:24:
/Developer/SDKs/MacOSX10.6.sdk/usr/include/stddef.h:74: error: two or 
more data types in declaration specifiers
In file included from ./gettextP.h:23,
                  from ./dgettext.c:23:
/Developer/SDKs/MacOSX10.6.sdk/usr/include/stddef.h:74: error: two or 
more data types in declaration specifiers
In file included from ./gettextP.h:23,
                  from ./dcgettext.c:23:
/Developer/SDKs/MacOSX10.6.sdk/usr/include/stddef.h:74: error: two or 
more data types in declaration specifiers
In file included from ./gettextP.h:23,
                  from ./gettext.c:30:
/Developer/SDKs/MacOSX10.6.sdk/usr/include/stddef.h:74: error: two or 
more data types in declaration specifiers
make[4]: *** [dgettext.lo] Error 1
make[4]: *** Waiting for unfinished jobs....
make[4]: *** [gettext.lo] Error 1
make[4]: *** [dcgettext.lo] Error 1
make[4]: *** [bindtextdom.lo] Error 1
make[3]: *** [all-recursive] Error 1
make[2]: *** [all] Error 2
make[1]: *** [all-recursive] Error 1
make: *** [.gettext] Error 2

--I'd appreciate your suggestions.

--If this is the wrong place to post this, please let me know the 
correct location.  Thanks.

Best regards,

Derek Scott






From komh78 at gmail.com  Tue Dec  6 14:55:19 2011
From: komh78 at gmail.com (KO Myung-Hun)
Date: Tue, 06 Dec 2011 22:55:19 +0900
Subject: [vlc-devel] [PATCH 6/9] Add missing getaddrinfo(),
 freeaddrinfo(), getnameinfo() and gai_strerror() for OS/2
In-Reply-To: <201112011926.33363.remi@remlab.net>
References: <1322287529-7242-1-git-send-email-komh@chollian.net>	<201111291655.40201.remi@remlab.net>	<4ED77CD2.5090707@chollian.net>
	<201112011926.33363.remi@remlab.net>
Message-ID: <4EDE1EC7.4000403@chollian.net>



R?mi Denis-Courmont wrote:
> Le jeudi 1 d?cembre 2011 15:10:42 KO Myung-Hun, vous avez ?crit :
>> R?mi Denis-Courmont wrote:
>>> I'm neither allowed nor willing to merge GPL stuff into LibVLC as of now.
>>
>> How about this ?
> 
> It looks OK but it does not apply on my tree.
> 

Ok. I removed the dependency on the other patch.

-- 
KO Myung-Hun

Using Mozilla SeaMonkey 2.0.14
Under OS/2 Warp 4 for Korean with FixPak #15
On AMD ThunderBird 1GHz with 512 MB RAM

Korean OS/2 User Community : http://www.ecomstation.co.kr

-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: 0006-Add-missing-getaddrinfo-freeaddrinfo-getnameinfo-and.patch
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111206/2aed42bb/attachment.ksh>

From komh78 at gmail.com  Tue Dec  6 14:58:27 2011
From: komh78 at gmail.com (KO Myung-Hun)
Date: Tue, 06 Dec 2011 22:58:27 +0900
Subject: [vlc-devel] [PATCH 5/9] Define IPv6 stuffs required in order to
 compile on OS/2.
In-Reply-To: <1322287529-7242-6-git-send-email-komh@chollian.net>
References: <1322287529-7242-1-git-send-email-komh@chollian.net>
	<1322287529-7242-6-git-send-email-komh@chollian.net>
Message-ID: <4EDE1F83.1080509@chollian.net>



KO Myung-Hun wrote:
> [...]

Updated to remove the dependency of the following patch on this patch.

    [PATCH 6/9] Add missing getaddrinfo(), freeaddrinfo(), getnameinfo()
and gai_strerror() for OS/2

-- 
KO Myung-Hun

Using Mozilla SeaMonkey 2.0.14
Under OS/2 Warp 4 for Korean with FixPak #15
On AMD ThunderBird 1GHz with 512 MB RAM

Korean OS/2 User Community : http://www.ecomstation.co.kr

-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: 0005-Define-IPv6-stuffs-required-in-order-to-compile-on-O.patch
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111206/6c3d0814/attachment.ksh>

From remi at remlab.net  Tue Dec  6 14:59:04 2011
From: remi at remlab.net (=?utf-8?q?R=C3=A9mi?= Denis-Courmont)
Date: Tue, 6 Dec 2011 15:59:04 +0200
Subject: [vlc-devel] [Patch] Adding an helper to convert iso8601
	durations to seconds
In-Reply-To: <CAK-C75SA9p6Sfv4YoGe6=V8d8WG00goDi9ACOmJzzkJDFM=PVw@mail.gmail.com>
References: <CAK-C75R5BPUud6z2QxrFNNnh1wMFktu48O59oXi2bAUqZ0mYMw@mail.gmail.com>
	<CAK-C75TPsc1Lh07NNXFnmzW5ZQgSQbkfm1S2Vfb=e2Nc11Z_Tw@mail.gmail.com>
	<CAK-C75SA9p6Sfv4YoGe6=V8d8WG00goDi9ACOmJzzkJDFM=PVw@mail.gmail.com>
Message-ID: <201112061559.04629.remi@remlab.net>

Le mardi 6 d?cembre 2011 14:11:10 Hugo Beauz?e-Luyssen, vous avez ?crit :
> On Wed, Nov 30, 2011 at 4:41 PM, Hugo Beauz?e-Luyssen
> 
> <beauze.h at gmail.com> wrote:
> > On Wed, Nov 30, 2011 at 2:45 PM, Hugo Beauz?e-Luyssen
> > 
> > <beauze.h at gmail.com> wrote:
> >> On Wed, Nov 30, 2011 at 2:08 PM, Tobias G?ntner <fatbull at web.de> wrote:
> >>> Am 30.11.2011 11:05, schrieb Hugo Beauz?e-Luyssen:
> >>>> +/*
> >>>> +  Decodes a duration as defined by ISO 8601
> >>>> +  http://en.wikipedia.org/wiki/ISO_8601#Durations
> >>>> +  @param str A null-terminated string to convert
> >>> 
> >>> Perhaps a short example of a valid input string would be useful, so
> >>> readers can see at first glance what kind of data this function
> >>> expects, without looking it up.
> >> 
> >> Yep I'm going to write some tests in a this afternoon or tomorrow, but
> >> I could add a link to the format description and an exemple in the
> >> comments.
> >> 
> >>>> +  @return: The duration in seconds. 0 if an error occured.
> >>> 
> >>> This function usually does not return 0 if the input is invalid.
> >> 
> >> Yep i'm going to change the return type as a signed int 64, so -1 is an
> >> error.
> >> 
> >>>> + */
> >>>> +uint64_t str_iso8601_duration_to_seconds( const char *psz_duration )
> >>>> +{
> >>>> +    bool        timeDesignatorReached = false;
> >>>> +    double      mul = 0;
> >>> 
> >>> I think an integer should be fine.
> >> 
> >> Well in the exemple file I've got
> >> (http://www-itec.uni-klu.ac.at/ftp/datasets/mmsys12/Selected_VLC_Plugin_
> >> MPDs/BigBuckBunny_15s_DASH_VLC_Compatible_4-Representations_720p.mpd)
> >> there are some decimal values, so I try to reduce the loss of
> >> precision when converting by using doubles.
> >> 
> >>>> +    mtime_t     res = 0;
> >>> 
> >>> Shouldn't this type and the return type be the same?
> >> 
> >> I was using mtime_t at first, it seems I forgot to change this one, my
> >> bad.
> >> 
> >> Since the line below is a direct pointer dereferencement without any
> >> prior check, this was intended, but a check could be added.
> >> 
> >>>> +
> >>>> +    if ( *psz_duration++ != 'P' )
> >>>> +        return 0;
> >>>> +    for ( const char* nptr = psz_duration; *psz_duration;
> >>>> ++psz_duration ) +    {
> >>>> +        switch( *psz_duration )
> >>>> +        {
> >>>> +            case 'M':
> >>>> +            {
> >>> 
> >>> I find this a little confusing because letters normally follow numbers
> >>> in the input, yet you somehow parse letters first. ;)
> >> 
> >> That is absolutely right...
> >> 
> >>>> +                //M can mean month or minutes, if the 'T' flag has
> >>>> been reached. +                //We don't handle months though.
> >>>> +                if ( timeDesignatorReached == true )
> >>>> +                    mul = 60.0;
> >>>> +                break ;
> >>>> +            }
> >>>> +            case 'Y':
> >>>> +            case 'W':
> >>>> +                break ; //Don't handle this duration.
> >>> 
> >>> Maybe return an error here?
> >> 
> >> I don't really know... warning should be mandatory, an error may be to
> >> agressive...
> >> 
> >>>> +            case 'D':
> >>>> +                mul = 86400.0;
> >>>> +                break ;
> >>>> +            case 'T':
> >>>> +                timeDesignatorReached = true;
> >>>> +                break ;
> >>>> +            case 'H':
> >>>> +                mul = 3600.0;
> >>>> +                break ;
> >>>> +            case 'S':
> >>>> +                mul = 1.0;
> >>>> +                break ;
> >>>> +            default:
> >>>> +                continue ;
> >>>> +        }
> >>>> +        //It means we went trough some digits. So we must compute a
> >>>> number +        if ( nptr != psz_duration )
> >>>> +        {
> >>>> +            double tmp = strtold( nptr, NULL );
> >>> 
> >>> I think you can achieve a slightly cleaner design if you utilize the
> >>> second parameter of that function. The strtoX functions store the
> >>> address of the first invalid character in there, so all you have to do
> >> 
> >>> is basically:
> >> Yep since I'm doing it in the wrong way, using the second parameter
> >> makes a lot of sense now :D
> >> 
> >>> 1. Call strtoul(psz_duration, &endptr, 10)
> >>> 2. Check the character at *endptr
> >>> 3. Repeat with psz_duration = endptr + 1
> >>> 
> >>>> +            res += (uint64_t)(tmp * mul);
> >>>> +            nptr = psz_duration;
> >>>> +        }
> >>>> +        nptr++;
> >>>> +    }
> >>>> +    return res;
> >>>> +}
> >>> 
> >>> Regards,
> >>> Tobias
> >> 
> >> Thanks for the review!
> >> Regards,
> > 
> > Here is a new version, along with some unit tests.
> > 
> > Regards,
> 
> Hi,
> 
> Could somebody review these patches?


+    if ( *psz_duration++ != 'P' )
+        return -1;

With *three* independent operators (dereference, increment and inequality) on 
psz_duration in a single expression, you really need more explicit parenthesis 
IMHO.

Also, the test case name is very poor IMHO.

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From remi at remlab.net  Tue Dec  6 15:02:02 2011
From: remi at remlab.net (=?iso-8859-15?q?R=E9mi?= Denis-Courmont)
Date: Tue, 6 Dec 2011 16:02:02 +0200
Subject: [vlc-devel] [PATCH 6/9] Add missing getaddrinfo(),
	freeaddrinfo(), getnameinfo() and gai_strerror() for OS/2
In-Reply-To: <4EDE1EC7.4000403@chollian.net>
References: <1322287529-7242-1-git-send-email-komh@chollian.net>
	<201112011926.33363.remi@remlab.net>
	<4EDE1EC7.4000403@chollian.net>
Message-ID: <201112061602.03652.remi@remlab.net>

Le mardi 6 d?cembre 2011 15:55:19 KO Myung-Hun, vous avez ?crit :
> R?mi Denis-Courmont wrote:
> > Le jeudi 1 d?cembre 2011 15:10:42 KO Myung-Hun, vous avez ?crit :
> >> R?mi Denis-Courmont wrote:
> >>> I'm neither allowed nor willing to merge GPL stuff into LibVLC as of
> >>> now.
> >> 
> >> How about this ?
> > 
> > It looks OK but it does not apply on my tree.
> 
> Ok. I removed the dependency on the other patch.

Merged, thanks.

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From beauze.h at gmail.com  Tue Dec  6 15:02:15 2011
From: beauze.h at gmail.com (=?UTF-8?Q?Hugo_Beauz=C3=A9e=2DLuyssen?=)
Date: Tue, 6 Dec 2011 15:02:15 +0100
Subject: [vlc-devel] [Patch] Adding an helper to convert iso8601
 durations to seconds
In-Reply-To: <201112061559.04629.remi@remlab.net>
References: <CAK-C75R5BPUud6z2QxrFNNnh1wMFktu48O59oXi2bAUqZ0mYMw@mail.gmail.com>
	<CAK-C75TPsc1Lh07NNXFnmzW5ZQgSQbkfm1S2Vfb=e2Nc11Z_Tw@mail.gmail.com>
	<CAK-C75SA9p6Sfv4YoGe6=V8d8WG00goDi9ACOmJzzkJDFM=PVw@mail.gmail.com>
	<201112061559.04629.remi@remlab.net>
Message-ID: <CAK-C75RQNA-+iYR8wKs2LfFAE54oDOswZ_e-PXcpBJwe8W2zqg@mail.gmail.com>

2011/12/6 R?mi Denis-Courmont <remi at remlab.net>:
> Le mardi 6 d?cembre 2011 14:11:10 Hugo Beauz?e-Luyssen, vous avez ?crit :
>> On Wed, Nov 30, 2011 at 4:41 PM, Hugo Beauz?e-Luyssen
>>
>> <beauze.h at gmail.com> wrote:
>> > On Wed, Nov 30, 2011 at 2:45 PM, Hugo Beauz?e-Luyssen
>> >
>> > <beauze.h at gmail.com> wrote:
>> >> On Wed, Nov 30, 2011 at 2:08 PM, Tobias G?ntner <fatbull at web.de> wrote:
>> >>> Am 30.11.2011 11:05, schrieb Hugo Beauz?e-Luyssen:
>> >>>> +/*
>> >>>> + ?Decodes a duration as defined by ISO 8601
>> >>>> + ?http://en.wikipedia.org/wiki/ISO_8601#Durations
>> >>>> + ?@param str A null-terminated string to convert
>> >>>
>> >>> Perhaps a short example of a valid input string would be useful, so
>> >>> readers can see at first glance what kind of data this function
>> >>> expects, without looking it up.
>> >>
>> >> Yep I'm going to write some tests in a this afternoon or tomorrow, but
>> >> I could add a link to the format description and an exemple in the
>> >> comments.
>> >>
>> >>>> + ?@return: The duration in seconds. 0 if an error occured.
>> >>>
>> >>> This function usually does not return 0 if the input is invalid.
>> >>
>> >> Yep i'm going to change the return type as a signed int 64, so -1 is an
>> >> error.
>> >>
>> >>>> + */
>> >>>> +uint64_t str_iso8601_duration_to_seconds( const char *psz_duration )
>> >>>> +{
>> >>>> + ? ?bool ? ? ? ?timeDesignatorReached = false;
>> >>>> + ? ?double ? ? ?mul = 0;
>> >>>
>> >>> I think an integer should be fine.
>> >>
>> >> Well in the exemple file I've got
>> >> (http://www-itec.uni-klu.ac.at/ftp/datasets/mmsys12/Selected_VLC_Plugin_
>> >> MPDs/BigBuckBunny_15s_DASH_VLC_Compatible_4-Representations_720p.mpd)
>> >> there are some decimal values, so I try to reduce the loss of
>> >> precision when converting by using doubles.
>> >>
>> >>>> + ? ?mtime_t ? ? res = 0;
>> >>>
>> >>> Shouldn't this type and the return type be the same?
>> >>
>> >> I was using mtime_t at first, it seems I forgot to change this one, my
>> >> bad.
>> >>
>> >> Since the line below is a direct pointer dereferencement without any
>> >> prior check, this was intended, but a check could be added.
>> >>
>> >>>> +
>> >>>> + ? ?if ( *psz_duration++ != 'P' )
>> >>>> + ? ? ? ?return 0;
>> >>>> + ? ?for ( const char* nptr = psz_duration; *psz_duration;
>> >>>> ++psz_duration ) + ? ?{
>> >>>> + ? ? ? ?switch( *psz_duration )
>> >>>> + ? ? ? ?{
>> >>>> + ? ? ? ? ? ?case 'M':
>> >>>> + ? ? ? ? ? ?{
>> >>>
>> >>> I find this a little confusing because letters normally follow numbers
>> >>> in the input, yet you somehow parse letters first. ;)
>> >>
>> >> That is absolutely right...
>> >>
>> >>>> + ? ? ? ? ? ? ? ?//M can mean month or minutes, if the 'T' flag has
>> >>>> been reached. + ? ? ? ? ? ? ? ?//We don't handle months though.
>> >>>> + ? ? ? ? ? ? ? ?if ( timeDesignatorReached == true )
>> >>>> + ? ? ? ? ? ? ? ? ? ?mul = 60.0;
>> >>>> + ? ? ? ? ? ? ? ?break ;
>> >>>> + ? ? ? ? ? ?}
>> >>>> + ? ? ? ? ? ?case 'Y':
>> >>>> + ? ? ? ? ? ?case 'W':
>> >>>> + ? ? ? ? ? ? ? ?break ; //Don't handle this duration.
>> >>>
>> >>> Maybe return an error here?
>> >>
>> >> I don't really know... warning should be mandatory, an error may be to
>> >> agressive...
>> >>
>> >>>> + ? ? ? ? ? ?case 'D':
>> >>>> + ? ? ? ? ? ? ? ?mul = 86400.0;
>> >>>> + ? ? ? ? ? ? ? ?break ;
>> >>>> + ? ? ? ? ? ?case 'T':
>> >>>> + ? ? ? ? ? ? ? ?timeDesignatorReached = true;
>> >>>> + ? ? ? ? ? ? ? ?break ;
>> >>>> + ? ? ? ? ? ?case 'H':
>> >>>> + ? ? ? ? ? ? ? ?mul = 3600.0;
>> >>>> + ? ? ? ? ? ? ? ?break ;
>> >>>> + ? ? ? ? ? ?case 'S':
>> >>>> + ? ? ? ? ? ? ? ?mul = 1.0;
>> >>>> + ? ? ? ? ? ? ? ?break ;
>> >>>> + ? ? ? ? ? ?default:
>> >>>> + ? ? ? ? ? ? ? ?continue ;
>> >>>> + ? ? ? ?}
>> >>>> + ? ? ? ?//It means we went trough some digits. So we must compute a
>> >>>> number + ? ? ? ?if ( nptr != psz_duration )
>> >>>> + ? ? ? ?{
>> >>>> + ? ? ? ? ? ?double tmp = strtold( nptr, NULL );
>> >>>
>> >>> I think you can achieve a slightly cleaner design if you utilize the
>> >>> second parameter of that function. The strtoX functions store the
>> >>> address of the first invalid character in there, so all you have to do
>> >>
>> >>> is basically:
>> >> Yep since I'm doing it in the wrong way, using the second parameter
>> >> makes a lot of sense now :D
>> >>
>> >>> 1. Call strtoul(psz_duration, &endptr, 10)
>> >>> 2. Check the character at *endptr
>> >>> 3. Repeat with psz_duration = endptr + 1
>> >>>
>> >>>> + ? ? ? ? ? ?res += (uint64_t)(tmp * mul);
>> >>>> + ? ? ? ? ? ?nptr = psz_duration;
>> >>>> + ? ? ? ?}
>> >>>> + ? ? ? ?nptr++;
>> >>>> + ? ?}
>> >>>> + ? ?return res;
>> >>>> +}
>> >>>
>> >>> Regards,
>> >>> Tobias
>> >>
>> >> Thanks for the review!
>> >> Regards,
>> >
>> > Here is a new version, along with some unit tests.
>> >
>> > Regards,
>>
>> Hi,
>>
>> Could somebody review these patches?
>
>
> + ? ?if ( *psz_duration++ != 'P' )
> + ? ? ? ?return -1;
>
> With *three* independent operators (dereference, increment and inequality) on
> psz_duration in a single expression, you really need more explicit parenthesis
> IMHO.
>

Okey, I thought that was ok but I will fix it.

> Also, the test case name is very poor IMHO.
>

It is. Will add some.

Regards,

-- 
Hugo Beauz?e-Luyssen


From jpsaman at videolan.org  Tue Dec  6 15:16:14 2011
From: jpsaman at videolan.org (Jean-Paul Saman)
Date: Tue, 6 Dec 2011 15:16:14 +0100
Subject: [vlc-devel] [PATCH] vaapi: fix profiles detection.
In-Reply-To: <1323177188-14117-1-git-send-email-fcvlcdev@free.fr>
References: <1323177188-14117-1-git-send-email-fcvlcdev@free.fr>
Message-ID: <CAK9US3AR9UYihxWkYbM42e1AGAMniCH3CQ+k1yEe-qiUGnsLPg@mail.gmail.com>

This patch crashes in my tests with an NVIDIA card. See review comments below:

On Tue, Dec 6, 2011 at 2:13 PM, Francois Cartegnie <fcvlcdev at free.fr> wrote:
> The selected profile wasn't checked against card's ones.
> (ex: resulting in sending MPEG2 to a VC1/h264 only card)
> ---
> ?modules/codec/avcodec/vaapi.c | ? 22 +++++++++++++++++++++-
> ?1 files changed, 21 insertions(+), 1 deletions(-)
>
> diff --git a/modules/codec/avcodec/vaapi.c b/modules/codec/avcodec/vaapi.c
> index 710fddd..2705025 100644
> --- a/modules/codec/avcodec/vaapi.c
> +++ b/modules/codec/avcodec/vaapi.c
> @@ -95,8 +95,10 @@ static vlc_va_vaapi_t *vlc_va_vaapi_Get( void *p_va )
> ?/* */
> ?static int Open( vlc_va_vaapi_t *p_va, int i_codec_id )
> ?{
> - ? ?VAProfile i_profile;
> + ? ?VAProfile i_profile, *p_profiles_list;
> ? ? int i_surface_count;
> + ? ?bool b_supported_profile = false;
> + ? ?int i_profiles_nb = 0;
>
> ? ? /* */
> ? ? switch( i_codec_id )
> @@ -144,6 +146,24 @@ static int Open( vlc_va_vaapi_t *p_va, int i_codec_id )
> ? ? if( vaInitialize( p_va->p_display, &p_va->i_version_major, &p_va->i_version_minor ) )
> ? ? ? ? goto error;
>
> + ? ?/* Check if the selected profile is supported */
> + ? ?i_profiles_nb = vaMaxNumEntrypoints( p_va->p_display );

==> cause of crash, it should be vaMaxNumProfiles( p_va->p_display );

> + ? ?p_profiles_list = malloc( i_profiles_nb * sizeof( VAProfile ) );

Use calloc() here: VAProfile *p_profiles_list = calloc( i_profiles_nb,
sizeof( VAProfile ) );

> + ? ?if ( p_profiles_list &&
> + ? ? ? ? vaQueryConfigProfiles( p_va->p_display, p_profiles_list, &i_profiles_nb ) == VA_STATUS_SUCCESS )

I would split this up, like this:

if (!p_profiles_list)
   goto error;

VAStatus status = vaQueryConfigProfiles( p_va->p_display,
p_profiles_list, &i_profiles_nb );
if ( status == VA_STATUS_SUCCESS )
{
    ...
}

> + ? ?{
> + ? ? ? ?while( --i_profiles_nb >= 0 )

The coding style in this file uses for-loops in this kind of
constructs. Therefor my suggestion is that you do the same.

> + ? ? ? ?{
> + ? ? ? ? ? ?if ( p_profiles_list[i_profiles_nb] == i_profile )
> + ? ? ? ? ? ?{
> + ? ? ? ? ? ? ? ?b_supported_profile = true;
> + ? ? ? ? ? ? ? ?break;
> + ? ? ? ? ? ?}
> + ? ? ? ?}
> + ? ?}
> + ? ?free( p_profiles_list );
> + ? ?if ( !b_supported_profile ) goto error;
> +
> ? ? /* Create a VA configuration */
> ? ? VAConfigAttrib attrib;
> ? ? memset( &attrib, 0, sizeof(attrib) );
> --
> 1.7.3.4
>
> _______________________________________________
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
> http://mailman.videolan.org/listinfo/vlc-devel

Kind regards,

Jean-Paul Saman


From remi at remlab.net  Tue Dec  6 15:20:40 2011
From: remi at remlab.net (=?iso-8859-1?q?R=E9mi?= Denis-Courmont)
Date: Tue, 6 Dec 2011 16:20:40 +0200
Subject: [vlc-devel] [PATCH] vaapi: fix profiles detection.
In-Reply-To: <CAK9US3AR9UYihxWkYbM42e1AGAMniCH3CQ+k1yEe-qiUGnsLPg@mail.gmail.com>
References: <1323177188-14117-1-git-send-email-fcvlcdev@free.fr>
	<CAK9US3AR9UYihxWkYbM42e1AGAMniCH3CQ+k1yEe-qiUGnsLPg@mail.gmail.com>
Message-ID: <201112061620.41148.remi@remlab.net>

Le mardi 6 d?cembre 2011 16:16:14 Jean-Paul Saman, vous avez ?crit :
> > +    p_profiles_list = malloc( i_profiles_nb * sizeof( VAProfile ) );
> 
> Use calloc() here: VAProfile *p_profiles_list = calloc( i_profiles_nb,
> sizeof( VAProfile ) );

Why bother the zero the memory here?


-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From jpsaman at videolan.org  Tue Dec  6 15:29:12 2011
From: jpsaman at videolan.org (Jean-Paul Saman)
Date: Tue, 6 Dec 2011 15:29:12 +0100
Subject: [vlc-devel] [PATCH] vaapi: fix profiles detection.
In-Reply-To: <201112061620.41148.remi@remlab.net>
References: <1323177188-14117-1-git-send-email-fcvlcdev@free.fr>
	<CAK9US3AR9UYihxWkYbM42e1AGAMniCH3CQ+k1yEe-qiUGnsLPg@mail.gmail.com>
	<201112061620.41148.remi@remlab.net>
Message-ID: <CAK9US3AVps+1_2LmSrdzs7kZ_UGFYHAivKJjDz+wnkZZu6trNQ@mail.gmail.com>

2011/12/6 R?mi Denis-Courmont <remi at remlab.net>:
> Le mardi 6 d?cembre 2011 16:16:14 Jean-Paul Saman, vous avez ?crit :
>> > + ? ?p_profiles_list = malloc( i_profiles_nb * sizeof( VAProfile ) );
>>
>> Use calloc() here: VAProfile *p_profiles_list = calloc( i_profiles_nb,
>> sizeof( VAProfile ) );
>
> Why bother the zero the memory here?

Cause vaMaxNumProfiles() is not guarenteed to fill it up. Although in
this patch it might not be necessary.

Kind regards

Jean-Paul Saman.


From beauze.h at gmail.com  Tue Dec  6 15:41:49 2011
From: beauze.h at gmail.com (=?UTF-8?Q?Hugo_Beauz=C3=A9e=2DLuyssen?=)
Date: Tue, 6 Dec 2011 15:41:49 +0100
Subject: [vlc-devel] [Patch] Adding an helper to convert iso8601
 durations to seconds
In-Reply-To: <CAK-C75RQNA-+iYR8wKs2LfFAE54oDOswZ_e-PXcpBJwe8W2zqg@mail.gmail.com>
References: <CAK-C75R5BPUud6z2QxrFNNnh1wMFktu48O59oXi2bAUqZ0mYMw@mail.gmail.com>
	<CAK-C75TPsc1Lh07NNXFnmzW5ZQgSQbkfm1S2Vfb=e2Nc11Z_Tw@mail.gmail.com>
	<CAK-C75SA9p6Sfv4YoGe6=V8d8WG00goDi9ACOmJzzkJDFM=PVw@mail.gmail.com>
	<201112061559.04629.remi@remlab.net>
	<CAK-C75RQNA-+iYR8wKs2LfFAE54oDOswZ_e-PXcpBJwe8W2zqg@mail.gmail.com>
Message-ID: <CAK-C75T4x2x4cBjFwvz-FtSseLsV1EoHe=RvfsqUAqb70mf7Aw@mail.gmail.com>

On Tue, Dec 6, 2011 at 3:02 PM, Hugo Beauz?e-Luyssen <beauze.h at gmail.com> wrote:
> 2011/12/6 R?mi Denis-Courmont <remi at remlab.net>:
>> Le mardi 6 d?cembre 2011 14:11:10 Hugo Beauz?e-Luyssen, vous avez ?crit :
>>> On Wed, Nov 30, 2011 at 4:41 PM, Hugo Beauz?e-Luyssen
>>>
>>> <beauze.h at gmail.com> wrote:
>>> > On Wed, Nov 30, 2011 at 2:45 PM, Hugo Beauz?e-Luyssen
>>> >
>>> > <beauze.h at gmail.com> wrote:
>>> >> On Wed, Nov 30, 2011 at 2:08 PM, Tobias G?ntner <fatbull at web.de> wrote:
>>> >>> Am 30.11.2011 11:05, schrieb Hugo Beauz?e-Luyssen:
>>> >>>> +/*
>>> >>>> + ?Decodes a duration as defined by ISO 8601
>>> >>>> + ?http://en.wikipedia.org/wiki/ISO_8601#Durations
>>> >>>> + ?@param str A null-terminated string to convert
>>> >>>
>>> >>> Perhaps a short example of a valid input string would be useful, so
>>> >>> readers can see at first glance what kind of data this function
>>> >>> expects, without looking it up.
>>> >>
>>> >> Yep I'm going to write some tests in a this afternoon or tomorrow, but
>>> >> I could add a link to the format description and an exemple in the
>>> >> comments.
>>> >>
>>> >>>> + ?@return: The duration in seconds. 0 if an error occured.
>>> >>>
>>> >>> This function usually does not return 0 if the input is invalid.
>>> >>
>>> >> Yep i'm going to change the return type as a signed int 64, so -1 is an
>>> >> error.
>>> >>
>>> >>>> + */
>>> >>>> +uint64_t str_iso8601_duration_to_seconds( const char *psz_duration )
>>> >>>> +{
>>> >>>> + ? ?bool ? ? ? ?timeDesignatorReached = false;
>>> >>>> + ? ?double ? ? ?mul = 0;
>>> >>>
>>> >>> I think an integer should be fine.
>>> >>
>>> >> Well in the exemple file I've got
>>> >> (http://www-itec.uni-klu.ac.at/ftp/datasets/mmsys12/Selected_VLC_Plugin_
>>> >> MPDs/BigBuckBunny_15s_DASH_VLC_Compatible_4-Representations_720p.mpd)
>>> >> there are some decimal values, so I try to reduce the loss of
>>> >> precision when converting by using doubles.
>>> >>
>>> >>>> + ? ?mtime_t ? ? res = 0;
>>> >>>
>>> >>> Shouldn't this type and the return type be the same?
>>> >>
>>> >> I was using mtime_t at first, it seems I forgot to change this one, my
>>> >> bad.
>>> >>
>>> >> Since the line below is a direct pointer dereferencement without any
>>> >> prior check, this was intended, but a check could be added.
>>> >>
>>> >>>> +
>>> >>>> + ? ?if ( *psz_duration++ != 'P' )
>>> >>>> + ? ? ? ?return 0;
>>> >>>> + ? ?for ( const char* nptr = psz_duration; *psz_duration;
>>> >>>> ++psz_duration ) + ? ?{
>>> >>>> + ? ? ? ?switch( *psz_duration )
>>> >>>> + ? ? ? ?{
>>> >>>> + ? ? ? ? ? ?case 'M':
>>> >>>> + ? ? ? ? ? ?{
>>> >>>
>>> >>> I find this a little confusing because letters normally follow numbers
>>> >>> in the input, yet you somehow parse letters first. ;)
>>> >>
>>> >> That is absolutely right...
>>> >>
>>> >>>> + ? ? ? ? ? ? ? ?//M can mean month or minutes, if the 'T' flag has
>>> >>>> been reached. + ? ? ? ? ? ? ? ?//We don't handle months though.
>>> >>>> + ? ? ? ? ? ? ? ?if ( timeDesignatorReached == true )
>>> >>>> + ? ? ? ? ? ? ? ? ? ?mul = 60.0;
>>> >>>> + ? ? ? ? ? ? ? ?break ;
>>> >>>> + ? ? ? ? ? ?}
>>> >>>> + ? ? ? ? ? ?case 'Y':
>>> >>>> + ? ? ? ? ? ?case 'W':
>>> >>>> + ? ? ? ? ? ? ? ?break ; //Don't handle this duration.
>>> >>>
>>> >>> Maybe return an error here?
>>> >>
>>> >> I don't really know... warning should be mandatory, an error may be to
>>> >> agressive...
>>> >>
>>> >>>> + ? ? ? ? ? ?case 'D':
>>> >>>> + ? ? ? ? ? ? ? ?mul = 86400.0;
>>> >>>> + ? ? ? ? ? ? ? ?break ;
>>> >>>> + ? ? ? ? ? ?case 'T':
>>> >>>> + ? ? ? ? ? ? ? ?timeDesignatorReached = true;
>>> >>>> + ? ? ? ? ? ? ? ?break ;
>>> >>>> + ? ? ? ? ? ?case 'H':
>>> >>>> + ? ? ? ? ? ? ? ?mul = 3600.0;
>>> >>>> + ? ? ? ? ? ? ? ?break ;
>>> >>>> + ? ? ? ? ? ?case 'S':
>>> >>>> + ? ? ? ? ? ? ? ?mul = 1.0;
>>> >>>> + ? ? ? ? ? ? ? ?break ;
>>> >>>> + ? ? ? ? ? ?default:
>>> >>>> + ? ? ? ? ? ? ? ?continue ;
>>> >>>> + ? ? ? ?}
>>> >>>> + ? ? ? ?//It means we went trough some digits. So we must compute a
>>> >>>> number + ? ? ? ?if ( nptr != psz_duration )
>>> >>>> + ? ? ? ?{
>>> >>>> + ? ? ? ? ? ?double tmp = strtold( nptr, NULL );
>>> >>>
>>> >>> I think you can achieve a slightly cleaner design if you utilize the
>>> >>> second parameter of that function. The strtoX functions store the
>>> >>> address of the first invalid character in there, so all you have to do
>>> >>
>>> >>> is basically:
>>> >> Yep since I'm doing it in the wrong way, using the second parameter
>>> >> makes a lot of sense now :D
>>> >>
>>> >>> 1. Call strtoul(psz_duration, &endptr, 10)
>>> >>> 2. Check the character at *endptr
>>> >>> 3. Repeat with psz_duration = endptr + 1
>>> >>>
>>> >>>> + ? ? ? ? ? ?res += (uint64_t)(tmp * mul);
>>> >>>> + ? ? ? ? ? ?nptr = psz_duration;
>>> >>>> + ? ? ? ?}
>>> >>>> + ? ? ? ?nptr++;
>>> >>>> + ? ?}
>>> >>>> + ? ?return res;
>>> >>>> +}
>>> >>>
>>> >>> Regards,
>>> >>> Tobias
>>> >>
>>> >> Thanks for the review!
>>> >> Regards,
>>> >
>>> > Here is a new version, along with some unit tests.
>>> >
>>> > Regards,
>>>
>>> Hi,
>>>
>>> Could somebody review these patches?
>>
>>
>> + ? ?if ( *psz_duration++ != 'P' )
>> + ? ? ? ?return -1;
>>
>> With *three* independent operators (dereference, increment and inequality) on
>> psz_duration in a single expression, you really need more explicit parenthesis
>> IMHO.
>>
>
> Okey, I thought that was ok but I will fix it.
>

Should be better, but to be honest I find it harder to read.

>> Also, the test case name is very poor IMHO.
>>
>
> It is. Will add some.
>

I've added three tests, though I think that's enough to test the basic
behaviour.

Regards,

-- 
Hugo Beauz?e-Luyssen
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0007-strings-Adding-an-helper-to-convert-iso8601-duration.patch
Type: text/x-patch
Size: 3112 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111206/44f08caf/attachment.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0008-tests-Running-new-strings-tests-in-the-test-suite.patch
Type: text/x-patch
Size: 3399 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111206/44f08caf/attachment-0001.bin>

From fcvlcdev at free.fr  Tue Dec  6 15:43:01 2011
From: fcvlcdev at free.fr (Francois Cartegnie)
Date: Tue, 06 Dec 2011 15:43:01 +0100
Subject: [vlc-devel] [PATCH] vaapi: fix profiles detection.
In-Reply-To: <CAK9US3AR9UYihxWkYbM42e1AGAMniCH3CQ+k1yEe-qiUGnsLPg@mail.gmail.com>
References: <1323177188-14117-1-git-send-email-fcvlcdev@free.fr>
	<CAK9US3AR9UYihxWkYbM42e1AGAMniCH3CQ+k1yEe-qiUGnsLPg@mail.gmail.com>
Message-ID: <4EDE29F5.40705@free.fr>

On 06/12/2011 15:16, Jean-Paul Saman wrote:
> ==> cause of crash, it should be vaMaxNumProfiles( p_va->p_display );

Of course that's wrong.
Scrolled through source and ended copying 3 lines lower :/

>> +        while( --i_profiles_nb >= 0 )
> 
> The coding style in this file uses for-loops in this kind of
> constructs. Therefor my suggestion is that you do the same.

I tried to avoid another top declaration as I can't see any inline var
decl here.

Francois


From jpsaman at videolan.org  Tue Dec  6 15:50:26 2011
From: jpsaman at videolan.org (Jean-Paul Saman)
Date: Tue, 6 Dec 2011 15:50:26 +0100
Subject: [vlc-devel] [PATCH] vaapi: fix profiles detection.
In-Reply-To: <4EDE29F5.40705@free.fr>
References: <1323177188-14117-1-git-send-email-fcvlcdev@free.fr>
	<CAK9US3AR9UYihxWkYbM42e1AGAMniCH3CQ+k1yEe-qiUGnsLPg@mail.gmail.com>
	<4EDE29F5.40705@free.fr>
Message-ID: <CAK9US3DqduBDo68tXesi+JGCLHYQeWnDvh49KF1L7NnE6Sdcvg@mail.gmail.com>

On Tue, Dec 6, 2011 at 3:43 PM, Francois Cartegnie <fcvlcdev at free.fr> wrote:
> On 06/12/2011 15:16, Jean-Paul Saman wrote:
>> ==> cause of crash, it should be vaMaxNumProfiles( p_va->p_display );
>
> Of course that's wrong.
> Scrolled through source and ended copying 3 lines lower :/
>
>>> + ? ? ? ?while( --i_profiles_nb >= 0 )
>>
>> The coding style in this file uses for-loops in this kind of
>> constructs. Therefor my suggestion is that you do the same.
>
> I tried to avoid another top declaration as I can't see any inline var
> decl here.

for( int i = 0 ; i < i_profiles_nb; i++ )
is allowed in VLC coding style.

Kind regards,
Jean-Paul Saman


From nkoriyama at gmail.com  Tue Dec  6 15:50:48 2011
From: nkoriyama at gmail.com (Naohiro KORIYAMA)
Date: Tue, 6 Dec 2011 23:50:48 +0900
Subject: [vlc-devel] [PATCH] yadif : Add SSSE3 and SSE2 support. porting
 from FFmpeg.
In-Reply-To: <20111205214242.GA27741@videolan.org>
References: <CAOa_K+vdRUzBj4cE9qoLyU19+NYnVuPx03UqdOEXgkT6fw60Gg@mail.gmail.com>
	<201111260919.16263.remi@remlab.net>
	<CAOa_K+uZt76_CgJ9DZHVa5-a0CD-wNJFy=scd1Q4xVr8DCx_xQ@mail.gmail.com>
	<20111126024120.2709a235@tok> <20111126112504.GA32214@videolan.org>
	<CAOa_K+svS3RvmmdpwuvgAvYJ=svjTf2CQqe93yDSO7TEwR23Mg@mail.gmail.com>
	<CAOa_K+vkZK2uO4jrbqUXDkdW4B75v+Dzgg4Joc0089hj4Kf3tg@mail.gmail.com>
	<20111205214242.GA27741@videolan.org>
Message-ID: <CAOa_K+vcob=0Af2vCLHH7jeQeBXhO5TnQtieo_LWNmC9_msgSQ@mail.gmail.com>

2011/12/6 Jean-Baptiste Kempf <jb at videolan.org>:
> http://pastebin.com/4uMWVbRV

I noticed the issue recently when I stated building my own win32 builds
for chasing the problem around UPnP.
Both OSX's clang and GCC don't complain  mm and xmm are in the clobber list.

GCC never allows the registers of unsupported instructions in the clobber list.
GCC with -mmmx (or -march=pentium-mmx or above), "unknown register mm"
errors disappear.
GCC with -msse (or -march=pentium3 or above), "unknown register xmm"
errors disappear.
GCC with -mtune doesn't help.

Here is the short test log on Mac OS X, native GCC vs cross mingw32 GCC.
http://pastebin.com/ZWA2JpRv


I found a discussion about the clobber.
http://ffmpeg.org/pipermail/ffmpeg-devel/2010-September/102164.html
http://ffmpeg.org/pipermail/ffmpeg-devel/2010-October/098782.html
As a result of the discussion,
- Check whether xmm clobbers are supported in configure like:
int main()
{
   asm volatile( "" ::: "xmm0");
}
- Define a macro that put xmm registers to the clobber list only if
xmm clobbers are supported.


But for this case, I don't need to think on a large scale.
I attached the new patch.
On this patch, checking GCC's predefined macro __MMX__ and __SSE__
to detect whether mm/xmm clobbers are supported.
These macros are set when GCC can use MMX/SSE instructions.

If it's better to check it properly with C program  like the above in
configure, I can easily rewrite.

-- 
KORIYAMA, Naohiro
nkoriyama at gmail.com
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-yadif-Add-SSSE3-and-SSE2-support.-porting-from-FFmpe.patch
Type: application/octet-stream
Size: 30788 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111206/02b5b345/attachment.obj>

From fcvlcdev at free.fr  Tue Dec  6 15:57:35 2011
From: fcvlcdev at free.fr (Francois Cartegnie)
Date: Tue,  6 Dec 2011 15:57:35 +0100
Subject: [vlc-devel] [PATCH] vaapi: fix profiles detection.
In-Reply-To: <CAK9US3AR9UYihxWkYbM42e1AGAMniCH3CQ+k1yEe-qiUGnsLPg@mail.gmail.com>
References: <CAK9US3AR9UYihxWkYbM42e1AGAMniCH3CQ+k1yEe-qiUGnsLPg@mail.gmail.com>
Message-ID: <1323183455-2238-1-git-send-email-fcvlcdev@free.fr>

The selected profile wasn't checked against card's ones.
(ex: resulting in sending MPEG2 to a VC1/h264 only card)
---
 modules/codec/avcodec/vaapi.c |   24 ++++++++++++++++++++++++
 1 files changed, 24 insertions(+), 0 deletions(-)

diff --git a/modules/codec/avcodec/vaapi.c b/modules/codec/avcodec/vaapi.c
index 710fddd..6ffeb99 100644
--- a/modules/codec/avcodec/vaapi.c
+++ b/modules/codec/avcodec/vaapi.c
@@ -96,7 +96,11 @@ static vlc_va_vaapi_t *vlc_va_vaapi_Get( void *p_va )
 static int Open( vlc_va_vaapi_t *p_va, int i_codec_id )
 {
     VAProfile i_profile;
+    VAProfile *p_profiles_list;
     int i_surface_count;
+    bool b_supported_profile = false;
+    int i_profiles_nb;
+    int i_status;
 
     /* */
     switch( i_codec_id )
@@ -144,6 +148,26 @@ static int Open( vlc_va_vaapi_t *p_va, int i_codec_id )
     if( vaInitialize( p_va->p_display, &p_va->i_version_major, &p_va->i_version_minor ) )
         goto error;
 
+    /* Check if the selected profile is supported */
+    i_profiles_nb = vaMaxNumProfiles( p_va->p_display );
+    p_profiles_list = calloc( i_profiles_nb, sizeof( VAProfile ) );
+    if ( !p_profiles_list ) goto error;
+
+    i_status = vaQueryConfigProfiles( p_va->p_display, p_profiles_list, &i_profiles_nb );
+    if ( i_status == VA_STATUS_SUCCESS )
+    {
+        for( int i = 0; i < i_profiles_nb; i++ )
+        {
+            if ( p_profiles_list[i] == i_profile )
+            {
+                b_supported_profile = true;
+                break;
+            }
+        }
+    }
+    free( p_profiles_list );
+    if ( !b_supported_profile ) goto error;
+
     /* Create a VA configuration */
     VAConfigAttrib attrib;
     memset( &attrib, 0, sizeof(attrib) );
-- 
1.7.3.4



From remi at remlab.net  Tue Dec  6 16:29:17 2011
From: remi at remlab.net (=?iso-8859-1?q?R=E9mi?= Denis-Courmont)
Date: Tue, 6 Dec 2011 17:29:17 +0200
Subject: [vlc-devel] [PATCH] vaapi: fix profiles detection.
In-Reply-To: <CAK9US3AVps+1_2LmSrdzs7kZ_UGFYHAivKJjDz+wnkZZu6trNQ@mail.gmail.com>
References: <1323177188-14117-1-git-send-email-fcvlcdev@free.fr>
	<201112061620.41148.remi@remlab.net>
	<CAK9US3AVps+1_2LmSrdzs7kZ_UGFYHAivKJjDz+wnkZZu6trNQ@mail.gmail.com>
Message-ID: <201112061729.17761.remi@remlab.net>

Le mardi 6 d?cembre 2011 16:29:12 Jean-Paul Saman, vous avez ?crit :
> 2011/12/6 R?mi Denis-Courmont <remi at remlab.net>:
> > Le mardi 6 d?cembre 2011 16:16:14 Jean-Paul Saman, vous avez ?crit :
> >> > +    p_profiles_list = malloc( i_profiles_nb * sizeof( VAProfile ) );
> >> 
> >> Use calloc() here: VAProfile *p_profiles_list = calloc( i_profiles_nb,
> >> sizeof( VAProfile ) );
> > 
> > Why bother the zero the memory here?
> 
> Cause vaMaxNumProfiles() is not guarenteed to fill it up.

So what?

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From funman at videolan.org  Tue Dec  6 16:44:30 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Tue, 6 Dec 2011 10:44:30 -0500
Subject: [vlc-devel] Compiling VLC in MacOS under Lion 10.7.2
In-Reply-To: <4EDD95A5.2020905@gmail.com>
References: <4EDD95A5.2020905@gmail.com>
Message-ID: <20111206104430.7d0c1210@tok>

Le Tue, 06 Dec 2011 11:10:13 +0700,
Derek Scott <cyzicusmedia at gmail.com> a ?crit :

> 
> Hello everyone,
> 
> I've been trying to compile VLC from the git repository using the 
> instructions at http://wiki.videolan.org/OSXCompile.
> 
> 1.  I successfully get the development tools by bootstrapping in 
> vlc/extras/tools.
> 
> 2.  I select 10.6 SDK by $ export OSX_VERSION=10.6.

IIUC there is no 10.7 SDK, that's it?

> 3.  Prepare 3rd Party Libraries.  After creating the osx directory in 
> contrib, I do
> 
>   $ ../bootstrap --host=x86_64-apple-darwin11 --build=x86_64-apple-darwin11
> 
> Since I'm running in 64-bit in OS 10.7.2, I thought my host should be 
> "--host=x86_64-apple-darwin11" and my build should be 
> "--build=x86_64-apple-darwin11."

Run gcc -dumpmachine to see what it is

> 4.  Prebuilt libraries.
> 
> I try make prebuilt, but
> 
> ftp://ftp.videolan.org/pub/videolan/contrib/x86_64-apple-darwin11/vlc-contrib-x86_64-apple-darwin11-latest.tar.bz2"
> 
> fails because there's no file there.
> 
> 5.  Build your own libraries.
> 
> I try
> 
> make -j4 .gettext
> 
> This errors out with the following:
> 
> In file included from ./bindtextdom.c:24:
> /Developer/SDKs/MacOSX10.6.sdk/usr/include/stddef.h:74: error: two or 


I'd start by looking what's at line 24 of bindtextdom.c and line 74 of stddef.h


> more data types in declaration specifiers
> In file included from ./gettextP.h:23,
>                   from ./dgettext.c:23:
> /Developer/SDKs/MacOSX10.6.sdk/usr/include/stddef.h:74: error: two or 
> more data types in declaration specifiers
> In file included from ./gettextP.h:23,
>                   from ./dcgettext.c:23:
> /Developer/SDKs/MacOSX10.6.sdk/usr/include/stddef.h:74: error: two or 
> more data types in declaration specifiers
> In file included from ./gettextP.h:23,
>                   from ./gettext.c:30:
> /Developer/SDKs/MacOSX10.6.sdk/usr/include/stddef.h:74: error: two or 
> more data types in declaration specifiers
> make[4]: *** [dgettext.lo] Error 1
> make[4]: *** Waiting for unfinished jobs....
> make[4]: *** [gettext.lo] Error 1
> make[4]: *** [dcgettext.lo] Error 1
> make[4]: *** [bindtextdom.lo] Error 1
> make[3]: *** [all-recursive] Error 1
> make[2]: *** [all] Error 2
> make[1]: *** [all-recursive] Error 1
> make: *** [.gettext] Error 2
> 
> --I'd appreciate your suggestions.

-- 
Rafa?l Carr?


From jpsaman at videolan.org  Tue Dec  6 16:49:03 2011
From: jpsaman at videolan.org (Jean-Paul Saman)
Date: Tue, 6 Dec 2011 16:49:03 +0100
Subject: [vlc-devel] [PATCH] vaapi: fix profiles detection.
In-Reply-To: <1323183455-2238-1-git-send-email-fcvlcdev@free.fr>
References: <CAK9US3AR9UYihxWkYbM42e1AGAMniCH3CQ+k1yEe-qiUGnsLPg@mail.gmail.com>
	<1323183455-2238-1-git-send-email-fcvlcdev@free.fr>
Message-ID: <CAK9US3AOCpzYt+SCwTgX3J2G6h2=j3esu4oghbbDisJ9W0ULJw@mail.gmail.com>

Patch applied with minor modifications.

Please provide a patch against vlc master-branch next time.

Kind regards,
Jean-Paul Saman


From typx at dinauz.org  Tue Dec  6 23:38:15 2011
From: typx at dinauz.org (Denis Charmet)
Date: Tue,  6 Dec 2011 23:38:15 +0100
Subject: [vlc-devel] [PATCH] Fix a crash in MKV when seeking to the last
	block of a cluster
Message-ID: <1323211095-16001-1-git-send-email-typx@dinauz.org>

---
 modules/demux/mkv/Ebml_parser.cpp |   11 +++++++----
 1 files changed, 7 insertions(+), 4 deletions(-)

diff --git a/modules/demux/mkv/Ebml_parser.cpp b/modules/demux/mkv/Ebml_parser.cpp
index 69676bf..6c3aa67 100644
--- a/modules/demux/mkv/Ebml_parser.cpp
+++ b/modules/demux/mkv/Ebml_parser.cpp
@@ -88,10 +88,13 @@ EbmlElement* EbmlParser::UnGet( uint64 i_block_pos, uint64 i_cluster_pos )
     {
         // seek to the previous Cluster
         m_es->I_O().setFilePointer( i_cluster_pos, seek_beginning );
-        mi_level--;
-        mi_user_level--;
-        delete m_el[mi_level];
-        m_el[mi_level] = NULL;
+        while(mi_level > 1)
+        {
+            mi_level--;
+            mi_user_level--;
+            delete m_el[mi_level];
+            m_el[mi_level] = NULL;
+        }
         return NULL;
     }
 }
-- 
1.7.7.3



From jb at videolan.org  Wed Dec  7 02:23:20 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Wed, 7 Dec 2011 02:23:20 +0100
Subject: [vlc-devel] Compiling VLC in MacOS under Lion 10.7.2
In-Reply-To: <4EDE1CF1.5030300@gmail.com>
References: <4EDE1CF1.5030300@gmail.com>
Message-ID: <20111207012320.GA7969@videolan.org>

Hello,

On Tue, Dec 06, 2011 at 08:47:29PM +0700, Derek Scott wrote :
> 3.  Prepare 3rd Party Libraries.  After creating the osx directory
> in contrib, I do
> 
>  $ ../bootstrap --host=x86_64-apple-darwin11 --build=x86_64-apple-darwin11

Use x86_64-apple-darwin10 and it will just work fine.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Wed Dec  7 02:32:40 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Wed, 7 Dec 2011 02:32:40 +0100
Subject: [vlc-devel] [PATCH] Fix a crash in MKV when seeking to the last
 block of a cluster
In-Reply-To: <1323211095-16001-1-git-send-email-typx@dinauz.org>
References: <1323211095-16001-1-git-send-email-typx@dinauz.org>
Message-ID: <20111207013240.GA10170@videolan.org>

On Tue, Dec 06, 2011 at 11:38:15PM +0100, Denis Charmet wrote :
>  modules/demux/mkv/Ebml_parser.cpp |   11 +++++++----

Applied.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Wed Dec  7 02:33:19 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Wed, 7 Dec 2011 02:33:19 +0100
Subject: [vlc-devel] [PATCH] NEWS about luaintf
In-Reply-To: <20111205023115.GA26079@via.ecp.fr>
References: <20111205023115.GA26079@via.ecp.fr>
Message-ID: <20111207013319.GB10170@videolan.org>

On Mon, Dec 05, 2011 at 03:31:16AM +0100, Pierre Ynard wrote :
> diff --git a/NEWS b/NEWS

Applied.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Wed Dec  7 02:48:50 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Wed, 7 Dec 2011 02:48:50 +0100
Subject: [vlc-devel] [PATCH] common/win32_fullscreen: relicensed as LGPL
In-Reply-To: <CADOtZwg3spBRFw28iz-+gG++KQSoeWxw3-4VWP_pQpGNmvbtqA@mail.gmail.com>
References: <CADOtZwg3spBRFw28iz-+gG++KQSoeWxw3-4VWP_pQpGNmvbtqA@mail.gmail.com>
Message-ID: <20111207014850.GA12615@videolan.org>

On Sun, Dec 04, 2011 at 10:51:17AM +0700, Sergey Radionov wrote :
> 

Done.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Wed Dec  7 02:49:42 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Wed, 7 Dec 2011 02:49:42 +0100
Subject: [vlc-devel] RTSP muxing
In-Reply-To: <b77e083e648ed932e98a07d22eb9d17b@fe.up.pt>
References: <b77e083e648ed932e98a07d22eb9d17b@fe.up.pt>
Message-ID: <20111207014942.GB12615@videolan.org>

On Fri, Dec 02, 2011 at 03:20:29PM +0000, Nuno Mota wrote :
> PS: How can I use ffmpeg muxing in this situation?

mux=ffmpeg{mux=...}

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From fatbull at web.de  Wed Dec  7 12:39:07 2011
From: fatbull at web.de (=?ISO-8859-15?Q?Tobias_G=FCntner?=)
Date: Wed, 07 Dec 2011 12:39:07 +0100
Subject: [vlc-devel] [Patches] Dash: mostly cosmetics
In-Reply-To: <CAK-C75SV21xePRH-8pnJ-P4HMK=USxbOQtMVhRvwMGZd2_fG9w@mail.gmail.com>
References: <CAK-C75SN=k0WL==GiO89Fm1VmiqapCCZ7k1uOefyg9brc=Aafw@mail.gmail.com>
	<CAK-C75TGwHkm5iTZpydXUe3JGeW5CASrxrsu7xe3CYc5veb9Eg@mail.gmail.com>
	<CAK-C75S4oXDKecXAxwZyP6qfFRh0Q+85iPgYtpt23Yx0SWF7mA@mail.gmail.com>
	<201111241914.49264.remi@remlab.net>
	<CAK-C75TcveJQ3LeKUxA4JzupLiN5JAyUMi5=tGC+QXiTrqFhPQ@mail.gmail.com>
	<CAK-C75QRRLSqe6EKSsNEfnJGwOnjA8b80_t-aa245=bq167Rrg@mail.gmail.com>
	<4ECFD86A.3050305@web.de>
	<CAK-C75QujbKk9EP2M4aDSw26ttq8+H2ALU=RhUiraUB-BzAu4g@mail.gmail.com>
	<4ED47565.1090602@web.de>
	<CAK-C75QcXCDoJF7cpjZQRDHf30soO4bK5-90K57r98ycNb11Kg@mail.gmail.com>
	<4ED4CC73.6010008@web.de>
	<CAK-C75SiVvUN_tnh+pvo+H5xCCNN+_Afi3qG+ZzsSMthAf8QaQ@mail.gmail.com>
	<4ED71FE0.5070003@web.de>
	<CAK-C75TCOTO2dUUHz8HDPk4X395ifWNnO5bdzS2=JBJtOtkL_Q@mail.gmail.com>
	<4EDE07EF.2030803@web.de>
	<CAK-C75SV21xePRH-8pnJ-P4HMK=USxbOQtMVhRvwMGZd2_fG9w@mail.gmail.com>
Message-ID: <4EDF505B.9000309@web.de>

Am 06.12.2011 13:43, schrieb Hugo Beauz?e-Luyssen:
> Most of the error handling could be done once and for all after the
> parsing occurred.

Hmmm.

Is the parsing really done if supposedly simple getters can still fail 
because of errors in an input file? Aren't Attribute- and 
ElementNotPresentException actually (tree) parsing errors? Why do most 
getters return strings, although their names suggest otherwise? Why do 
getters returning numbers have to call atoi first, i.e. parse their own 
value? Shouldn't BasicCMParser take care of all these issues *before* 
creating MPD objects?

Regards,
Tobias



From beauze.h at gmail.com  Wed Dec  7 13:30:14 2011
From: beauze.h at gmail.com (=?UTF-8?Q?Hugo_Beauz=C3=A9e=2DLuyssen?=)
Date: Wed, 7 Dec 2011 13:30:14 +0100
Subject: [vlc-devel] [Patches] Dash: mostly cosmetics
In-Reply-To: <4EDF505B.9000309@web.de>
References: <CAK-C75SN=k0WL==GiO89Fm1VmiqapCCZ7k1uOefyg9brc=Aafw@mail.gmail.com>
	<CAK-C75TGwHkm5iTZpydXUe3JGeW5CASrxrsu7xe3CYc5veb9Eg@mail.gmail.com>
	<CAK-C75S4oXDKecXAxwZyP6qfFRh0Q+85iPgYtpt23Yx0SWF7mA@mail.gmail.com>
	<201111241914.49264.remi@remlab.net>
	<CAK-C75TcveJQ3LeKUxA4JzupLiN5JAyUMi5=tGC+QXiTrqFhPQ@mail.gmail.com>
	<CAK-C75QRRLSqe6EKSsNEfnJGwOnjA8b80_t-aa245=bq167Rrg@mail.gmail.com>
	<4ECFD86A.3050305@web.de>
	<CAK-C75QujbKk9EP2M4aDSw26ttq8+H2ALU=RhUiraUB-BzAu4g@mail.gmail.com>
	<4ED47565.1090602@web.de>
	<CAK-C75QcXCDoJF7cpjZQRDHf30soO4bK5-90K57r98ycNb11Kg@mail.gmail.com>
	<4ED4CC73.6010008@web.de>
	<CAK-C75SiVvUN_tnh+pvo+H5xCCNN+_Afi3qG+ZzsSMthAf8QaQ@mail.gmail.com>
	<4ED71FE0.5070003@web.de>
	<CAK-C75TCOTO2dUUHz8HDPk4X395ifWNnO5bdzS2=JBJtOtkL_Q@mail.gmail.com>
	<4EDE07EF.2030803@web.de>
	<CAK-C75SV21xePRH-8pnJ-P4HMK=USxbOQtMVhRvwMGZd2_fG9w@mail.gmail.com>
	<4EDF505B.9000309@web.de>
Message-ID: <CAK-C75QzQhj2257cCVH4eKFOG18=ZiG7c+UHULGcepG=hhPoCA@mail.gmail.com>

On Wed, Dec 7, 2011 at 12:39 PM, Tobias G?ntner <fatbull at web.de> wrote:
> Am 06.12.2011 13:43, schrieb Hugo Beauz?e-Luyssen:
>
>> Most of the error handling could be done once and for all after the
>> parsing occurred.
>
>
> Hmmm.
>
> Is the parsing really done if supposedly simple getters can still fail
> because of errors in an input file? Aren't Attribute- and
> ElementNotPresentException actually (tree) parsing errors? Why do most
> getters return strings, although their names suggest otherwise? Why do
> getters returning numbers have to call atoi first, i.e. parse their own
> value? Shouldn't BasicCMParser take care of all these issues *before*
> creating MPD objects?
>

I was thinking about some missing elements. Some attributes are
mandatory and should indeed be parsed in the ctor, so the getter
doesn't have to run atoi.
Some getters return strings but shouldn't. Some patches I sent fix the
problem for a few of them, some other need to be fixed.
And indeed, the parser should run some input sanity check before
trying to use the elements themselves.

>
> Regards,
> Tobias
>

Regards,


-- 
Hugo Beauz?e-Luyssen


From komh78 at gmail.com  Wed Dec  7 15:09:15 2011
From: komh78 at gmail.com (KO Myung-Hun)
Date: Wed,  7 Dec 2011 23:09:15 +0900
Subject: [vlc-devel] [PATCH 6/6] Do not define ENABLE_ROOTWRAP on OS/2
Message-ID: <1323266955-28662-1-git-send-email-komh@chollian.net>

---
 src/network/rootbind.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/src/network/rootbind.c b/src/network/rootbind.c
index f8c42cc..3d3229f 100644
--- a/src/network/rootbind.c
+++ b/src/network/rootbind.c
@@ -24,7 +24,7 @@
 
 #define _XPG4_2 /* ancilliary data on Solaris */
 
-#if !defined (WIN32)
+#if !defined (WIN32) && !defined (__OS2__)
 # define ENABLE_ROOTWRAP 1
 #endif
 
-- 
1.7.3.2



From fatbull at web.de  Wed Dec  7 17:20:23 2011
From: fatbull at web.de (=?ISO-8859-15?Q?Tobias_G=FCntner?=)
Date: Wed, 07 Dec 2011 17:20:23 +0100
Subject: [vlc-devel] [Patches] Dash: mostly cosmetics
In-Reply-To: <CAK-C75QzQhj2257cCVH4eKFOG18=ZiG7c+UHULGcepG=hhPoCA@mail.gmail.com>
References: <CAK-C75SN=k0WL==GiO89Fm1VmiqapCCZ7k1uOefyg9brc=Aafw@mail.gmail.com>
	<CAK-C75TGwHkm5iTZpydXUe3JGeW5CASrxrsu7xe3CYc5veb9Eg@mail.gmail.com>
	<CAK-C75S4oXDKecXAxwZyP6qfFRh0Q+85iPgYtpt23Yx0SWF7mA@mail.gmail.com>
	<201111241914.49264.remi@remlab.net>
	<CAK-C75TcveJQ3LeKUxA4JzupLiN5JAyUMi5=tGC+QXiTrqFhPQ@mail.gmail.com>
	<CAK-C75QRRLSqe6EKSsNEfnJGwOnjA8b80_t-aa245=bq167Rrg@mail.gmail.com>
	<4ECFD86A.3050305@web.de>
	<CAK-C75QujbKk9EP2M4aDSw26ttq8+H2ALU=RhUiraUB-BzAu4g@mail.gmail.com>
	<4ED47565.1090602@web.de>
	<CAK-C75QcXCDoJF7cpjZQRDHf30soO4bK5-90K57r98ycNb11Kg@mail.gmail.com>
	<4ED4CC73.6010008@web.de>
	<CAK-C75SiVvUN_tnh+pvo+H5xCCNN+_Afi3qG+ZzsSMthAf8QaQ@mail.gmail.com>
	<4ED71FE0.5070003@web.de>
	<CAK-C75TCOTO2dUUHz8HDPk4X395ifWNnO5bdzS2=JBJtOtkL_Q@mail.gmail.com>
	<4EDE07EF.2030803@web.de>
	<CAK-C75SV21xePRH-8pnJ-P4HMK=USxbOQtMVhRvwMGZd2_fG9w@mail.gmail.com>
	<4EDF505B.9000309@web.de>
	<CAK-C75QzQhj2257cCVH4eKFOG18=ZiG7c+UHULGcepG=hhPoCA@mail.gmail.com>
Message-ID: <4EDF9247.90301@web.de>

Am 07.12.2011 13:30, schrieb Hugo Beauz?e-Luyssen:
> On Wed, Dec 7, 2011 at 12:39 PM, Tobias G?ntner<fatbull at web.de>  wrote:
>> Am 06.12.2011 13:43, schrieb Hugo Beauz?e-Luyssen:
>>
>>> Most of the error handling could be done once and for all after the
>>> parsing occurred.
>>
>>
>> Hmmm.
>>
>> Is the parsing really done if supposedly simple getters can still fail
>> because of errors in an input file? Aren't Attribute- and
>> ElementNotPresentException actually (tree) parsing errors? Why do most
>> getters return strings, although their names suggest otherwise? Why do
>> getters returning numbers have to call atoi first, i.e. parse their own
>> value? Shouldn't BasicCMParser take care of all these issues *before*
>> creating MPD objects?
>>
>
> I was thinking about some missing elements. Some attributes are
> mandatory and should indeed be parsed in the ctor, so the getter
> doesn't have to run atoi.
> Some getters return strings but shouldn't. Some patches I sent fix the
> problem for a few of them, some other need to be fixed.
> And indeed, the parser should run some input sanity check before
> trying to use the elements themselves.

That's not what I meant. Let me try again.

I suggest parsing that stuff in a real parser, not in constructors. And 
by "parsing" I mean the transformation of an XML DOM tree into actual 
MPD objects; not the transformation of a text blob into an XML DOM tree.

Parsing is more or less "on demand" right now, i.e., whenever a getter 
is called, it looks in a map whether an attribute was present in the 
input file and parses it, or throws an exception if there was a problem. 
This means every user of the MPD classes has to deal with parsing 
errors. Wouldn't it be a lot easier to handle this in one place?

* Let's remove all parsing logic from MPD classes, specifically the 
SomethingNotPresentExceptions and the attribute maps. Instead, create 
setters for individual attributes or have the constructors initialize 
them. Either way, attribute values are always valid and have the correct 
type. Only (trivial) getters and setters and perhaps a few special 
purpose methods remain. No exceptions are thrown.

* Have BasicCMParser read and parse all attributes. If attributes are 
missing, the parser can throw an AttributeNotPresentException. If some 
attributes are optional, the parser can instead set a reasonable default 
value or the class can set it in its ctor.

* Descend recursively to parse all XML nodes. If all child nodes and 
attributes could be parsed and are valid, create the appropriate MPD 
object and return it. If some elements are mandatory or may not be 
empty, this is the place to handle that, e.g., by throwing an 
ElementNotPresentException.

Pseudo code:

class BasicCMParser {
	string attribute(Node* parent, string name) {
		return attribute value or throw AttributeNotPresentException;
	}
	string attribute(Node* parent, string name, string default) {
		return attribute value or return default;
	}
	int intAttribute(Node* parent, string name, int default, int min, int 
max) {
		return attribute value or return default or throw 
InvalidAttributeValueException;
	}
	// ditto for other attribute types (int, date, enum, etc.)

	MPD* parseMPD(Node* root) {
		// parse attributes
		minBufferTime = intAttribute(root, "minBufferTime", 0, 0, MAX_TIME);
		// parse child elements
		vector<Period> periods;
		for(...) periods.add(parsePeriod(child));
		// input is valid, create the object
		return new MPD(...);
	}
	Period* parsePeriod(Node* node) {
		...
		parseGroup(...);
		...
	}
	...
	Representation* parseRepresentation(Node* node, SegmentInfoDefault* 
default) {
		...
		if(segment info is present)
			parseSegmentInfo(...);
		else
			use SegmentInfoDefault;
		...
	}
}

When the parser returns, users (like BasicCMManager) can call getters 
without having to worry about missing attributes. Also, callers don't 
have to worry about optional elements, either. For example, a 
Representation object can always store a SegmentInfo if the parser 
ensures that it is properly filled (i.e. from a SegmentInfo or 
SegmentInfoDefault element (or whatever the standard says; I'm not that 
familiar with it)).

Regards,
Tobias

PS:
The call to strstr in DOMParser::isDash looks suspicious. stream_Peek 
doesn't return a NUL terminated string, does it?



From funman at videolan.org  Wed Dec  7 17:44:59 2011
From: funman at videolan.org (=?UTF-8?q?Rafa=C3=ABl=20Carr=C3=A9?=)
Date: Wed,  7 Dec 2011 11:44:59 -0500
Subject: [vlc-devel] [PATCH] contrib: remove directx and directshow headers
Message-ID: <1323276299-13924-1-git-send-email-funman@videolan.org>

As part of the Windows SDK, they must be provided by the dev environment
mingw-64, for example provide them

bug reports should be addressed to mingw32 and/or distributions
---
 contrib/src/README             |    4 ++--
 contrib/src/directx/SHA512SUMS |    2 --
 contrib/src/directx/rules.mak  |   22 ----------------------
 contrib/src/dshow/SHA512SUMS   |    2 --
 contrib/src/dshow/rules.mak    |   28 ----------------------------
 contrib/src/ffmpeg/rules.mak   |    1 -
 6 files changed, 2 insertions(+), 57 deletions(-)
 delete mode 100644 contrib/src/directx/SHA512SUMS
 delete mode 100644 contrib/src/directx/rules.mak
 delete mode 100644 contrib/src/dshow/SHA512SUMS
 delete mode 100644 contrib/src/dshow/rules.mak

diff --git a/contrib/src/README b/contrib/src/README
index 5d3752c..d900f98 100644
--- a/contrib/src/README
+++ b/contrib/src/README
@@ -58,9 +58,9 @@ should be listed in the PKGS special variable. See some examples:
 	# FFmpeg is always useful
 	PKGS += ffmpeg
 
-	# DirectX headers are useful only on Windows
+	# pthreads is only needed on Windows
 	ifdef HAVE_WIN32
-	PKGS += directx
+	PKGS += pthreads
 	endif
 
 	# x264 is only useful when stream output is enabled
diff --git a/contrib/src/directx/SHA512SUMS b/contrib/src/directx/SHA512SUMS
deleted file mode 100644
index dc244fa..0000000
--- a/contrib/src/directx/SHA512SUMS
+++ /dev/null
@@ -1,2 +0,0 @@
-03620b89ed8f9e9e339be9f7024d1f9a71468cba0ac2389eabd68ed1f1aa0a2d6e67c42c93f9a91eb43cf06964112dc9942ae49f8b532ff5c69db53d6b03edf8  directx-oss.tar.bz2
-e62868480379fbb7bc7a433836c6884d51e35ca2d903b9b1f0944250a9d87814be087237240ce84e6d173ea84faed5ea834f2b38800e5ac2c2d8365726defb6a  dxva2api.h
diff --git a/contrib/src/directx/rules.mak b/contrib/src/directx/rules.mak
deleted file mode 100644
index 1739994..0000000
--- a/contrib/src/directx/rules.mak
+++ /dev/null
@@ -1,22 +0,0 @@
-# DirectX headers
-
-DX_HEADERS_URL := $(CONTRIB_VIDEOLAN)/directx-oss.tar.bz2
-DXVA2_URL := $(CONTRIB_VIDEOLAN)/dxva2api.h
-
-ifdef HAVE_WIN32
-PKGS += directx
-endif
-
-$(TARBALLS)/directx-oss.tar.bz2:
-	$(call download,$(DX_HEADERS_URL))
-
-$(TARBALLS)/dxva2api.h:
-	$(call download,$(DXVA2_URL))
-
-.sum-directx: directx-oss.tar.bz2 dxva2api.h
-
-.directx: directx-oss.tar.bz2 dxva2api.h .sum-directx
-	mkdir -p -- "$(PREFIX)/include"
-	tar xvjf $< -C "$(PREFIX)/include"
-	cp $(TARBALLS)/dxva2api.h "$(PREFIX)/include/"
-	touch $@
diff --git a/contrib/src/dshow/SHA512SUMS b/contrib/src/dshow/SHA512SUMS
deleted file mode 100644
index 9d7779a..0000000
--- a/contrib/src/dshow/SHA512SUMS
+++ /dev/null
@@ -1,2 +0,0 @@
-779c4a36a4276c6e18017b898903cdf162c186b2f8de22ac12d995b316fe40d8277374eb88c695e8a760497239ad176a85ea4ecccd0df841283f95ff591997a7  dshow-headers-oss.tar.bz2
-90258b440b3ee1be13a274d5ce29d8f3bf22ad5540310e33b5f862cb8e802fc9aa7060973d066861c8236592791a666c98d8e29a4462d548cc3b9380ab2340b9  d2d_headers.tar.gz
diff --git a/contrib/src/dshow/rules.mak b/contrib/src/dshow/rules.mak
deleted file mode 100644
index 1ce4324..0000000
--- a/contrib/src/dshow/rules.mak
+++ /dev/null
@@ -1,28 +0,0 @@
-# DirectX headers
-
-DSHOW_HEADERS_URL := $(CONTRIB_VIDEOLAN)/dshow-headers-oss.tar.bz2
-D2D_HASH := b1affb70c021200b410eccd377ad5aeef2c5a85b
-D2D_URL := http://nodeload.github.com/2of1/d2d1headers/tarball/master
-# FIXME: ^ D2D not working
-
-ifdef HAVE_WIN32
-PKGS += dshow
-endif
-
-$(TARBALLS)/dshow-headers-oss.tar.bz2:
-	$(call download,$(DSHOW_HEADERS_URL))
-
-$(TARBALLS)/d2d_headers.tar.gz:
-	$(call download,$(D2D_URL))
-
-DSHOW_SOURCES := dshow-headers-oss.tar.bz2 d2d_headers.tar.gz
-
-.sum-dshow: $(DSHOW_SOURCES)
-
-.dshow: $(DSHOW_SOURCES) .sum-dshow
-	mkdir -p -- "$(PREFIX)/include"
-	tar xjf $< -C "$(PREFIX)/include" \
-		--wildcards --no-anchored '*.h' --strip-components=1
-	tar xzf $(TARBALLS)/d2d_headers.tar.gz -C "$(PREFIX)/include" \
-		 --wildcards --no-anchored '*.h' --strip-components=1
-	touch $@
diff --git a/contrib/src/ffmpeg/rules.mak b/contrib/src/ffmpeg/rules.mak
index a4ca133..efd7d46 100644
--- a/contrib/src/ffmpeg/rules.mak
+++ b/contrib/src/ffmpeg/rules.mak
@@ -71,7 +71,6 @@ endif
 
 # Windows
 ifdef HAVE_WIN32
-DEPS_ffmpeg += directx
 FFMPEGCONF += --target-os=mingw32 --enable-memalign-hack
 FFMPEGCONF += --enable-w32threads \
 	--disable-bzlib --disable-bsfs \
-- 
1.7.7.3


From jb at videolan.org  Wed Dec  7 17:58:22 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Wed, 7 Dec 2011 17:58:22 +0100
Subject: [vlc-devel] [PATCH] contrib: remove directx and directshow
 headers
In-Reply-To: <1323276299-13924-1-git-send-email-funman@videolan.org>
References: <1323276299-13924-1-git-send-email-funman@videolan.org>
Message-ID: <20111207165822.GA20191@videolan.org>

On Wed, Dec 07, 2011 at 11:44:59AM -0500, Rafa?l Carr? wrote :
> As part of the Windows SDK, they must be provided by the dev environment
> mingw-64, for example provide them
> 
> bug reports should be addressed to mingw32 and/or distributions

Rejected. Mingw headers are buggy. Mingw-w64 is far from being ready.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Wed Dec  7 18:01:27 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Wed, 7 Dec 2011 18:01:27 +0100
Subject: [vlc-devel] [vlc-commits] Dxva2 is only available in Windows
 Vista and above
In-Reply-To: <20111207164127.CF384144F23@albiero.videolan.org>
References: <20111207164127.CF384144F23@albiero.videolan.org>
Message-ID: <20111207170127.GB20191@videolan.org>

On Wed, Dec 07, 2011 at 05:41:27PM +0100, Rafa?l Carr? wrote :
> +#  undef _WIN32_WINNT

This is a horrible hack.
You should have discussed that before pushing.

> +#       define COBJMACROS

Hack again. Why ?

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Wed Dec  7 18:04:23 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Wed, 7 Dec 2011 18:04:23 +0100
Subject: [vlc-devel] [vlc-commits] contrib: change dxva2api.h checksum
In-Reply-To: <20111207170231.334EC144629@albiero.videolan.org>
References: <20111207170231.334EC144629@albiero.videolan.org>
Message-ID: <20111207170423.GC20191@videolan.org>

On Wed, Dec 07, 2011 at 06:02:31PM +0100, Rafa?l Carr? wrote :
> vlc | branch: master | Rafa?l Carr? <funman at videolan.org> | Wed Dec  7 12:01:18 2011 -0500| [8fd4111acb17d4d887f163badd25c92a9465a51e] | committer: Rafa?l Carr?
> 
> contrib: change dxva2api.h checksum
> 
> I added a workaround for mingw64
> 

What work-around?


-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From funman at videolan.org  Wed Dec  7 18:09:24 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Wed, 7 Dec 2011 12:09:24 -0500
Subject: [vlc-devel] [PATCH] contrib: remove directx and directshow
 headers
In-Reply-To: <20111207165822.GA20191@videolan.org>
References: <1323276299-13924-1-git-send-email-funman@videolan.org>
	<20111207165822.GA20191@videolan.org>
Message-ID: <20111207120924.446300d3@tok>

Le Wed, 7 Dec 2011 17:58:22 +0100,
Jean-Baptiste Kempf <jb at videolan.org> a ?crit :

> On Wed, Dec 07, 2011 at 11:44:59AM -0500, Rafa?l Carr? wrote :
> > As part of the Windows SDK, they must be provided by the dev environment
> > mingw-64, for example provide them
> > 
> > bug reports should be addressed to mingw32 and/or distributions
> 
> Rejected. Mingw headers are buggy.

Yes that's the point

> Mingw-w64 is far from being ready.

True but upstream guys are helpful and we're making good progress

-- 
Rafa?l Carr?


From funman at videolan.org  Wed Dec  7 18:10:21 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Wed, 7 Dec 2011 12:10:21 -0500
Subject: [vlc-devel] [vlc-commits] contrib: change dxva2api.h checksum
In-Reply-To: <20111207170423.GC20191@videolan.org>
References: <20111207170231.334EC144629@albiero.videolan.org>
	<20111207170423.GC20191@videolan.org>
Message-ID: <20111207121021.12b430d4@tok>

Le Wed, 7 Dec 2011 18:04:23 +0100,
Jean-Baptiste Kempf <jb at videolan.org> a ?crit :

> On Wed, Dec 07, 2011 at 06:02:31PM +0100, Rafa?l Carr? wrote :
> > vlc | branch: master | Rafa?l Carr? <funman at videolan.org> | Wed Dec  7 12:01:18 2011 -0500| [8fd4111acb17d4d887f163badd25c92a9465a51e] | committer: Rafa?l Carr?
> > 
> > contrib: change dxva2api.h checksum
> > 
> > I added a workaround for mingw64
> > 
> 
> What work-around?

#ifndef _COM_interface
#define _COM_interface struct
#endif

-- 
Rafa?l Carr?


From jb at videolan.org  Wed Dec  7 18:11:07 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Wed, 7 Dec 2011 18:11:07 +0100
Subject: [vlc-devel] [PATCH] contrib: remove directx and directshow
 headers
In-Reply-To: <20111207120924.446300d3@tok>
References: <1323276299-13924-1-git-send-email-funman@videolan.org>
	<20111207165822.GA20191@videolan.org> <20111207120924.446300d3@tok>
Message-ID: <20111207171107.GA28811@videolan.org>

On Wed, Dec 07, 2011 at 12:09:24PM -0500, Rafa?l Carr? wrote :
> True but upstream guys are helpful and we're making good progress

By adding work-arounds to their headers in both libav and on our FTP?

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From john at zavgren.com  Wed Dec  7 18:12:43 2011
From: john at zavgren.com (John Zavgren)
Date: Wed, 7 Dec 2011 12:12:43 -0500
Subject: [vlc-devel] Tap, tap, is this thing "on"
Message-ID: <CADF-EHGJiH9Do7EUPby_sHE7siq-pTBwwvaHJYpu0x7gSZO7FQ@mail.gmail.com>

Greetings:
I've been trying to post a question to vlc-devel, but my attempts seem
to be failing silently.



-- 
I refuse to believe that corporations are actually people until the
state of Texas executes one.


John Zavgren
603-371-0513 (home)
603-801-2094 (cell)
johnzavgren (skype)
603-821-0904 (skype)
john at zavgren.com


From jb at videolan.org  Wed Dec  7 18:14:19 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Wed, 7 Dec 2011 18:14:19 +0100
Subject: [vlc-devel] [vlc-commits] contrib: change dxva2api.h checksum
In-Reply-To: <20111207121021.12b430d4@tok>
References: <20111207170231.334EC144629@albiero.videolan.org>
	<20111207170423.GC20191@videolan.org> <20111207121021.12b430d4@tok>
Message-ID: <20111207171419.GB28811@videolan.org>

On Wed, Dec 07, 2011 at 12:10:21PM -0500, Rafa?l Carr? wrote :
> Le Wed, 7 Dec 2011 18:04:23 +0100,
> Jean-Baptiste Kempf <jb at videolan.org> a ?crit :
> 
> > On Wed, Dec 07, 2011 at 06:02:31PM +0100, Rafa?l Carr? wrote :
> > > vlc | branch: master | Rafa?l Carr? <funman at videolan.org> | Wed Dec  7 12:01:18 2011 -0500| [8fd4111acb17d4d887f163badd25c92a9465a51e] | committer: Rafa?l Carr?
> > > 
> > > contrib: change dxva2api.h checksum
> > > 
> > > I added a workaround for mingw64
> > > 
> > 
> > What work-around?
> 
> #ifndef _COM_interface
> #define _COM_interface struct
> #endif

Not defining COM_interface is a major global bug of mingw-w64
Not defining COM_interface is a major global bug of mingw-w64


Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From funman at videolan.org  Wed Dec  7 18:14:22 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Wed, 7 Dec 2011 12:14:22 -0500
Subject: [vlc-devel] [vlc-commits] Dxva2 is only available in Windows
 Vista and above
In-Reply-To: <20111207170127.GB20191@videolan.org>
References: <20111207164127.CF384144F23@albiero.videolan.org>
	<20111207170127.GB20191@videolan.org>
Message-ID: <20111207121422.5ac11de3@tok>

Le Wed, 7 Dec 2011 18:01:27 +0100,
Jean-Baptiste Kempf <jb at videolan.org> a ?crit :

> On Wed, Dec 07, 2011 at 05:41:27PM +0100, Rafa?l Carr? wrote :
> > +#  undef _WIN32_WINNT
> 
> This is a horrible hack.

We use AC_DEFINE to define it to XP compatibility before checking dxva2 headers so it's required

> You should have discussed that before pushing.



> > +#       define COBJMACROS
> 
> Hack again. Why ?

I found this in libav, it's required to get some helpers like:
#define IDirectXVideoDecoder_Release(This) (This)->lpVtbl->Release(This)

It's mentioned in msdn but I didn't a definition though

-- 
Rafa?l Carr?


From funman at videolan.org  Wed Dec  7 18:20:34 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Wed, 7 Dec 2011 12:20:34 -0500
Subject: [vlc-devel] [PATCH] contrib: remove directx and directshow
 headers
In-Reply-To: <20111207171107.GA28811@videolan.org>
References: <1323276299-13924-1-git-send-email-funman@videolan.org>
	<20111207165822.GA20191@videolan.org> <20111207120924.446300d3@tok>
	<20111207171107.GA28811@videolan.org>
Message-ID: <20111207122034.58b6d947@tok>

Le Wed, 7 Dec 2011 18:11:07 +0100,
Jean-Baptiste Kempf <jb at videolan.org> a ?crit :

> On Wed, Dec 07, 2011 at 12:09:24PM -0500, Rafa?l Carr? wrote :
> > True but upstream guys are helpful and we're making good progress
> 
> By adding work-arounds to their headers in both libav and on our FTP?

Nope, by fixing the mingw32 workaround that you refuse to remove.

The commit log of the patch I sent to libav should be clear I think

-- 
Rafa?l Carr?


From funman at videolan.org  Wed Dec  7 18:22:11 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Wed, 7 Dec 2011 12:22:11 -0500
Subject: [vlc-devel] [vlc-commits] contrib: change dxva2api.h checksum
In-Reply-To: <20111207171419.GB28811@videolan.org>
References: <20111207170231.334EC144629@albiero.videolan.org>
	<20111207170423.GC20191@videolan.org> <20111207121021.12b430d4@tok>
	<20111207171419.GB28811@videolan.org>
Message-ID: <20111207122211.3a4e1598@tok>

Le Wed, 7 Dec 2011 18:14:19 +0100,
Jean-Baptiste Kempf <jb at videolan.org> a ?crit :

> On Wed, Dec 07, 2011 at 12:10:21PM -0500, Rafa?l Carr? wrote :
> > Le Wed, 7 Dec 2011 18:04:23 +0100,
> > Jean-Baptiste Kempf <jb at videolan.org> a ?crit :
> > 
> > > On Wed, Dec 07, 2011 at 06:02:31PM +0100, Rafa?l Carr? wrote :
> > > > vlc | branch: master | Rafa?l Carr? <funman at videolan.org> | Wed Dec  7 12:01:18 2011 -0500| [8fd4111acb17d4d887f163badd25c92a9465a51e] | committer: Rafa?l Carr?
> > > > 
> > > > contrib: change dxva2api.h checksum
> > > > 
> > > > I added a workaround for mingw64
> > > > 
> > > 
> > > What work-around?
> > 
> > #ifndef _COM_interface
> > #define _COM_interface struct
> > #endif
> 
> Not defining COM_interface is a major global bug of mingw-w64
> Not defining COM_interface is a major global bug of mingw-w64

mingw64 dxva2api.h works properly, but our own custom broken version is
used before it

-- 
Rafa?l Carr?


From jb at videolan.org  Wed Dec  7 18:22:35 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Wed, 7 Dec 2011 18:22:35 +0100
Subject: [vlc-devel] [vlc-commits] Dxva2 is only available in Windows
 Vista and above
In-Reply-To: <20111207121422.5ac11de3@tok>
References: <20111207164127.CF384144F23@albiero.videolan.org>
	<20111207170127.GB20191@videolan.org> <20111207121422.5ac11de3@tok>
Message-ID: <20111207172235.GD28811@videolan.org>

On Wed, Dec 07, 2011 at 12:14:22PM -0500, Rafa?l Carr? wrote :
> Le Wed, 7 Dec 2011 18:01:27 +0100,
> Jean-Baptiste Kempf <jb at videolan.org> a ?crit :
> 
> > On Wed, Dec 07, 2011 at 05:41:27PM +0100, Rafa?l Carr? wrote :
> > > +#  undef _WIN32_WINNT
> > 
> > This is a horrible hack.
> 
> We use AC_DEFINE to define it to XP compatibility before checking dxva2 headers so it's required

It isn't required, since we need to check the header presence, not to
compile them.

> I found this in libav, it's required to get some helpers like:
> #define IDirectXVideoDecoder_Release(This) (This)->lpVtbl->Release(This)

It isn't needed, if you have correctly defined the CINTERFACE, as you
can clearly see at the bottom of the dxva2api.h file.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Wed Dec  7 18:23:36 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Wed, 7 Dec 2011 18:23:36 +0100
Subject: [vlc-devel] [vlc-commits] contrib: change dxva2api.h checksum
In-Reply-To: <20111207122211.3a4e1598@tok>
References: <20111207170231.334EC144629@albiero.videolan.org>
	<20111207170423.GC20191@videolan.org> <20111207121021.12b430d4@tok>
	<20111207171419.GB28811@videolan.org> <20111207122211.3a4e1598@tok>
Message-ID: <20111207172336.GE28811@videolan.org>

On Wed, Dec 07, 2011 at 12:22:11PM -0500, Rafa?l Carr? wrote :
> Le Wed, 7 Dec 2011 18:14:19 +0100,
> Jean-Baptiste Kempf <jb at videolan.org> a ?crit :
> 
> > On Wed, Dec 07, 2011 at 12:10:21PM -0500, Rafa?l Carr? wrote :
> > > Le Wed, 7 Dec 2011 18:04:23 +0100,
> > > Jean-Baptiste Kempf <jb at videolan.org> a ?crit :
> > > 
> > > > On Wed, Dec 07, 2011 at 06:02:31PM +0100, Rafa?l Carr? wrote :
> > > > > vlc | branch: master | Rafa?l Carr? <funman at videolan.org> | Wed Dec  7 12:01:18 2011 -0500| [8fd4111acb17d4d887f163badd25c92a9465a51e] | committer: Rafa?l Carr?
> > > > > 
> > > > > contrib: change dxva2api.h checksum
> > > > > 
> > > > > I added a workaround for mingw64
> > > > > 
> > > > 
> > > > What work-around?
> > > 
> > > #ifndef _COM_interface
> > > #define _COM_interface struct
> > > #endif
> > 
> > Not defining COM_interface is a major global bug of mingw-w64
> > Not defining COM_interface is a major global bug of mingw-w64
> 
> mingw64 dxva2api.h works properly, but our own custom broken version is
> used before it

You add _COM_interface for mingw64, but then OUR version is buggy? I
don't get it.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Wed Dec  7 18:24:50 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Wed, 7 Dec 2011 18:24:50 +0100
Subject: [vlc-devel] [PATCH] contrib: remove directx and directshow
 headers
In-Reply-To: <20111207122034.58b6d947@tok>
References: <1323276299-13924-1-git-send-email-funman@videolan.org>
	<20111207165822.GA20191@videolan.org> <20111207120924.446300d3@tok>
	<20111207171107.GA28811@videolan.org> <20111207122034.58b6d947@tok>
Message-ID: <20111207172450.GF28811@videolan.org>

On Wed, Dec 07, 2011 at 12:20:34PM -0500, Rafa?l Carr? wrote :
> Le Wed, 7 Dec 2011 18:11:07 +0100,
> Jean-Baptiste Kempf <jb at videolan.org> a ?crit :
> 
> > On Wed, Dec 07, 2011 at 12:09:24PM -0500, Rafa?l Carr? wrote :
> > > True but upstream guys are helpful and we're making good progress
> > 
> > By adding work-arounds to their headers in both libav and on our FTP?
> 
> Nope, by fixing the mingw32 workaround that you refuse to remove.
> 
> The commit log of the patch I sent to libav should be clear I think

Dxva2api.h needs d3d9.h, therefore dxva2api.h should include it, not the
code including dxva2api.h

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From funman at videolan.org  Wed Dec  7 18:25:50 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Wed, 7 Dec 2011 12:25:50 -0500
Subject: [vlc-devel] [vlc-commits] Dxva2 is only available in Windows
 Vista and above
In-Reply-To: <20111207172235.GD28811@videolan.org>
References: <20111207164127.CF384144F23@albiero.videolan.org>
	<20111207170127.GB20191@videolan.org> <20111207121422.5ac11de3@tok>
	<20111207172235.GD28811@videolan.org>
Message-ID: <20111207122550.333f5f22@tok>

Le Wed, 7 Dec 2011 18:22:35 +0100,
Jean-Baptiste Kempf <jb at videolan.org> a ?crit :

> On Wed, Dec 07, 2011 at 12:14:22PM -0500, Rafa?l Carr? wrote :
> > Le Wed, 7 Dec 2011 18:01:27 +0100,
> > Jean-Baptiste Kempf <jb at videolan.org> a ?crit :
> > 
> > > On Wed, Dec 07, 2011 at 05:41:27PM +0100, Rafa?l Carr? wrote :
> > > > +#  undef _WIN32_WINNT
> > > 
> > > This is a horrible hack.
> > 
> > We use AC_DEFINE to define it to XP compatibility before checking dxva2 headers so it's required
> 
> It isn't required, since we need to check the header presence, not to
> compile them.

Please read autoconf documentation, we do compile them.

> > I found this in libav, it's required to get some helpers like:
> > #define IDirectXVideoDecoder_Release(This) (This)->lpVtbl->Release(This)
> 
> It isn't needed, if you have correctly defined the CINTERFACE, as you
> can clearly see at the bottom of the dxva2api.h file.

I don't know how geal and fenrir wrote this file but if they could
comment that'd be welcome of course.

From libavcodec/dxva2_internal.h:

#define _WIN32_WINNT 0x0600
#define COBJMACROS
#include "dxva2.h"


-- 
Rafa?l Carr?


From jb at videolan.org  Wed Dec  7 18:27:44 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Wed, 7 Dec 2011 18:27:44 +0100
Subject: [vlc-devel] [vlc-commits] Dxva2 is only available in Windows
 Vista and above
In-Reply-To: <20111207122550.333f5f22@tok>
References: <20111207164127.CF384144F23@albiero.videolan.org>
	<20111207170127.GB20191@videolan.org> <20111207121422.5ac11de3@tok>
	<20111207172235.GD28811@videolan.org> <20111207122550.333f5f22@tok>
Message-ID: <20111207172744.GA32159@videolan.org>

On Wed, Dec 07, 2011 at 12:25:50PM -0500, Rafa?l Carr? wrote :
> Please read autoconf documentation, we do compile them.

Not having WINNT at 600 should not prevent compilation.


-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From funman at videolan.org  Wed Dec  7 18:33:42 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Wed, 7 Dec 2011 12:33:42 -0500
Subject: [vlc-devel] [PATCH] contrib: remove directx and directshow
 headers
In-Reply-To: <20111207172450.GF28811@videolan.org>
References: <1323276299-13924-1-git-send-email-funman@videolan.org>
	<20111207165822.GA20191@videolan.org> <20111207120924.446300d3@tok>
	<20111207171107.GA28811@videolan.org> <20111207122034.58b6d947@tok>
	<20111207172450.GF28811@videolan.org>
Message-ID: <20111207123342.56ac8e4b@tok>

Le Wed, 7 Dec 2011 18:24:50 +0100,
Jean-Baptiste Kempf <jb at videolan.org> a ?crit :

> On Wed, Dec 07, 2011 at 12:20:34PM -0500, Rafa?l Carr? wrote :
> > Le Wed, 7 Dec 2011 18:11:07 +0100,
> > Jean-Baptiste Kempf <jb at videolan.org> a ?crit :
> > 
> > > On Wed, Dec 07, 2011 at 12:09:24PM -0500, Rafa?l Carr? wrote :
> > > > True but upstream guys are helpful and we're making good progress
> > > 
> > > By adding work-arounds to their headers in both libav and on our FTP?
> > 
> > Nope, by fixing the mingw32 workaround that you refuse to remove.
> > 
> > The commit log of the patch I sent to libav should be clear I think
> 
> Dxva2api.h needs d3d9.h, therefore dxva2api.h should include it, not the
> code including dxva2api.h

I checked MSDN documentation, the libav commit should have a link to it.

-- 
Rafa?l Carr?


From remi at remlab.net  Wed Dec  7 18:36:12 2011
From: remi at remlab.net (=?iso-8859-1?q?R=E9mi?= Denis-Courmont)
Date: Wed, 7 Dec 2011 19:36:12 +0200
Subject: [vlc-devel] [vlc-commits] Dxva2 is only available in Windows
	Vista and above
In-Reply-To: <20111207172235.GD28811@videolan.org>
References: <20111207164127.CF384144F23@albiero.videolan.org>
	<20111207121422.5ac11de3@tok> <20111207172235.GD28811@videolan.org>
Message-ID: <201112071936.12551.remi@remlab.net>

Le mercredi 7 d?cembre 2011 19:22:35 Jean-Baptiste Kempf, vous avez ?crit :
> On Wed, Dec 07, 2011 at 12:14:22PM -0500, Rafa?l Carr? wrote :
> > Le Wed, 7 Dec 2011 18:01:27 +0100,
> > 
> > Jean-Baptiste Kempf <jb at videolan.org> a ?crit :
> > > On Wed, Dec 07, 2011 at 05:41:27PM +0100, Rafa?l Carr? wrote :
> > > > +#  undef _WIN32_WINNT
> > > 
> > > This is a horrible hack.
> > 
> > We use AC_DEFINE to define it to XP compatibility before checking dxva2
> > headers so it's required
> 
> It isn't required, since we need to check the header presence, not to
> compile them.

WHAAAAAT? autoconf checks for headers precompilation, not just file presence.


-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From funman at videolan.org  Wed Dec  7 18:36:43 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Wed, 7 Dec 2011 12:36:43 -0500
Subject: [vlc-devel] [vlc-commits] contrib: change dxva2api.h checksum
In-Reply-To: <20111207172336.GE28811@videolan.org>
References: <20111207170231.334EC144629@albiero.videolan.org>
	<20111207170423.GC20191@videolan.org> <20111207121021.12b430d4@tok>
	<20111207171419.GB28811@videolan.org> <20111207122211.3a4e1598@tok>
	<20111207172336.GE28811@videolan.org>
Message-ID: <20111207123643.37efc03a@tok>

Le Wed, 7 Dec 2011 18:23:36 +0100,
Jean-Baptiste Kempf <jb at videolan.org> a ?crit :

> On Wed, Dec 07, 2011 at 12:22:11PM -0500, Rafa?l Carr? wrote :
> > Le Wed, 7 Dec 2011 18:14:19 +0100,
> > Jean-Baptiste Kempf <jb at videolan.org> a ?crit :
> > 
> > > On Wed, Dec 07, 2011 at 12:10:21PM -0500, Rafa?l Carr? wrote :
> > > > Le Wed, 7 Dec 2011 18:04:23 +0100,
> > > > Jean-Baptiste Kempf <jb at videolan.org> a ?crit :
> > > > 
> > > > > On Wed, Dec 07, 2011 at 06:02:31PM +0100, Rafa?l Carr? wrote :
> > > > > > vlc | branch: master | Rafa?l Carr? <funman at videolan.org> | Wed Dec  7 12:01:18 2011 -0500| [8fd4111acb17d4d887f163badd25c92a9465a51e] | committer: Rafa?l Carr?
> > > > > > 
> > > > > > contrib: change dxva2api.h checksum
> > > > > > 
> > > > > > I added a workaround for mingw64
> > > > > > 
> > > > > 
> > > > > What work-around?
> > > > 
> > > > #ifndef _COM_interface
> > > > #define _COM_interface struct
> > > > #endif
> > > 
> > > Not defining COM_interface is a major global bug of mingw-w64
> > > Not defining COM_interface is a major global bug of mingw-w64
> > 
> > mingw64 dxva2api.h works properly, but our own custom broken version is
> > used before it
> 
> You add _COM_interface for mingw64, but then OUR version is buggy? I
> don't get it.
> 
> Best regards,

Our dxva2api.h fails to define _COM_interface

-- 
Rafa?l Carr?


From remi at remlab.net  Wed Dec  7 18:42:47 2011
From: remi at remlab.net (=?iso-8859-1?q?R=E9mi?= Denis-Courmont)
Date: Wed, 7 Dec 2011 19:42:47 +0200
Subject: [vlc-devel] [vlc-commits] Dxva2 is only available in Windows
	Vista and above
In-Reply-To: <20111207172744.GA32159@videolan.org>
References: <20111207164127.CF384144F23@albiero.videolan.org>
	<20111207122550.333f5f22@tok> <20111207172744.GA32159@videolan.org>
Message-ID: <201112071942.47845.remi@remlab.net>

Le mercredi 7 d?cembre 2011 19:27:44 Jean-Baptiste Kempf, vous avez ?crit :
> On Wed, Dec 07, 2011 at 12:25:50PM -0500, Rafa?l Carr? wrote :
> > Please read autoconf documentation, we do compile them.
> 
> Not having WINNT at 600 should not prevent compilation.

Failing to precompiling would a problem (and IMHO better).

But in any case, headers must be tested in autoconf with the same definitions 
as they will be included with. autoconf normally does that automatically. But 
if some definitions is overriden, then I don't see any clean alternative to 
Rafa?l's hack.

Otherwise the reciprocal problem shows up:
the header succeeds in autoconf and fails in make!


So unless you're suggesting to drop Windows XP/2003 support (fine with me), I 
don't see any better solution.

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From jb at videolan.org  Wed Dec  7 18:43:43 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Wed, 7 Dec 2011 18:43:43 +0100
Subject: [vlc-devel] [vlc-commits] contrib: change dxva2api.h checksum
In-Reply-To: <20111207123643.37efc03a@tok>
References: <20111207170231.334EC144629@albiero.videolan.org>
	<20111207170423.GC20191@videolan.org> <20111207121021.12b430d4@tok>
	<20111207171419.GB28811@videolan.org> <20111207122211.3a4e1598@tok>
	<20111207172336.GE28811@videolan.org> <20111207123643.37efc03a@tok>
Message-ID: <20111207174342.GB32159@videolan.org>

On Wed, Dec 07, 2011 at 12:36:43PM -0500, Rafa?l Carr? wrote :
> Our dxva2api.h fails to define _COM_interface

Then, the right way is to fix it correctly, by removing reference to
_COM_interface


-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From remi at remlab.net  Wed Dec  7 18:44:24 2011
From: remi at remlab.net (=?iso-8859-1?q?R=E9mi?= Denis-Courmont)
Date: Wed, 7 Dec 2011 19:44:24 +0200
Subject: [vlc-devel] [PATCH] contrib: remove directx and directshow
	headers
In-Reply-To: <1323276299-13924-1-git-send-email-funman@videolan.org>
References: <1323276299-13924-1-git-send-email-funman@videolan.org>
Message-ID: <201112071944.24365.remi@remlab.net>

Le mercredi 7 d?cembre 2011 18:44:59 Rafa?l Carr?, vous avez ?crit :
> diff --git a/contrib/src/README b/contrib/src/README
> index 5d3752c..d900f98 100644
> --- a/contrib/src/README
> +++ b/contrib/src/README
> @@ -58,9 +58,9 @@ should be listed in the PKGS special variable. See some
> examples: # FFmpeg is always useful
>  	PKGS += ffmpeg
> 
> -	# DirectX headers are useful only on Windows
> +	# pthreads is only needed on Windows
>  	ifdef HAVE_WIN32
> -	PKGS += directx
> +	PKGS += pthreads
>  	endif

What does pthread have to do with this path?!

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From jb at videolan.org  Wed Dec  7 18:47:50 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Wed, 7 Dec 2011 18:47:50 +0100
Subject: [vlc-devel] [PATCH] contrib: remove directx and directshow
 headers
In-Reply-To: <20111207123342.56ac8e4b@tok>
References: <1323276299-13924-1-git-send-email-funman@videolan.org>
	<20111207165822.GA20191@videolan.org> <20111207120924.446300d3@tok>
	<20111207171107.GA28811@videolan.org> <20111207122034.58b6d947@tok>
	<20111207172450.GF28811@videolan.org> <20111207123342.56ac8e4b@tok>
Message-ID: <20111207174750.GC32159@videolan.org>

On Wed, Dec 07, 2011 at 12:33:42PM -0500, Rafa?l Carr? wrote :
> I checked MSDN documentation, the libav commit should have a link to it.

It doesn't.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From remi at remlab.net  Wed Dec  7 19:00:18 2011
From: remi at remlab.net (=?iso-8859-1?q?R=E9mi?= Denis-Courmont)
Date: Wed, 7 Dec 2011 20:00:18 +0200
Subject: [vlc-devel] [PATCH 6/6] Do not define ENABLE_ROOTWRAP on OS/2
In-Reply-To: <1323266955-28662-1-git-send-email-komh@chollian.net>
References: <1323266955-28662-1-git-send-email-komh@chollian.net>
Message-ID: <201112072000.18753.remi@remlab.net>

Merged, thanks.

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From john at zavgren.com  Wed Dec  7 19:02:30 2011
From: john at zavgren.com (John Zavgren)
Date: Wed, 7 Dec 2011 13:02:30 -0500
Subject: [vlc-devel] Video packet reordering
Message-ID: <CADF-EHFB=9OuBC0O8vkroE5J=osEvC+6_1wAhQM1G4BAv+_wbw@mail.gmail.com>

Greetings:

I'm sending video frames (via vlc) over an IP network that reorders
datagrams... so much so that sequential arrivals are rare events.

I'm using RTP/UDP. The results are horrible.

Are there packet resequencing functions in vlc that I can activate
that leverage off the sequence numbers in RTP, and reverse these
network impairments? Allegedly, RTP can compensate, and I assume the
issue is whether vlc implements the full RFC?

Thanks!
John Zavgren
john at zavgren.com


-- 
I refuse to believe that corporations are actually people until the
state of Texas executes one.


John Zavgren
603-371-0513 (home)
603-801-2094 (cell)
johnzavgren (skype)
603-821-0904 (skype)
john at zavgren.com


From remi at remlab.net  Wed Dec  7 19:02:58 2011
From: remi at remlab.net (=?iso-8859-15?q?R=E9mi?= Denis-Courmont)
Date: Wed, 7 Dec 2011 20:02:58 +0200
Subject: [vlc-devel] [PATCH] live555 : enable/disable tracks on the fly
In-Reply-To: <1322730254.10961.3.camel@stim-desktop>
References: <1322730254.10961.3.camel@stim-desktop>
Message-ID: <201112072002.59492.remi@remlab.net>

Le jeudi 1 d?cembre 2011 11:04:14 S?bastien Escudier, vous avez ?crit :
> 3 patches attached.
> I did not test multiple audio or multiple video tracks case because I
> don't have rtsp servers like that.

All three patches failed to merge.

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From remi at remlab.net  Wed Dec  7 19:10:23 2011
From: remi at remlab.net (=?iso-8859-1?q?R=E9mi?= Denis-Courmont)
Date: Wed, 7 Dec 2011 20:10:23 +0200
Subject: [vlc-devel] Video packet reordering
In-Reply-To: <CADF-EHFB=9OuBC0O8vkroE5J=osEvC+6_1wAhQM1G4BAv+_wbw@mail.gmail.com>
References: <CADF-EHFB=9OuBC0O8vkroE5J=osEvC+6_1wAhQM1G4BAv+_wbw@mail.gmail.com>
Message-ID: <201112072010.23659.remi@remlab.net>

Le mercredi 7 d?cembre 2011 20:02:30 John Zavgren, vous avez ?crit :
> Are there packet resequencing functions in vlc that I can activate
> that leverage off the sequence numbers in RTP, and reverse these
> network impairments?

If you use the built-in RTP receiver, you can change the maximum drop-out, 
maximum misorder and timeout from the preferences.

> Allegedly, RTP can compensate, and I assume the
> issue is whether vlc implements the full RFC?

That question means nothing as there are many different RFC's involved and 
there is no single standardized reordering process (as far as I know).

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From funman at videolan.org  Wed Dec  7 19:25:03 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Wed, 7 Dec 2011 13:25:03 -0500
Subject: [vlc-devel] [PATCH] contrib: remove directx and directshow
 headers
In-Reply-To: <20111207174750.GC32159@videolan.org>
References: <1323276299-13924-1-git-send-email-funman@videolan.org>
	<20111207165822.GA20191@videolan.org> <20111207120924.446300d3@tok>
	<20111207171107.GA28811@videolan.org> <20111207122034.58b6d947@tok>
	<20111207172450.GF28811@videolan.org> <20111207123342.56ac8e4b@tok>
	<20111207174750.GC32159@videolan.org>
Message-ID: <20111207132503.15da22fc@tok>

Le Wed, 7 Dec 2011 18:47:50 +0100,
Jean-Baptiste Kempf <jb at videolan.org> a ?crit :

> On Wed, Dec 07, 2011 at 12:33:42PM -0500, Rafa?l Carr? wrote :
> > I checked MSDN documentation, the libav commit should have a link to it.
> 
> It doesn't.

Hm I might have forgotten it, or libav's mailman might have removed it.

In any case: http://msdn.microsoft.com/en-us/library/windows/desktop/bb205892(v=vs.85).aspx

LPDIRECT3DSURFACE9 needs d3d9.h

-- 
Rafa?l Carr?


From funman at videolan.org  Wed Dec  7 19:27:27 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Wed, 7 Dec 2011 13:27:27 -0500
Subject: [vlc-devel] [PATCH] contrib: remove directx and directshow
 headers
In-Reply-To: <201112071944.24365.remi@remlab.net>
References: <1323276299-13924-1-git-send-email-funman@videolan.org>
	<201112071944.24365.remi@remlab.net>
Message-ID: <20111207132727.3b0cc507@tok>

Le Wed, 7 Dec 2011 19:44:24 +0200,
"R?mi Denis-Courmont" <remi at remlab.net> a ?crit :

> Le mercredi 7 d?cembre 2011 18:44:59 Rafa?l Carr?, vous avez ?crit :
> > diff --git a/contrib/src/README b/contrib/src/README
> > index 5d3752c..d900f98 100644
> > --- a/contrib/src/README
> > +++ b/contrib/src/README
> > @@ -58,9 +58,9 @@ should be listed in the PKGS special variable. See some
> > examples: # FFmpeg is always useful
> >  	PKGS += ffmpeg
> > 
> > -	# DirectX headers are useful only on Windows
> > +	# pthreads is only needed on Windows
> >  	ifdef HAVE_WIN32
> > -	PKGS += directx
> > +	PKGS += pthreads
> >  	endif
> 
> What does pthread have to do with this path?!

I quickly found something to add as an example instead of directx (note
the file being modified).

It doesn't really matter since that patch was not meant to be applied,
at least not immediately.

We don't want to break our win32 setup, right?

-- 
Rafa?l Carr?


From funman at videolan.org  Wed Dec  7 19:28:51 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Wed, 7 Dec 2011 13:28:51 -0500
Subject: [vlc-devel] [vlc-commits] contrib: change dxva2api.h checksum
In-Reply-To: <20111207174342.GB32159@videolan.org>
References: <20111207170231.334EC144629@albiero.videolan.org>
	<20111207170423.GC20191@videolan.org> <20111207121021.12b430d4@tok>
	<20111207171419.GB28811@videolan.org> <20111207122211.3a4e1598@tok>
	<20111207172336.GE28811@videolan.org> <20111207123643.37efc03a@tok>
	<20111207174342.GB32159@videolan.org>
Message-ID: <20111207132851.187228e5@tok>

Le Wed, 7 Dec 2011 18:43:43 +0100,
Jean-Baptiste Kempf <jb at videolan.org> a ?crit :

> On Wed, Dec 07, 2011 at 12:36:43PM -0500, Rafa?l Carr? wrote :
> > Our dxva2api.h fails to define _COM_interface
> 
> Then, the right way is to fix it correctly, by removing reference to
> _COM_interface

I'm OK with someone else doing that, though if they do, I don't mind
them asking me if that works on mingw64 if they don't check themselves.

-- 
Rafa?l Carr?


From remi at remlab.net  Wed Dec  7 19:33:27 2011
From: remi at remlab.net (=?utf-8?q?R=C3=A9mi?= Denis-Courmont)
Date: Wed, 7 Dec 2011 20:33:27 +0200
Subject: [vlc-devel] [PATCH] contrib: remove directx and directshow
	headers
In-Reply-To: <20111207132727.3b0cc507@tok>
References: <1323276299-13924-1-git-send-email-funman@videolan.org>
	<201112071944.24365.remi@remlab.net> <20111207132727.3b0cc507@tok>
Message-ID: <201112072033.28325.remi@remlab.net>

Le mercredi 7 d?cembre 2011 20:27:27 Rafa?l Carr?, vous avez ?crit :
> We don't want to break our win32 setup, right?

I don't know if "we" want to break it now. But now is a better time to break 
it in vlc.git than two weeks before the 1.3.0 freeze.

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From jb at videolan.org  Wed Dec  7 19:36:40 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Wed, 7 Dec 2011 19:36:40 +0100
Subject: [vlc-devel] [PATCH] contrib: remove directx and directshow
 headers
In-Reply-To: <20111207132503.15da22fc@tok>
References: <1323276299-13924-1-git-send-email-funman@videolan.org>
	<20111207165822.GA20191@videolan.org> <20111207120924.446300d3@tok>
	<20111207171107.GA28811@videolan.org> <20111207122034.58b6d947@tok>
	<20111207172450.GF28811@videolan.org> <20111207123342.56ac8e4b@tok>
	<20111207174750.GC32159@videolan.org> <20111207132503.15da22fc@tok>
Message-ID: <20111207183640.GA11126@videolan.org>

On Wed, Dec 07, 2011 at 01:25:03PM -0500, Rafa?l Carr? wrote :
> Le Wed, 7 Dec 2011 18:47:50 +0100,
> Jean-Baptiste Kempf <jb at videolan.org> a ?crit :
> 
> > On Wed, Dec 07, 2011 at 12:33:42PM -0500, Rafa?l Carr? wrote :
> > > I checked MSDN documentation, the libav commit should have a link to it.
> > 
> > It doesn't.
> 
> Hm I might have forgotten it, or libav's mailman might have removed it.
> 
> In any case: http://msdn.microsoft.com/en-us/library/windows/desktop/bb205892(v=vs.85).aspx
> 
> LPDIRECT3DSURFACE9 needs d3d9.h

Sure, but what I am saying is that other things in dxva2api.h do need
d3d9.h, like D3DFORMAT, D3DPOOL, so the include should be in dxva2api.h,
not in the client code.

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From funman at videolan.org  Wed Dec  7 19:41:17 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Wed, 7 Dec 2011 13:41:17 -0500
Subject: [vlc-devel] [PATCH] contrib: remove directx and directshow
 headers
In-Reply-To: <20111207183640.GA11126@videolan.org>
References: <1323276299-13924-1-git-send-email-funman@videolan.org>
	<20111207165822.GA20191@videolan.org> <20111207120924.446300d3@tok>
	<20111207171107.GA28811@videolan.org> <20111207122034.58b6d947@tok>
	<20111207172450.GF28811@videolan.org> <20111207123342.56ac8e4b@tok>
	<20111207174750.GC32159@videolan.org> <20111207132503.15da22fc@tok>
	<20111207183640.GA11126@videolan.org>
Message-ID: <20111207134117.0aba6bf5@tok>

Le Wed, 7 Dec 2011 19:36:40 +0100,
Jean-Baptiste Kempf <jb at videolan.org> a ?crit :

> On Wed, Dec 07, 2011 at 01:25:03PM -0500, Rafa?l Carr? wrote :
> > Le Wed, 7 Dec 2011 18:47:50 +0100,
> > Jean-Baptiste Kempf <jb at videolan.org> a ?crit :
> > 
> > > On Wed, Dec 07, 2011 at 12:33:42PM -0500, Rafa?l Carr? wrote :
> > > > I checked MSDN documentation, the libav commit should have a link to it.
> > > 
> > > It doesn't.
> > 
> > Hm I might have forgotten it, or libav's mailman might have removed it.
> > 
> > In any case: http://msdn.microsoft.com/en-us/library/windows/desktop/bb205892(v=vs.85).aspx
> > 
> > LPDIRECT3DSURFACE9 needs d3d9.h
> 
> Sure, but what I am saying is that other things in dxva2api.h do need
> d3d9.h, like D3DFORMAT, D3DPOOL, so the include should be in dxva2api.h,
> not in the client code.

mingw64's dxva2api.h effectively includes d3d9.h

If the client code uses directly d3d9.h features it should also include
it directly.

Relying on headers including other ones can cause headache, as proved
by our custom dxva2api.h failure to do this.

-- 
Rafa?l Carr?


From finlayson at live555.com  Wed Dec  7 20:07:36 2011
From: finlayson at live555.com (Ross Finlayson)
Date: Wed, 7 Dec 2011 12:07:36 -0700
Subject: [vlc-devel] Video packet reordering
In-Reply-To: <201112072010.23659.remi@remlab.net>
References: <CADF-EHFB=9OuBC0O8vkroE5J=osEvC+6_1wAhQM1G4BAv+_wbw@mail.gmail.com>
	<201112072010.23659.remi@remlab.net>
Message-ID: <A690633E-006C-48CF-8327-B83D257BA8C7@live555.com>

The "LIVE555 Streaming Media" libraries - that VLC uses to receive RTP packets - automatically reorders incoming RTP packets if they arrive out-of-order, so, VLC's decoding/rendering code will never see out-of-order data.

The only exception to this is that if an out-of-order packet arrives later than a pre-specified threshold (for VLC, 200ms), then the out-of-order packet will be dropped (i.e. treated as if it had been lost).

So, as long as your network does not delay out-of-order packets more than 200ms, you should be OK.  (In any case, as I noted above, VLC's decoding/rendering code will never see out-of-order data.)

Ross Finlayson
Live Networks, Inc.
http://www.live555.com/

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111207/e3c69a40/attachment.html>

From fenrir at elivagar.org  Wed Dec  7 20:19:56 2011
From: fenrir at elivagar.org (Laurent Aimar)
Date: Wed, 7 Dec 2011 20:19:56 +0100
Subject: [vlc-devel] [vlc-commits] contrib: change dxva2api.h checksum
In-Reply-To: <20111207121021.12b430d4@tok>
References: <20111207170231.334EC144629@albiero.videolan.org>
	<20111207170423.GC20191@videolan.org> <20111207121021.12b430d4@tok>
Message-ID: <20111207191956.GA1043@elivagar.org>

On Wed, Dec 07, 2011 at 12:10:21PM -0500, Rafa?l Carr? wrote:
> Le Wed, 7 Dec 2011 18:04:23 +0100,
> Jean-Baptiste Kempf <jb at videolan.org> a ?crit :
> 
> > On Wed, Dec 07, 2011 at 06:02:31PM +0100, Rafa?l Carr? wrote :
> > > vlc | branch: master | Rafa?l Carr? <funman at videolan.org> | Wed Dec  7 12:01:18 2011 -0500| [8fd4111acb17d4d887f163badd25c92a9465a51e] | committer: Rafa?l Carr?
> > > 
> > > contrib: change dxva2api.h checksum
> > > 
> > > I added a workaround for mingw64
> > > 
> > 
> > What work-around?
> 
> #ifndef _COM_interface
> #define _COM_interface struct
> #endif
 From what I understand, the right fix would be to use "interface" instead of
"_COM_interface".

-- 
fenrir



From sebastien-devel at celeos.eu  Wed Dec  7 20:20:12 2011
From: sebastien-devel at celeos.eu (=?utf-8?b?U8OpYmFzdGllbg==?= Escudier)
Date: Wed, 07 Dec 2011 20:20:12 +0100
Subject: [vlc-devel] [PATCH] live555 : enable/disable tracks on the fly
In-Reply-To: <201112072002.59492.remi@remlab.net>
References: <1322730254.10961.3.camel@stim-desktop>
	<201112072002.59492.remi@remlab.net>
Message-ID: <20111207202012.10885iuybnrthhmo@ssl0.ovh.net>

R?mi Denis-Courmont <remi at remlab.net> a ?crit?:

> Le jeudi 1 d?cembre 2011 11:04:14 S?bastien Escudier, vous avez ?crit :
>> 3 patches attached.
>> I did not test multiple audio or multiple video tracks case because I
>> don't have rtsp servers like that.
>
> All three patches failed to merge.

I already commited them.




From funman at videolan.org  Wed Dec  7 20:28:58 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Wed, 7 Dec 2011 14:28:58 -0500
Subject: [vlc-devel] [PATCH] contrib: remove directx and directshow
 headers
In-Reply-To: <20111207165822.GA20191@videolan.org>
References: <1323276299-13924-1-git-send-email-funman@videolan.org>
	<20111207165822.GA20191@videolan.org>
Message-ID: <20111207142858.3e81d5e5@tok>

Le Wed, 7 Dec 2011 17:58:22 +0100,
Jean-Baptiste Kempf <jb at videolan.org> a ?crit :

> On Wed, Dec 07, 2011 at 11:44:59AM -0500, Rafa?l Carr? wrote :
> > As part of the Windows SDK, they must be provided by the dev environment
> > mingw-64, for example provide them
> > 
> > bug reports should be addressed to mingw32 and/or distributions
> 
> Rejected. Mingw headers are buggy.

I just remember that Debian ships an antique version anyway and we have
to use http://people.videolan.org/~jb/debian/

Perhaps we could put dshow/directx headers in this .deb?

Of course that'd break other distros though

-- 
Rafa?l Carr?


From lukem.pdr at googlemail.com  Wed Dec  7 21:08:19 2011
From: lukem.pdr at googlemail.com (Luke McQuade)
Date: Wed, 07 Dec 2011 20:08:19 +0000
Subject: [vlc-devel] Media Library plans?
In-Reply-To: <20111207142858.3e81d5e5@tok>
References: <1323276299-13924-1-git-send-email-funman@videolan.org>
	<20111207165822.GA20191@videolan.org> <20111207142858.3e81d5e5@tok>
Message-ID: <4EDFC7B3.5000602@gmail.com>

Hi,
Newb here. I was wondering what the status/plans for the Media Library 
are? e.g. how much, if any time is being spent on it?
Thanks,
Luke McQuade



From jb at videolan.org  Wed Dec  7 21:36:19 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Wed, 7 Dec 2011 21:36:19 +0100
Subject: [vlc-devel] [vlc-commits] contrib: change dxva2api.h checksum
In-Reply-To: <20111207191956.GA1043@elivagar.org>
References: <20111207170231.334EC144629@albiero.videolan.org>
	<20111207170423.GC20191@videolan.org> <20111207121021.12b430d4@tok>
	<20111207191956.GA1043@elivagar.org>
Message-ID: <20111207203619.GA3905@videolan.org>

On Wed, Dec 07, 2011 at 08:19:56PM +0100, Laurent Aimar wrote :
> On Wed, Dec 07, 2011 at 12:10:21PM -0500, Rafa?l Carr? wrote:
> > Le Wed, 7 Dec 2011 18:04:23 +0100,
> > Jean-Baptiste Kempf <jb at videolan.org> a ?crit :
> > 
> > > On Wed, Dec 07, 2011 at 06:02:31PM +0100, Rafa?l Carr? wrote :
> > > > vlc | branch: master | Rafa?l Carr? <funman at videolan.org> | Wed Dec  7 12:01:18 2011 -0500| [8fd4111acb17d4d887f163badd25c92a9465a51e] | committer: Rafa?l Carr?
> > > > 
> > > > contrib: change dxva2api.h checksum
> > > > 
> > > > I added a workaround for mingw64
> > > > 
> > > 
> > > What work-around?
> > 
> > #ifndef _COM_interface
> > #define _COM_interface struct
> > #endif
>  From what I understand, the right fix would be to use "interface" instead of
> "_COM_interface".

yes, that's what I said on IRC. DECLARE_INTERFACE macro might be of help
here.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Thu Dec  8 02:27:51 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Thu, 8 Dec 2011 02:27:51 +0100
Subject: [vlc-devel] Video XP bug on VLC 1.2
Message-ID: <20111208012751.GA23860@videolan.org>

So, tonight, I took quite some time to understand what was
happening on the #5527 bug, meaning no video or very few video, with a
very strong deadlock on Windows XP.

As you might know, this was one of biggest issue of VLC 1.2.x on
Windows.

So, tonight, I installed XP, and that took a few hours (3 or so, even
starting from SP3 MSDN) to have a working setup (using the new HDD
bought for the white Macbook) on my desktop machine.

Audio was crashing quite a lot, but with --no-audio, reproduction of the
bug was quite easy to get.

Trying all the video output changed barely anything...

Then, I remembered the issues we were having with the webplugin, and I
deactivated the -mt of libavcodec... Forcing the thread-count to 1 fixed
it, indeed.

Therefore, updating to latest libavcodec (after the win32 thread commit)
+ a full contribs recompile, fixed it partially for me:
 - Usually, it works. Sometimes, it doesn't. Sometimes in the same
   instance, it works, then not.
 - Seeking is a way to kill it quite often.
 - Deactivating audio helps, but doesn't make it working all the time.
 - Forcing 2 threads (on a QuadCore) makes it less bad, but not
   perfectly.

Here is the report I could make.

Patches and cool ideas are welcome.
I can help testing now.

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From funman at videolan.org  Thu Dec  8 02:38:51 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Wed, 7 Dec 2011 20:38:51 -0500
Subject: [vlc-devel] Video XP bug on VLC 1.2
In-Reply-To: <20111208012751.GA23860@videolan.org>
References: <20111208012751.GA23860@videolan.org>
Message-ID: <20111207203851.24ac581c@tok>

Le Thu, 8 Dec 2011 02:27:51 +0100,
Jean-Baptiste Kempf <jb at videolan.org> a ?crit :

> So, tonight, I took quite some time to understand what was
> happening on the #5527 bug, meaning no video or very few video, with a
> very strong deadlock on Windows XP.
> 
> As you might know, this was one of biggest issue of VLC 1.2.x on
> Windows.
> 
> So, tonight, I installed XP, and that took a few hours (3 or so, even
> starting from SP3 MSDN) to have a working setup (using the new HDD
> bought for the white Macbook) on my desktop machine.
> 
> Audio was crashing quite a lot, but with --no-audio, reproduction of the
> bug was quite easy to get.
> 
> Trying all the video output changed barely anything...
> 
> Then, I remembered the issues we were having with the webplugin, and I
> deactivated the -mt of libavcodec... Forcing the thread-count to 1 fixed
> it, indeed.
> 
> Therefore, updating to latest libavcodec (after the win32 thread commit)
> + a full contribs recompile, fixed it partially for me:
>  - Usually, it works. Sometimes, it doesn't. Sometimes in the same
>    instance, it works, then not.
>  - Seeking is a way to kill it quite often.
>  - Deactivating audio helps, but doesn't make it working all the time.
>  - Forcing 2 threads (on a QuadCore) makes it less bad, but not
>    perfectly.
> 
> Here is the report I could make.
> 
> Patches and cool ideas are welcome.
> I can help testing now.

do you know where it crash?

if you have a coredump + debug symbols we can analyze it with the
script on jones, especially since aslr isn't enabled on XP, right?

-- 
Rafa?l Carr?


From jb at videolan.org  Thu Dec  8 02:42:41 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Thu, 8 Dec 2011 02:42:41 +0100
Subject: [vlc-devel] Video XP bug on VLC 1.2
In-Reply-To: <20111207203851.24ac581c@tok>
References: <20111208012751.GA23860@videolan.org> <20111207203851.24ac581c@tok>
Message-ID: <20111208014241.GB23860@videolan.org>

On Wed, Dec 07, 2011 at 08:38:51PM -0500, Rafa?l Carr? wrote :
> do you know where it crash?

It does not crash, that is the issue.

> if you have a coredump + debug symbols we can analyze it with the
> script on jones, especially since aslr isn't enabled on XP, right?

Sure, if I could crash it. Deadlocking is a bit harder, though...

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Thu Dec  8 03:42:28 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Thu, 8 Dec 2011 03:42:28 +0100
Subject: [vlc-devel] Video XP bug on VLC 1.2
In-Reply-To: <20111208012751.GA23860@videolan.org>
References: <20111208012751.GA23860@videolan.org>
Message-ID: <20111208024228.GA9259@videolan.org>

On Thu, Dec 08, 2011 at 02:27:51AM +0100, Jean-Baptiste Kempf wrote :
> Patches and cool ideas are welcome.

One idea was to use winpthread from mingw-w64, which isn't the crap that
the old win32pthread (GC2) was.

After patching it quite heavily, I managed to compile it, recompile
libavcodec against it, and recompile the whole mess and re-run it. (took
the time between the 2 mails).

It does work, more or less, which would mean that the issue is indeed in
the libavcodec win32 thread wrapper.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From funman at videolan.org  Thu Dec  8 04:53:40 2011
From: funman at videolan.org (=?UTF-8?q?Rafa=C3=ABl=20Carr=C3=A9?=)
Date: Wed,  7 Dec 2011 22:53:40 -0500
Subject: [vlc-devel] [PATCH 2/2] contrib: remove unneeded gecko
In-Reply-To: <1323316420-2207-1-git-send-email-funman@videolan.org>
References: <1323316420-2207-1-git-send-email-funman@videolan.org>
Message-ID: <1323316420-2207-2-git-send-email-funman@videolan.org>

---
 contrib/src/gecko/rules.mak |   23 -----------------------
 1 files changed, 0 insertions(+), 23 deletions(-)
 delete mode 100644 contrib/src/gecko/rules.mak

diff --git a/contrib/src/gecko/rules.mak b/contrib/src/gecko/rules.mak
deleted file mode 100644
index 50e4fa5..0000000
--- a/contrib/src/gecko/rules.mak
+++ /dev/null
@@ -1,23 +0,0 @@
-# gecko (mozilla plugin headers)
-
-PKGS += gecko
-
-NPAPI_HEADERS_SVN_URL := http://npapi-sdk.googlecode.com/svn/trunk/headers
-NPAPI_HEADERS_SVN_REV := HEAD # revision number, or just HEAD for the latest
-
-$(TARBALLS)/gecko-svn.tar.xz:
-	rm -Rf gecko-svn
-	$(SVN) export $(NPAPI_HEADERS_SVN_URL) gecko-svn -r $(NPAPI_HEADERS_SVN_REV)
-	tar cvJ gecko-svn > $@
-
-.sum-gecko: gecko-svn.tar.xz
-	$(warning Integrity check skipped.)
-	touch $@
-
-gecko: gecko-svn.tar.xz .sum-gecko
-	$(UNPACK)
-	$(MOVE)
-
-.gecko: gecko
-	cp $</*.h "$(PREFIX)/include"
-	touch $@
-- 
1.7.7.3


From funman at videolan.org  Thu Dec  8 04:53:39 2011
From: funman at videolan.org (=?UTF-8?q?Rafa=C3=ABl=20Carr=C3=A9?=)
Date: Wed,  7 Dec 2011 22:53:39 -0500
Subject: [vlc-devel] [PATCH 1/2] npapi: fetch gecko headers required for
	windows
Message-ID: <1323316420-2207-1-git-send-email-funman@videolan.org>

---
 extras/package/npapi.am |    8 ++++++--
 1 files changed, 6 insertions(+), 2 deletions(-)

diff --git a/extras/package/npapi.am b/extras/package/npapi.am
index 759082f..8bde4c4 100644
--- a/extras/package/npapi.am
+++ b/extras/package/npapi.am
@@ -2,6 +2,10 @@
 
 build-npapi: stamp-npapi-build
 
+stamp-gecko-fetch:
+	svn export http://npapi-sdk.googlecode.com/svn/trunk/headers gecko -r HEAD
+	touch $@
+
 stamp-npapi-fetch:
 	rm -Rf npapi-vlc
 	git clone git://git.videolan.org/npapi-vlc.git npapi-vlc
@@ -19,7 +23,7 @@ else
 CONF_ACTIVEX =
 endif
 
-stamp-npapi-configure: stamp-npapi-autogen
+stamp-npapi-configure: stamp-npapi-autogen stamp-gecko-fetch
 	cd npapi-vlc && \
 	$(SHELL) ./configure \
 		--prefix=/ \
@@ -29,7 +33,7 @@ stamp-npapi-configure: stamp-npapi-autogen
 		CC="$(CC)" CFLAGS="$(CFLAGS)" \
 		CXX="$(CXX)" CXXFLAGS="$(CXXFLAGS)" \
 		LD="$(LD)" LDFLAGS="$(LDFLAGS)" \
-		MOZILLA_CFLAGS="-I$(CONTRIB_DIR)/gecko-sdk/include" \
+		MOZILLA_CFLAGS="-I$(abs_builddir)/gecko" \
 		SHELL="$(SHELL)" \
 		PKG_CONFIG_LIBDIR="../$(srcdir)/extras/contrib/lib/pkgconfig:../lib"
 	touch $@
-- 
1.7.7.3


From beauze.h at gmail.com  Thu Dec  8 10:25:05 2011
From: beauze.h at gmail.com (=?UTF-8?Q?Hugo_Beauz=C3=A9e=2DLuyssen?=)
Date: Thu, 8 Dec 2011 10:25:05 +0100
Subject: [vlc-devel] Tap, tap, is this thing "on"
In-Reply-To: <CADF-EHGJiH9Do7EUPby_sHE7siq-pTBwwvaHJYpu0x7gSZO7FQ@mail.gmail.com>
References: <CADF-EHGJiH9Do7EUPby_sHE7siq-pTBwwvaHJYpu0x7gSZO7FQ@mail.gmail.com>
Message-ID: <CAK-C75Qm4cxaW8CP=4ZEs-MgaAwkb+VH0k71Rw+ATV1bX7pZ0g@mail.gmail.com>

On Wed, Dec 7, 2011 at 6:12 PM, John Zavgren <john at zavgren.com> wrote:
> Greetings:
> I've been trying to post a question to vlc-devel, but my attempts seem
> to be failing silently.
>
>
>

Hi,

This one hasn't.
If you were sending big files before, this is why it may have failed.

Regards,

-- 
Hugo Beauz?e-Luyssen


From komh78 at gmail.com  Thu Dec  8 12:48:20 2011
From: komh78 at gmail.com (KO Myung-Hun)
Date: Thu,  8 Dec 2011 20:48:20 +0900
Subject: [vlc-devel] [PATCH 7/7] Fix compilation error due to pthread on OS/2
Message-ID: <1323344900-59383-1-git-send-email-komh@chollian.net>

---
 bin/vlc.c |    9 ++++++++-
 1 files changed, 8 insertions(+), 1 deletions(-)

diff --git a/bin/vlc.c b/bin/vlc.c
index 6969a1b..4d20f89 100644
--- a/bin/vlc.c
+++ b/bin/vlc.c
@@ -35,13 +35,20 @@
 #include <stdbool.h>
 #include <locale.h>
 #include <signal.h>
-#include <pthread.h>
+#ifdef HAVE_PTHREAD_H
+# include <pthread.h>
+#endif
 #include <unistd.h>
 
 #ifdef __APPLE__
 #include <string.h>
 #endif
 
+#ifdef __OS2__
+# define pthread_t      int
+# define pthread_self() _gettid()
+#endif
+
 
 /* Explicit HACK */
 extern void LocaleFree (const char *);
-- 
1.7.3.2



From komh78 at gmail.com  Thu Dec  8 13:32:48 2011
From: komh78 at gmail.com (KO Myung-Hun)
Date: Thu,  8 Dec 2011 21:32:48 +0900
Subject: [vlc-devel] [PATCH] Implement vlc_rand_bytes() for OS/2
Message-ID: <1323347568-59805-1-git-send-email-komh@chollian.net>

---
 src/misc/rand.c |   97 +++++++++++++++++++++++++++++++------------------------
 1 files changed, 55 insertions(+), 42 deletions(-)

diff --git a/src/misc/rand.c b/src/misc/rand.c
index 3b84de6..038f79e 100644
--- a/src/misc/rand.c
+++ b/src/misc/rand.c
@@ -26,7 +26,61 @@
 #include <vlc_common.h>
 #include <vlc_rand.h>
 
-#ifndef WIN32
+#ifdef WIN32
+#include <wincrypt.h>
+
+void vlc_rand_bytes (void *buf, size_t len)
+{
+    HCRYPTPROV hProv;
+    size_t count = len;
+    uint8_t *p_buf = (uint8_t *)buf;
+
+    /* fill buffer with pseudo-random data */
+    while (count > 0)
+    {
+        unsigned int val;
+        val = rand();
+        if (count < sizeof (val))
+        {
+            memcpy (p_buf, &val, count);
+            break;
+        }
+
+        memcpy (p_buf, &val, sizeof (val));
+        count -= sizeof (val);
+        p_buf += sizeof (val);
+    }
+
+    /* acquire default encryption context */
+    if( CryptAcquireContext(
+        &hProv,            // Variable to hold returned handle.
+        NULL,              // Use default key container.
+        MS_DEF_PROV,       // Use default CSP.
+        PROV_RSA_FULL,     // Type of provider to acquire.
+        0) )
+    {
+        /* fill buffer with pseudo-random data, intial buffer content
+           is used as auxillary random seed */
+        CryptGenRandom(hProv, len, buf);
+        CryptReleaseContext(hProv, 0);
+    }
+}
+
+#elif defined (__OS2__)
+void vlc_rand_bytes (void *buf, size_t len)
+{
+    QWORD qwTime;
+    uint8_t *p_buf = (uint8_t *)buf;
+
+    while (len > 0)
+    {
+        DosTmrQueryTime( &qwTime );
+
+        *p_buf++ = ( qwTime.ulLo * rand()) & 0xFF;
+        len--;
+    }
+}
+#else
 #include <stdint.h>
 #include <string.h>
 #include <stdlib.h>
@@ -121,47 +175,6 @@ void vlc_rand_bytes (void *buf, size_t len)
         buf = ((uint8_t *)buf) + 16;
     }
 }
-
-#else /* WIN32 */
-
-#include <wincrypt.h>
-
-void vlc_rand_bytes (void *buf, size_t len)
-{
-    HCRYPTPROV hProv;
-    size_t count = len;
-    uint8_t *p_buf = (uint8_t *)buf;
-
-    /* fill buffer with pseudo-random data */
-    while (count > 0)
-    {
-        unsigned int val;
-        val = rand();
-        if (count < sizeof (val))
-        {
-            memcpy (p_buf, &val, count);
-            break;
-        }
- 
-        memcpy (p_buf, &val, sizeof (val));
-        count -= sizeof (val);
-        p_buf += sizeof (val);
-    }
-
-    /* acquire default encryption context */
-    if( CryptAcquireContext(
-        &hProv,            // Variable to hold returned handle.
-        NULL,              // Use default key container.
-        MS_DEF_PROV,       // Use default CSP.
-        PROV_RSA_FULL,     // Type of provider to acquire.
-        0) )
-    {
-        /* fill buffer with pseudo-random data, intial buffer content
-           is used as auxillary random seed */
-        CryptGenRandom(hProv, len, buf);
-        CryptReleaseContext(hProv, 0);
-    }
-}
 #endif
 
 static struct
-- 
1.7.3.2



From david.glaude at gmail.com  Thu Dec  8 14:45:23 2011
From: david.glaude at gmail.com (David Glaude)
Date: Thu, 8 Dec 2011 14:45:23 +0100
Subject: [vlc-devel] [PATCH] live555 : enable/disable tracks on the fly
In-Reply-To: <1322730254.10961.3.camel@stim-desktop>
References: <1322730254.10961.3.camel@stim-desktop>
Message-ID: <CAEDB7osxirbH5mHGRSwcDrNVD4aZ1q=8ujiF8ONAKc7GMOO7Yw@mail.gmail.com>

2011/12/1 S?bastien Escudier <sebastien-devel at celeos.eu>

> I did not test multiple audio or multiple video tracks case because I
> don't have rtsp servers like that.
>

Hi,

I do not fully understand the patch and how it is related, but I am curious
about this comment because I am concern with everything that is related to
multiple alternative audio track (or video quality for that matter).

Even if you do not have one rtsp server that support that...

Do you know if there is any RTSP server that support streaming
multi-audio content = not all simultaniously but only those requested (I am
thinking in particular to something supporting 3GPP v6 and the
specification in MP4 file of alternate track)?

Do you know if there is any RTSP player (VLC excluded) that support sending
request to stop sending one language and start sending an alternat language
(and has a UI or javascript control to let the user/web developper pick a
language).

All I know si that of this negotiation and language change take place
between a Windows Media Player and a Windows Media Streaming Server... both
that can talk something very RTSP like (rtspu and rtspt).

Your input is welkom.

David Glaude
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111208/dccff6bc/attachment.html>

From sebastien-devel at celeos.eu  Thu Dec  8 14:49:39 2011
From: sebastien-devel at celeos.eu (=?ISO-8859-1?Q?S=E9bastien?= Escudier)
Date: Thu, 08 Dec 2011 14:49:39 +0100
Subject: [vlc-devel] [PATCH] live555 : enable/disable tracks on the fly
In-Reply-To: <CAEDB7osxirbH5mHGRSwcDrNVD4aZ1q=8ujiF8ONAKc7GMOO7Yw@mail.gmail.com>
References: <1322730254.10961.3.camel@stim-desktop>
	<CAEDB7osxirbH5mHGRSwcDrNVD4aZ1q=8ujiF8ONAKc7GMOO7Yw@mail.gmail.com>
Message-ID: <1323352179.1253.11.camel@stim-desktop>

On Thu, 2011-12-08 at 14:45 +0100, David Glaude wrote:
> Do you know if there is any RTSP server that support streaming
> multi-audio content = not all simultaniously but only those requested
> (I am thinking in particular to something supporting 3GPP v6 and the
> specification in MP4 file of alternate track)?

No, but I would be glad to test one.




From jb at videolan.org  Thu Dec  8 14:57:53 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Thu, 8 Dec 2011 14:57:53 +0100
Subject: [vlc-devel] [PATCH] live555 : enable/disable tracks on the fly
In-Reply-To: <CAEDB7osxirbH5mHGRSwcDrNVD4aZ1q=8ujiF8ONAKc7GMOO7Yw@mail.gmail.com>
References: <1322730254.10961.3.camel@stim-desktop>
	<CAEDB7osxirbH5mHGRSwcDrNVD4aZ1q=8ujiF8ONAKc7GMOO7Yw@mail.gmail.com>
Message-ID: <20111208135753.GA9371@videolan.org>

On Thu, Dec 08, 2011 at 02:45:23PM +0100, David Glaude wrote :
> Do you know if there is any RTSP player (VLC excluded) that support sending
> request to stop sending one language and start sending an alternat language
> (and has a UI or javascript control to let the user/web developper pick a
> language).

Did you ask Anevia ?

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From remi at remlab.net  Thu Dec  8 15:04:51 2011
From: remi at remlab.net (=?UTF-8?Q?R=C3=A9mi_Denis-Courmont?=)
Date: Thu, 08 Dec 2011 15:04:51 +0100
Subject: [vlc-devel] [PATCH] live555 : enable/disable tracks on the fly
In-Reply-To: <CAEDB7osxirbH5mHGRSwcDrNVD4aZ1q=8ujiF8ONAKc7GMOO7Yw@mail.gmail.com>
References: <1322730254.10961.3.camel@stim-desktop>
	<CAEDB7osxirbH5mHGRSwcDrNVD4aZ1q=8ujiF8ONAKc7GMOO7Yw@mail.gmail.com>
Message-ID: <1d58d649f71adbaaf56ada43af21bf8f@chewa.net>

   Hello,

On Thu, 8 Dec 2011 14:45:23 +0100, David Glaude <david.glaude at gmail.com>
wrote:
> I do not fully understand the patch and how it is related, but I am
curious
> about this comment because I am concern with everything that is related
to
> multiple alternative audio track (or video quality for that matter).
> 
> Even if you do not have one rtsp server that support that...

The VLC RTSP server does (and always did).

> Do you know if there is any RTSP server that support streaming
> multi-audio content = not all simultaniously but only those requested (I
am
> thinking in particular to something supporting 3GPP v6 and the
> specification in MP4 file of alternate track)?

Do you know any RTSP server that would _not_ support track selection?

I mean, I do not think it is even possible, since the RTSP SETUP request
can only enable one track at a time.

-- 
R?mi Denis-Courmont
http://www.remlab.net/


From remi at remlab.net  Thu Dec  8 15:32:12 2011
From: remi at remlab.net (=?UTF-8?Q?R=C3=A9mi_Denis-Courmont?=)
Date: Thu, 08 Dec 2011 15:32:12 +0100
Subject: [vlc-devel] Video XP bug on VLC 1.2
In-Reply-To: <20111208012751.GA23860@videolan.org>
References: <20111208012751.GA23860@videolan.org>
Message-ID: <8e0aae98a072b60e268a33b6c13b4f46@chewa.net>

On Thu, 8 Dec 2011 02:27:51 +0100, Jean-Baptiste Kempf <jb at videolan.org>
wrote:
> Patches and cool ideas are welcome.

Pragmatically, the best option consists of removing -mt support on
Windows. Moreover, the would-be bug in libavcodec thread wrappers on
Windows XP should affect other Windows versions too. Maybe it is just not
as frequent due to pure timing luck.

Alternatively, do not release 1.2.0 on Windows and until contribs gets
fixed in 1.2.x for some strictly positive value of x. But that has two
major problems: Firstly other bugs will remain unnoticed, since the user
base will be shrunk. Secondly, security/critical updates for Windows would
be blocked since VLC 1.1.x is now dead.

I would not wait for UID 65534 to provide a fix. That would be insane.

-- 
R?mi Denis-Courmont
http://www.remlab.net/


From jb at videolan.org  Thu Dec  8 15:35:08 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Thu, 8 Dec 2011 15:35:08 +0100
Subject: [vlc-devel] Video XP bug on VLC 1.2
In-Reply-To: <8e0aae98a072b60e268a33b6c13b4f46@chewa.net>
References: <20111208012751.GA23860@videolan.org>
	<8e0aae98a072b60e268a33b6c13b4f46@chewa.net>
Message-ID: <20111208143508.GA16596@videolan.org>

On Thu, Dec 08, 2011 at 03:32:12PM +0100, R?mi Denis-Courmont wrote :
> On Thu, 8 Dec 2011 02:27:51 +0100, Jean-Baptiste Kempf <jb at videolan.org>
> wrote:
> > Patches and cool ideas are welcome.
> 
> Pragmatically, the best option consists of removing -mt support on
> Windows. Moreover, the would-be bug in libavcodec thread wrappers on
> Windows XP should affect other Windows versions too. Maybe it is just not
> as frequent due to pure timing luck.

What I could suggest too is to introduce the winpthread in contribs,
patch it as necessary to build with mingw/debian mingw/msys as I did,
later this night, and use that until libavcodec-mt is fixed upstream wrt
win32threads.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From david.glaude at gmail.com  Thu Dec  8 16:06:47 2011
From: david.glaude at gmail.com (David Glaude)
Date: Thu, 8 Dec 2011 16:06:47 +0100
Subject: [vlc-devel] [PATCH] live555 : enable/disable tracks on the fly
In-Reply-To: <1d58d649f71adbaaf56ada43af21bf8f@chewa.net>
References: <1322730254.10961.3.camel@stim-desktop>
	<CAEDB7osxirbH5mHGRSwcDrNVD4aZ1q=8ujiF8ONAKc7GMOO7Yw@mail.gmail.com>
	<1d58d649f71adbaaf56ada43af21bf8f@chewa.net>
Message-ID: <CAEDB7ovB2RcfvtatuSyzcmDyiZJ4JSvtRP4RPiOm+75DR1jBbQ@mail.gmail.com>

Le 8 d?cembre 2011 15:04, R?mi Denis-Courmont <remi at remlab.net> a ?crit :

>   Hello,
>
> On Thu, 8 Dec 2011 14:45:23 +0100, David Glaude <david.glaude at gmail.com>
> wrote:
> > Even if you do not have one rtsp server that support that...
>
> The VLC RTSP server does (and always did).
>

I need to check this (see below.


> > Do you know if there is any RTSP server that support streaming
> > multi-audio content = not all simultaniously but only those requested (I
> am
> > thinking in particular to something supporting 3GPP v6 and the
> > specification in MP4 file of alternate track)?
>
> Do you know any RTSP server that would _not_ support track selection?
>
> I mean, I do not think it is even possible, since the RTSP SETUP request
> can only enable one track at a time.


Selecting track in the SETUP is one thing...
But for the server to present all the available alternative as alternative
in the SDP;
Or for the client to understand those alternative and let the client be
offered the choice is another thing.

While the standard describe all this, I fail to find product that implement
all this.

For more info on this, 'tsel' box you can check a related ticket here
http://trac.videolan.org/vlc/ticket/3970

Google found this PDF version of the 3GPPv6 documentation where it is
explain in 5.3.3.3 and after (but also with a sample in Annex A exemple 4,
5, 6 and 7) in this http://www.quintillion.co.jp/3GPP/Specs/26234-650.pdf

Thanks anyway.

David Glaude
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111208/80f15c49/attachment.html>

From funman at videolan.org  Thu Dec  8 17:15:35 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Thu, 8 Dec 2011 11:15:35 -0500
Subject: [vlc-devel] [PATCH 7/7] Fix compilation error due to pthread on
 OS/2
In-Reply-To: <1323344900-59383-1-git-send-email-komh@chollian.net>
References: <1323344900-59383-1-git-send-email-komh@chollian.net>
Message-ID: <20111208111535.6f924ef7@tok>

Le Thu,  8 Dec 2011 20:48:20 +0900,
KO Myung-Hun <komh78 at gmail.com> a ?crit :

> ---
>  bin/vlc.c |    9 ++++++++-
>  1 files changed, 8 insertions(+), 1 deletions(-)

Shouldn't you use winvlc.c on OS/2 ?

-- 
Rafa?l Carr?


From fcvlcdev at free.fr  Thu Dec  8 18:07:53 2011
From: fcvlcdev at free.fr (Francois Cartegnie)
Date: Thu,  8 Dec 2011 18:07:53 +0100
Subject: [vlc-devel] [PATCH] add dvb channels list .conf as filetype.
Message-ID: <1323364073-29242-1-git-send-email-fcvlcdev@free.fr>

(see demux/playlist/dvb. mplayer's std being channels.conf)
---
 include/vlc_interface.h |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/include/vlc_interface.h b/include/vlc_interface.h
index 3157373..76a57f5 100644
--- a/include/vlc_interface.h
+++ b/include/vlc_interface.h
@@ -227,7 +227,7 @@ typedef enum vlc_dialog {
                          "*.ogg;*.ogm;*.ogv;*.ogx;*.ps;" \
                          "*.rec;*.rm;*.rmvb;*.tod;*.ts;*.tts;*.vob;*.vro;*.webm;*.wmv"
 
-#define EXTENSIONS_PLAYLIST "*.asx;*.b4s;*.cue;*.ifo;*.m3u;*.m3u8;*.pls;*.ram;*.rar;*.sdp;*.vlc;*.xspf;*.zip"
+#define EXTENSIONS_PLAYLIST "*.asx;*.b4s;*.cue;*.ifo;*.m3u;*.m3u8;*.pls;*.ram;*.rar;*.sdp;*.vlc;*.xspf;*.zip;*.conf"
 
 #define EXTENSIONS_MEDIA EXTENSIONS_VIDEO ";" EXTENSIONS_AUDIO ";" \
                           EXTENSIONS_PLAYLIST
-- 
1.7.3.4



From remi at remlab.net  Thu Dec  8 18:46:45 2011
From: remi at remlab.net (=?utf-8?q?R=C3=A9mi?= Denis-Courmont)
Date: Thu, 8 Dec 2011 19:46:45 +0200
Subject: [vlc-devel] [PATCH 7/7] Fix compilation error due to pthread on
	OS/2
In-Reply-To: <20111208111535.6f924ef7@tok>
References: <1323344900-59383-1-git-send-email-komh@chollian.net>
	<20111208111535.6f924ef7@tok>
Message-ID: <201112081946.46142.remi@remlab.net>

Le jeudi 8 d?cembre 2011 18:15:35 Rafa?l Carr?, vous avez ?crit :
> Le Thu,  8 Dec 2011 20:48:20 +0900,
> 
> KO Myung-Hun <komh78 at gmail.com> a ?crit :
> > ---
> > 
> >  bin/vlc.c |    9 ++++++++-
> >  1 files changed, 8 insertions(+), 1 deletions(-)
> 
> Shouldn't you use winvlc.c on OS/2 ?

OS/2 does not use WinMain(). It's a problem that OS/2 is like DOS/Windows in 
some respects, like Unix in some other, and often like a very old system.


-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From remi at remlab.net  Thu Dec  8 18:51:41 2011
From: remi at remlab.net (=?iso-8859-1?q?R=E9mi?= Denis-Courmont)
Date: Thu, 8 Dec 2011 19:51:41 +0200
Subject: [vlc-devel] [PATCH] add dvb channels list .conf as filetype.
In-Reply-To: <1323364073-29242-1-git-send-email-fcvlcdev@free.fr>
References: <1323364073-29242-1-git-send-email-fcvlcdev@free.fr>
Message-ID: <201112081951.41863.remi@remlab.net>

Le jeudi 8 d?cembre 2011 19:07:53 Francois Cartegnie, vous avez ?crit :
> (see demux/playlist/dvb. mplayer's std being channels.conf)

I would think this comes from VDR, not mplayer.

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From finlayson at live555.com  Thu Dec  8 20:16:20 2011
From: finlayson at live555.com (Ross Finlayson)
Date: Thu, 8 Dec 2011 12:16:20 -0700
Subject: [vlc-devel] [PATCH] live555 : enable/disable tracks on the fly
In-Reply-To: <CAEDB7ovB2RcfvtatuSyzcmDyiZJ4JSvtRP4RPiOm+75DR1jBbQ@mail.gmail.com>
References: <1322730254.10961.3.camel@stim-desktop>
	<CAEDB7osxirbH5mHGRSwcDrNVD4aZ1q=8ujiF8ONAKc7GMOO7Yw@mail.gmail.com>
	<1d58d649f71adbaaf56ada43af21bf8f@chewa.net>
	<CAEDB7ovB2RcfvtatuSyzcmDyiZJ4JSvtRP4RPiOm+75DR1jBbQ@mail.gmail.com>
Message-ID: <4D9E105C-95F8-4971-B315-FC2069CF0AC5@live555.com>

> Do you know any RTSP server that would _not_ support track selection?
> 
> I mean, I do not think it is even possible, since the RTSP SETUP request
> can only enable one track at a time.
> 
> Selecting track in the SETUP is one thing... 
> But for the server to present all the available alternative as alternative in the SDP;

FWIW, the LIVE555 RTSP server implementation supports this; For each 'session' (stream), you (the server developer) can include as many different 'subsessions' (tracks) as you want (and, of course, the client can, in principle, select whichever one(s) it wants).  To see how to do this, look at the code for our "testOnDemandRTSPServer" demo application.


Ross Finlayson
Live Networks, Inc.
http://www.live555.com/

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111208/8d438cf1/attachment.html>

From fenrir at elivagar.org  Thu Dec  8 20:23:14 2011
From: fenrir at elivagar.org (Laurent Aimar)
Date: Thu, 8 Dec 2011 20:23:14 +0100
Subject: [vlc-devel] Video XP bug on VLC 1.2
In-Reply-To: <20111208143508.GA16596@videolan.org>
References: <20111208012751.GA23860@videolan.org>
	<8e0aae98a072b60e268a33b6c13b4f46@chewa.net>
	<20111208143508.GA16596@videolan.org>
Message-ID: <20111208192314.GA10151@elivagar.org>

On Thu, Dec 08, 2011 at 03:35:08PM +0100, Jean-Baptiste Kempf wrote:
> On Thu, Dec 08, 2011 at 03:32:12PM +0100, R?mi Denis-Courmont wrote :
> > On Thu, 8 Dec 2011 02:27:51 +0100, Jean-Baptiste Kempf <jb at videolan.org>
> > wrote:
> > > Patches and cool ideas are welcome.
> > 
> > Pragmatically, the best option consists of removing -mt support on
> > Windows. Moreover, the would-be bug in libavcodec thread wrappers on
> > Windows XP should affect other Windows versions too. Maybe it is just not
> > as frequent due to pure timing luck.
> 
> What I could suggest too is to introduce the winpthread in contribs,
> patch it as necessary to build with mingw/debian mingw/msys as I did,
> later this night, and use that until libavcodec-mt is fixed upstream wrt
> win32threads.

 Btw, I haven't updated ffmpeg in some time, but on my PC on linux, I also
experience deadlock (on slightly corrupted streams like H264 TS coming
from DVB-T) with ffmpeg-mt.
 Am I the only one?

Regards,

-- 
fenrir


From remi at remlab.net  Thu Dec  8 20:24:50 2011
From: remi at remlab.net (=?iso-8859-15?q?R=E9mi?= Denis-Courmont)
Date: Thu, 8 Dec 2011 21:24:50 +0200
Subject: [vlc-devel] [PATCH] live555 : enable/disable tracks on the fly
In-Reply-To: <CAEDB7ovB2RcfvtatuSyzcmDyiZJ4JSvtRP4RPiOm+75DR1jBbQ@mail.gmail.com>
References: <1322730254.10961.3.camel@stim-desktop>
	<1d58d649f71adbaaf56ada43af21bf8f@chewa.net>
	<CAEDB7ovB2RcfvtatuSyzcmDyiZJ4JSvtRP4RPiOm+75DR1jBbQ@mail.gmail.com>
Message-ID: <201112082124.51249.remi@remlab.net>

Le jeudi 8 d?cembre 2011 17:06:47 David Glaude, vous avez ?crit :
> Le 8 d?cembre 2011 15:04, R?mi Denis-Courmont <remi at remlab.net> a ?crit :
> >   Hello,
> > 
> > On Thu, 8 Dec 2011 14:45:23 +0100, David Glaude <david.glaude at gmail.com>
> > 
> > wrote:
> > > Even if you do not have one rtsp server that support that...
> > 
> > The VLC RTSP server does (and always did).
> 
> I need to check this (see below.

VLC outputs one ES per type by default, as this is the normal player 
behaviour. It outputs alls (supported) ES with the --sout-all command line 
option in concert with --sout '...'.

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From jb at videolan.org  Thu Dec  8 20:46:39 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Thu, 8 Dec 2011 20:46:39 +0100
Subject: [vlc-devel] Video XP bug on VLC 1.2
In-Reply-To: <20111208192314.GA10151@elivagar.org>
References: <20111208012751.GA23860@videolan.org>
	<8e0aae98a072b60e268a33b6c13b4f46@chewa.net>
	<20111208143508.GA16596@videolan.org>
	<20111208192314.GA10151@elivagar.org>
Message-ID: <20111208194639.GA29098@videolan.org>

On Thu, Dec 08, 2011 at 08:23:14PM +0100, Laurent Aimar wrote :
>  Btw, I haven't updated ffmpeg in some time, but on my PC on linux, I also
> experience deadlock (on slightly corrupted streams like H264 TS coming
> from DVB-T) with ffmpeg-mt.
>  Am I the only one?

I haven't seen that, so far. Samples?

Laurent, do you think I should add an option to deactivate ffmpeg-mt in
Simple Preferences?

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From remi at remlab.net  Thu Dec  8 20:51:15 2011
From: remi at remlab.net (=?iso-8859-1?q?R=E9mi?= Denis-Courmont)
Date: Thu, 8 Dec 2011 21:51:15 +0200
Subject: [vlc-devel] Video XP bug on VLC 1.2
In-Reply-To: <20111208194639.GA29098@videolan.org>
References: <20111208012751.GA23860@videolan.org>
	<20111208192314.GA10151@elivagar.org>
	<20111208194639.GA29098@videolan.org>
Message-ID: <201112082151.16000.remi@remlab.net>

Le jeudi 8 d?cembre 2011 21:46:39 Jean-Baptiste Kempf, vous avez ?crit :
> On Thu, Dec 08, 2011 at 08:23:14PM +0100, Laurent Aimar wrote :
> >  Btw, I haven't updated ffmpeg in some time, but on my PC on linux, I
> >  also
> > 
> > experience deadlock (on slightly corrupted streams like H264 TS coming
> > from DVB-T) with ffmpeg-mt.
> > 
> >  Am I the only one?
> 
> I haven't seen that, so far. Samples?
> 
> Laurent, do you think I should add an option to deactivate ffmpeg-mt in
> Simple Preferences?

If it's so bad, it should be an option to ENABLE it.

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From aballier at gentoo.org  Fri Dec  9 00:08:13 2011
From: aballier at gentoo.org (Alexis Ballier)
Date: Thu, 8 Dec 2011 20:08:13 -0300
Subject: [vlc-devel] [PATCH?] xosd and xlib init
Message-ID: <20111208200813.206d1142@gentoo.org>

Hi,

I got a bug report [1] where vlc is killed at startup by the
XInitThreads checks. It seems the culprit is xosd, the solution is to
call vlc_xlib_init() at xosd's module Open(), right ?

If this is right, patch is enclosed.

Regards,

Alexis.

[1] https://bugs.gentoo.org/show_bug.cgi?id=393439
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-xosd-call-vlc_xlib_init-first.patch
Type: text/x-patch
Size: 942 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111208/34542ef2/attachment.bin>

From jb at videolan.org  Fri Dec  9 00:23:39 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Fri, 9 Dec 2011 00:23:39 +0100
Subject: [vlc-devel] [PATCH] add dvb channels list .conf as filetype.
In-Reply-To: <1323364073-29242-1-git-send-email-fcvlcdev@free.fr>
References: <1323364073-29242-1-git-send-email-fcvlcdev@free.fr>
Message-ID: <20111208232339.GA21033@videolan.org>

On Thu, Dec 08, 2011 at 06:07:53PM +0100, Francois Cartegnie wrote :
> -#define EXTENSIONS_PLAYLIST "*.asx;*.b4s;*.cue;*.ifo;*.m3u;*.m3u8;*.pls;*.ram;*.rar;*.sdp;*.vlc;*.xspf;*.zip"
> +#define EXTENSIONS_PLAYLIST "*.asx;*.b4s;*.cue;*.ifo;*.m3u;*.m3u8;*.pls;*.ram;*.rar;*.sdp;*.vlc;*.xspf;*.zip;*.conf"

Seems ok for me.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Fri Dec  9 00:25:37 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Fri, 9 Dec 2011 00:25:37 +0100
Subject: [vlc-devel] [PATCH] Implement vlc_rand_bytes() for OS/2
In-Reply-To: <1323347568-59805-1-git-send-email-komh@chollian.net>
References: <1323347568-59805-1-git-send-email-komh@chollian.net>
Message-ID: <20111208232537.GB21033@videolan.org>

On Thu, Dec 08, 2011 at 09:32:48PM +0900, KO Myung-Hun wrote :
>  src/misc/rand.c |   97 +++++++++++++++++++++++++++++++------------------------
>  1 files changed, 55 insertions(+), 42 deletions(-)

Sorry to ask, but shouldn't vlc_rand_bytes be splitted in
os2/rand.c win32/rand.c and so on?

- 
Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Fri Dec  9 00:27:31 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Fri, 9 Dec 2011 00:27:31 +0100
Subject: [vlc-devel] [PATCH 1/2] npapi: fetch gecko headers required for
 windows
In-Reply-To: <1323316420-2207-1-git-send-email-funman@videolan.org>
References: <1323316420-2207-1-git-send-email-funman@videolan.org>
Message-ID: <20111208232731.GA21964@videolan.org>

Both seem fine.

I would think that this would be better in npapi-vlc.git, though.

On Wed, Dec 07, 2011 at 10:53:39PM -0500, Rafa?l Carr? wrote :
> ---
>  extras/package/npapi.am |    8 ++++++--
>  1 files changed, 6 insertions(+), 2 deletions(-)
> 
> diff --git a/extras/package/npapi.am b/extras/package/npapi.am
> index 759082f..8bde4c4 100644
> --- a/extras/package/npapi.am
> +++ b/extras/package/npapi.am
> @@ -2,6 +2,10 @@
>  
>  build-npapi: stamp-npapi-build
>  
> +stamp-gecko-fetch:
> +	svn export http://npapi-sdk.googlecode.com/svn/trunk/headers gecko -r HEAD
> +	touch $@
> +
>  stamp-npapi-fetch:
>  	rm -Rf npapi-vlc
>  	git clone git://git.videolan.org/npapi-vlc.git npapi-vlc
> @@ -19,7 +23,7 @@ else
>  CONF_ACTIVEX =
>  endif
>  
> -stamp-npapi-configure: stamp-npapi-autogen
> +stamp-npapi-configure: stamp-npapi-autogen stamp-gecko-fetch
>  	cd npapi-vlc && \
>  	$(SHELL) ./configure \
>  		--prefix=/ \
> @@ -29,7 +33,7 @@ stamp-npapi-configure: stamp-npapi-autogen
>  		CC="$(CC)" CFLAGS="$(CFLAGS)" \
>  		CXX="$(CXX)" CXXFLAGS="$(CXXFLAGS)" \
>  		LD="$(LD)" LDFLAGS="$(LDFLAGS)" \
> -		MOZILLA_CFLAGS="-I$(CONTRIB_DIR)/gecko-sdk/include" \
> +		MOZILLA_CFLAGS="-I$(abs_builddir)/gecko" \
>  		SHELL="$(SHELL)" \
>  		PKG_CONFIG_LIBDIR="../$(srcdir)/extras/contrib/lib/pkgconfig:../lib"
>  	touch $@
> -- 
> 1.7.7.3
> _______________________________________________
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
> http://mailman.videolan.org/listinfo/vlc-devel

-- 
Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From rogerdpack2 at gmail.com  Fri Dec  9 00:48:46 2011
From: rogerdpack2 at gmail.com (Roger Pack)
Date: Thu, 8 Dec 2011 16:48:46 -0700
Subject: [vlc-devel] Bounty: fix dvd menu highlighting
In-Reply-To: <CAL1QdWf0pBCiGYzw2kdadAx1R1KeXMmB1CfomXeRJ4vc=FYfuQ@mail.gmail.com>
References: <CAL1QdWeyzrGFwR37t29iQ3qz2BAhrVxw4fxv7nqQ0uFVhmWB_A@mail.gmail.com>
	<20111116201513.68d95796@tok>
	<CAL1QdWf0pBCiGYzw2kdadAx1R1KeXMmB1CfomXeRJ4vc=FYfuQ@mail.gmail.com>
Message-ID: <CAL1QdWdVyzaKOTwd2LcsfYMUD35Fr2To6gqgAb60JgQkrXC5MA@mail.gmail.com>

>>> $150 to fix DVD Menu highlighting, which is messed up at times.

Ok, after looking at it myself, I noticed the following:
problem is not present in VLC 0.9.9
problem is present with 1.0.0 (which, perhaps coincidentally, was the
first release to include libyuvp_plugin.dll)
problem is not present in 1.2.0-nightly in windows (though highlighted
sections are too small for some reason).
problem is present with source build on linux (version 1.3.0-git-143-g3ce16b3)

My current recommendation would be to just disable yuvp temporarily
until it's fixed, if that's an option.
Barring that, I couldn't tell any immediate problem with the code--it
seems to look right.  So the problem may be in the "display" module in
how it displays the YUVA image on the screen (works windows, fails
linux ["YUV SDL output"]).

Anyway I'm giving up on it for now, and raising the bounty to $200 if
someone else can figure it out.
Thank you.
-r


From funman at videolan.org  Fri Dec  9 01:00:33 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Thu, 8 Dec 2011 19:00:33 -0500
Subject: [vlc-devel] [PATCH 1/2] npapi: fetch gecko headers required for
 windows
In-Reply-To: <20111208232731.GA21964@videolan.org>
References: <1323316420-2207-1-git-send-email-funman@videolan.org>
	<20111208232731.GA21964@videolan.org>
Message-ID: <20111208190033.71a8365f@tok>

Le Fri, 9 Dec 2011 00:27:31 +0100,
Jean-Baptiste Kempf <jb at videolan.org> a ?crit :

> Both seem fine.

Pushed

> I would think that this would be better in npapi-vlc.git, though.

We could even include the headers in the repo since they represent a
stable* interface afaiu.

* more or less stable since some function now returns const while it
didn't in the past

-- 
Rafa?l Carr?


From remi at remlab.net  Fri Dec  9 16:15:30 2011
From: remi at remlab.net (=?iso-8859-15?q?R=E9mi?= Denis-Courmont)
Date: Fri, 9 Dec 2011 17:15:30 +0200
Subject: [vlc-devel] [PATCH?] xosd and xlib init
In-Reply-To: <20111208200813.206d1142@gentoo.org>
References: <20111208200813.206d1142@gentoo.org>
Message-ID: <201112091715.32782.remi@remlab.net>

Le vendredi 9 d?cembre 2011 01:08:13 Alexis Ballier, vous avez ?crit :
> Hi,
> 
> I got a bug report [1] where vlc is killed at startup by the
> XInitThreads checks. It seems the culprit is xosd, the solution is to
> call vlc_xlib_init() at xosd's module Open(), right ?

vlc_xlib_init() is one solution. The better solution involves porting xosd to 
XCB but that would be out of scope.

> If this is right, patch is enclosed.

I'll merged this. But does this correctly ensure that X11 development files 
are present?

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From aballier at gentoo.org  Fri Dec  9 16:27:17 2011
From: aballier at gentoo.org (Alexis Ballier)
Date: Fri, 9 Dec 2011 12:27:17 -0300
Subject: [vlc-devel] [PATCH?] xosd and xlib init
In-Reply-To: <201112091715.32782.remi@remlab.net>
References: <20111208200813.206d1142@gentoo.org>
	<201112091715.32782.remi@remlab.net>
Message-ID: <20111209122717.6308ee87@gentoo.org>

On Fri, 9 Dec 2011 17:15:30 +0200
"R?mi Denis-Courmont" <remi at remlab.net> wrote:

> > If this is right, patch is enclosed.
> 
> I'll merged this. But does this correctly ensure that X11 development
> files are present?
> 


hmm actually, no
they will be present because xosd requires libX11 but this is not
correct...

whats the correct/recommended way to do this ?
adding $X_CFLAGS to xosd CFLAGS, and ${X_LIBS} ${X_PRE_LIBS} -lX11 to
xosd LIBS ? (like eg, done for libva), or checking for no_x first and
disabling xosd / bailing out like done for pulseaudio ?

Alexis.


From jb at videolan.org  Fri Dec  9 16:28:11 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Fri,  9 Dec 2011 16:28:11 +0100
Subject: [vlc-devel] [PATCH] Detect AVX-enabled processors
Message-ID: <1323444491-10657-1-git-send-email-jb@videolan.org>

---
 configure.ac      |   22 ++++++++++++++++++++++
 include/vlc_cpu.h |   11 +++++++++++
 src/misc/cpu.c    |   15 +++++++++++++++
 3 files changed, 48 insertions(+), 0 deletions(-)

diff --git a/configure.ac b/configure.ac
index d0538e2..869b9d2 100644
--- a/configure.ac
+++ b/configure.ac
@@ -1336,6 +1336,28 @@ AS_IF([test "${ac_cv_3dnow_inline}" != "no"], [
 ])
 AM_CONDITIONAL([HAVE_3DNOW], [test "$have_3dnow" = "yes"])
 
+VLC_SAVE_FLAGS
+CFLAGS="${CFLAGS} -mavx"
+have_avx="no"
+AC_CACHE_CHECK([if $CC groks AVX inline assembly], [ac_cv_avx_inline], [
+  AC_COMPILE_IFELSE([AC_LANG_PROGRAM(,[[
+void *p;
+asm volatile("vzeroall"::"r"(p));
+]])
+  ], [
+    ac_cv_avx_inline=yes
+  ], [
+    ac_cv_avx_inline=no
+  ])
+])
+VLC_RESTORE_FLAGS
+AS_IF([test "${ac_cv_avx_inline}" != "no"], [
+  AC_DEFINE(CAN_COMPILE_AVX, 1,
+            [Define to 1 if AVX inline assembly is available.])
+  have_3dnow="yes"
+])
+AM_CONDITIONAL([HAVE_AVX], [test "$have_3dnow" = "yes"])
+
 
 AC_ARG_ENABLE(neon,
   [AS_HELP_STRING([--disable-neon],
diff --git a/include/vlc_cpu.h b/include/vlc_cpu.h
index 63cc978..a4dc66b 100644
--- a/include/vlc_cpu.h
+++ b/include/vlc_cpu.h
@@ -37,6 +37,7 @@
 #  define CPU_CAPABILITY_SSE4_1  (1<<10)
 #  define CPU_CAPABILITY_SSE4_2  (1<<11)
 #  define CPU_CAPABILITY_SSE4A   (1<<12)
+#  define CPU_CAPABILITY_AVX     (1<<13)
 
 # if defined (__MMX__)
 #  define VLC_MMX
@@ -54,6 +55,15 @@
 #  define VLC_SSE VLC_SSE_is_not_implemented_on_this_compiler
 # endif
 
+# if defined (__AVX__)
+#  define VLC_AVX
+# elif VLC_GCC_VERSION(4, 6)
+#  define VLC_AVX __attribute__ ((__target__ ("avx")))
+# else
+#  define VLC_AVX VLC_AVX_is_not_implemented_on_this_compiler
+# endif
+
+
 # else
 #  define CPU_CAPABILITY_MMX     (0)
 #  define CPU_CAPABILITY_3DNOW   (0)
@@ -65,6 +75,7 @@
 #  define CPU_CAPABILITY_SSE4_1  (0)
 #  define CPU_CAPABILITY_SSE4_2  (0)
 #  define CPU_CAPABILITY_SSE4A   (0)
+#  define CPU_CAPABILITY_AVX     (0)
 # endif
 
 # if defined (__ppc__) || defined (__ppc64__) || defined (__powerpc__)
diff --git a/src/misc/cpu.c b/src/misc/cpu.c
index 7c28b25..97826a9 100644
--- a/src/misc/cpu.c
+++ b/src/misc/cpu.c
@@ -125,6 +125,13 @@ VLC_MMX static void ThreeD_Now_test (void)
     asm volatile ("pfadd %%mm0,%%mm0\n" "femms\n" : : : "mm0");
 }
 #endif
+#if defined (CAN_COMPILE_AVX) && !defined (__AVX__)
+VLC_AVX static void AVX_test (void)
+{
+    asm volatile ("vzeroall\n" );
+}
+#endif
+
 
 #if defined (CAN_COMPILE_ALTIVEC)
 static void Altivec_text (void)
@@ -262,6 +269,13 @@ void vlc_CPU_init (void)
         i_capabilities |= CPU_CAPABILITY_SSE4_2;
 # endif
 
+# if defined (__AVX__)
+    i_capabilities |= CPU_CAPABILITY_AVX;
+# elif defined (CAN_COMPILE_AVX)
+    if ((i_ecx & (1 << 28)) && vlc_CPU_check ("AVX", AVX_test))
+        i_capabilities |= CPU_CAPABILITY_AVX;
+# endif
+
     /* test for additional capabilities */
     cpuid( 0x80000000 );
 
@@ -384,6 +398,7 @@ void vlc_CPU_dump (vlc_object_t *obj)
     PRINT_CAPABILITY(CPU_CAPABILITY_SSE4_1, "SSE4.1");
     PRINT_CAPABILITY(CPU_CAPABILITY_SSE4_2, "SSE4.2");
     PRINT_CAPABILITY(CPU_CAPABILITY_SSE4A,  "SSE4A");
+    PRINT_CAPABILITY(CPU_CAPABILITY_AVX,  "AVX");
 
 #elif defined (__powerpc__) || defined (__ppc__) || defined (__ppc64__)
     PRINT_CAPABILITY(CPU_CAPABILITY_ALTIVEC, "AltiVec");
-- 
1.7.8



From remi at remlab.net  Fri Dec  9 16:56:08 2011
From: remi at remlab.net (=?utf-8?q?R=C3=A9mi?= Denis-Courmont)
Date: Fri, 9 Dec 2011 17:56:08 +0200
Subject: [vlc-devel] [PATCH?] xosd and xlib init
In-Reply-To: <20111209122717.6308ee87@gentoo.org>
References: <20111208200813.206d1142@gentoo.org>
	<201112091715.32782.remi@remlab.net>
	<20111209122717.6308ee87@gentoo.org>
Message-ID: <201112091756.08711.remi@remlab.net>

Le vendredi 9 d?cembre 2011 17:27:17 Alexis Ballier, vous avez ?crit :
> whats the correct/recommended way to do this ?
> adding $X_CFLAGS to xosd CFLAGS, and ${X_LIBS} ${X_PRE_LIBS} -lX11 to
> xosd LIBS ? (like eg, done for libva), or checking for no_x first and
> disabling xosd / bailing out like done for pulseaudio ?

Both?

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From remi at remlab.net  Fri Dec  9 17:02:07 2011
From: remi at remlab.net (=?iso-8859-1?q?R=E9mi?= Denis-Courmont)
Date: Fri, 9 Dec 2011 18:02:07 +0200
Subject: [vlc-devel] [PATCH] Detect AVX-enabled processors
In-Reply-To: <1323444491-10657-1-git-send-email-jb@videolan.org>
References: <1323444491-10657-1-git-send-email-jb@videolan.org>
Message-ID: <201112091802.08415.remi@remlab.net>

Le vendredi 9 d?cembre 2011 17:28:11 Jean-Baptiste Kempf, vous avez ?crit :
> ---
>  configure.ac      |   22 ++++++++++++++++++++++
>  include/vlc_cpu.h |   11 +++++++++++
>  src/misc/cpu.c    |   15 +++++++++++++++
>  3 files changed, 48 insertions(+), 0 deletions(-)
> 
> diff --git a/configure.ac b/configure.ac
> index d0538e2..869b9d2 100644
> --- a/configure.ac
> +++ b/configure.ac
> @@ -1336,6 +1336,28 @@ AS_IF([test "${ac_cv_3dnow_inline}" != "no"], [
>  ])
>  AM_CONDITIONAL([HAVE_3DNOW], [test "$have_3dnow" = "yes"])
> 
> +VLC_SAVE_FLAGS
> +CFLAGS="${CFLAGS} -mavx"
> +have_avx="no"
> +AC_CACHE_CHECK([if $CC groks AVX inline assembly], [ac_cv_avx_inline], [
> +  AC_COMPILE_IFELSE([AC_LANG_PROGRAM(,[[
> +void *p;
> +asm volatile("vzeroall"::"r"(p));
> +]])
> +  ], [
> +    ac_cv_avx_inline=yes
> +  ], [
> +    ac_cv_avx_inline=no
> +  ])
> +])
> +VLC_RESTORE_FLAGS
> +AS_IF([test "${ac_cv_avx_inline}" != "no"], [
> +  AC_DEFINE(CAN_COMPILE_AVX, 1,
> +            [Define to 1 if AVX inline assembly is available.])
> +  have_3dnow="yes"

Cut&paste typo?

> +])
> +AM_CONDITIONAL([HAVE_AVX], [test "$have_3dnow" = "yes"])
> +
> 
>  AC_ARG_ENABLE(neon,
>    [AS_HELP_STRING([--disable-neon],
> diff --git a/include/vlc_cpu.h b/include/vlc_cpu.h
> index 63cc978..a4dc66b 100644
> --- a/include/vlc_cpu.h
> +++ b/include/vlc_cpu.h
> @@ -37,6 +37,7 @@
>  #  define CPU_CAPABILITY_SSE4_1  (1<<10)
>  #  define CPU_CAPABILITY_SSE4_2  (1<<11)
>  #  define CPU_CAPABILITY_SSE4A   (1<<12)
> +#  define CPU_CAPABILITY_AVX     (1<<13)
> 
>  # if defined (__MMX__)
>  #  define VLC_MMX
> @@ -54,6 +55,15 @@
>  #  define VLC_SSE VLC_SSE_is_not_implemented_on_this_compiler
>  # endif
> 
> +# if defined (__AVX__)
> +#  define VLC_AVX
> +# elif VLC_GCC_VERSION(4, 6)
> +#  define VLC_AVX __attribute__ ((__target__ ("avx")))
> +# else
> +#  define VLC_AVX VLC_AVX_is_not_implemented_on_this_compiler
> +# endif

If you need that, then you should probably include an AVX register in the 
inline assembly test within configure.ac.

> +
> +
>  # else
>  #  define CPU_CAPABILITY_MMX     (0)
>  #  define CPU_CAPABILITY_3DNOW   (0)
> @@ -65,6 +75,7 @@
>  #  define CPU_CAPABILITY_SSE4_1  (0)
>  #  define CPU_CAPABILITY_SSE4_2  (0)
>  #  define CPU_CAPABILITY_SSE4A   (0)
> +#  define CPU_CAPABILITY_AVX     (0)
>  # endif
> 
>  # if defined (__ppc__) || defined (__ppc64__) || defined (__powerpc__)
> diff --git a/src/misc/cpu.c b/src/misc/cpu.c
> index 7c28b25..97826a9 100644
> --- a/src/misc/cpu.c
> +++ b/src/misc/cpu.c
> @@ -125,6 +125,13 @@ VLC_MMX static void ThreeD_Now_test (void)
>      asm volatile ("pfadd %%mm0,%%mm0\n" "femms\n" : : : "mm0");
>  }
>  #endif
> +#if defined (CAN_COMPILE_AVX) && !defined (__AVX__)
> +VLC_AVX static void AVX_test (void)
> +{
> +    asm volatile ("vzeroall\n" );
> +}
> +#endif
> +

Side topic... I think those fork/SIGILL tests are not really good idea. fork() 
can (and seems to) cause problems with libraries using pthread_atfork().

Besides, I don't really understand why we have those tests. A number of 
libraries just check for CPUID, including some libraries used by VLC. I wonder 
how this test can fail? Can the OS block those instructions, or can it block 
the MMX and/or SSE registers?

>  #if defined (CAN_COMPILE_ALTIVEC)
>  static void Altivec_text (void)
> @@ -262,6 +269,13 @@ void vlc_CPU_init (void)
>          i_capabilities |= CPU_CAPABILITY_SSE4_2;
>  # endif
> 
> +# if defined (__AVX__)
> +    i_capabilities |= CPU_CAPABILITY_AVX;
> +# elif defined (CAN_COMPILE_AVX)
> +    if ((i_ecx & (1 << 28)) && vlc_CPU_check ("AVX", AVX_test))
> +        i_capabilities |= CPU_CAPABILITY_AVX;
> +# endif
> +
>      /* test for additional capabilities */
>      cpuid( 0x80000000 );
> 
> @@ -384,6 +398,7 @@ void vlc_CPU_dump (vlc_object_t *obj)
>      PRINT_CAPABILITY(CPU_CAPABILITY_SSE4_1, "SSE4.1");
>      PRINT_CAPABILITY(CPU_CAPABILITY_SSE4_2, "SSE4.2");
>      PRINT_CAPABILITY(CPU_CAPABILITY_SSE4A,  "SSE4A");
> +    PRINT_CAPABILITY(CPU_CAPABILITY_AVX,  "AVX");
> 
>  #elif defined (__powerpc__) || defined (__ppc__) || defined (__ppc64__)
>      PRINT_CAPABILITY(CPU_CAPABILITY_ALTIVEC, "AltiVec");


-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From jb at videolan.org  Fri Dec  9 17:08:02 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Fri, 9 Dec 2011 17:08:02 +0100
Subject: [vlc-devel] [PATCH] Detect AVX-enabled processors
In-Reply-To: <201112091802.08415.remi@remlab.net>
References: <1323444491-10657-1-git-send-email-jb@videolan.org>
	<201112091802.08415.remi@remlab.net>
Message-ID: <20111209160802.GA17626@videolan.org>

On Fri, Dec 09, 2011 at 06:02:07PM +0200, R?mi Denis-Courmont wrote :
> Side topic... I think those fork/SIGILL tests are not really good idea. fork() 
> can (and seems to) cause problems with libraries using pthread_atfork().
> 
> Besides, I don't really understand why we have those tests. A number of 
> libraries just check for CPUID, including some libraries used by VLC. I wonder 
> how this test can fail? Can the OS block those instructions, or can it block 
> the MMX and/or SSE registers?

>From what I understand, we need to have the OS support to get access to
the YMM registers.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From remi at remlab.net  Fri Dec  9 17:18:58 2011
From: remi at remlab.net (=?iso-8859-1?q?R=E9mi?= Denis-Courmont)
Date: Fri, 9 Dec 2011 18:18:58 +0200
Subject: [vlc-devel] [PATCH] Detect AVX-enabled processors
In-Reply-To: <20111209160802.GA17626@videolan.org>
References: <1323444491-10657-1-git-send-email-jb@videolan.org>
	<201112091802.08415.remi@remlab.net>
	<20111209160802.GA17626@videolan.org>
Message-ID: <201112091818.59311.remi@remlab.net>

Le vendredi 9 d?cembre 2011 18:08:02 Jean-Baptiste Kempf, vous avez ?crit :
> On Fri, Dec 09, 2011 at 06:02:07PM +0200, R?mi Denis-Courmont wrote :
> > Side topic... I think those fork/SIGILL tests are not really good idea.
> > fork() can (and seems to) cause problems with libraries using
> > pthread_atfork().
> > 
> > Besides, I don't really understand why we have those tests. A number of
> > libraries just check for CPUID, including some libraries used by VLC. I
> > wonder how this test can fail? Can the OS block those instructions, or
> > can it block the MMX and/or SSE registers?
> 
> From what I understand, we need to have the OS support to get access to
> the YMM registers.

That actualy makes sense. If the OS does not know the registers, it cannot 
save and restore them during context switch. But then we only need to test 
MMX, SSE1 and AVX, not _everything_.

Also then:
- there is a problem on Windows (no tests at all),
- parsing /proc/cpuinfo is safer than fork+SIGILL.


-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From jb at videolan.org  Fri Dec  9 17:37:53 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Fri, 9 Dec 2011 17:37:53 +0100
Subject: [vlc-devel] [PATCH] Detect AVX-enabled processors
In-Reply-To: <201112091818.59311.remi@remlab.net>
References: <1323444491-10657-1-git-send-email-jb@videolan.org>
	<201112091802.08415.remi@remlab.net>
	<20111209160802.GA17626@videolan.org>
	<201112091818.59311.remi@remlab.net>
Message-ID: <20111209163753.GA20898@videolan.org>

On Fri, Dec 09, 2011 at 06:18:58PM +0200, R?mi Denis-Courmont wrote :
> That actualy makes sense. If the OS does not know the registers, it cannot 
> save and restore them during context switch. But then we only need to test 
> MMX, SSE1 and AVX, not _everything_.

Yes, this is exactly my understanding: we need to test MMX, XMM and YMM
registers but not anything else.

> Also then:
> - there is a problem on Windows (no tests at all),

Win XP SP2 supports MMX and SSE, so no issue here.

AVX is a different beast, and Only Win7/Win8 supports it.

> - parsing /proc/cpuinfo is safer than fork+SIGILL.

No opinion.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From remi at remlab.net  Fri Dec  9 18:41:15 2011
From: remi at remlab.net (=?iso-8859-15?q?R=E9mi?= Denis-Courmont)
Date: Fri, 9 Dec 2011 19:41:15 +0200
Subject: [vlc-devel] [PATCH] yadif : Add SSSE3 and SSE2 support. porting
	from FFmpeg.
In-Reply-To: <CAOa_K+vcob=0Af2vCLHH7jeQeBXhO5TnQtieo_LWNmC9_msgSQ@mail.gmail.com>
References: <CAOa_K+vdRUzBj4cE9qoLyU19+NYnVuPx03UqdOEXgkT6fw60Gg@mail.gmail.com>
	<20111205214242.GA27741@videolan.org>
	<CAOa_K+vcob=0Af2vCLHH7jeQeBXhO5TnQtieo_LWNmC9_msgSQ@mail.gmail.com>
Message-ID: <201112091941.15730.remi@remlab.net>

Le mardi 6 d?cembre 2011 16:50:48 Naohiro KORIYAMA, vous avez ?crit :
> I noticed the issue recently when I stated building my own win32 builds
> for chasing the problem around UPnP.
> Both OSX's clang and GCC don't complain  mm and xmm are in the clobber
> list.

That's because you have SSE enabled. This won't work were SSE is detected at 
run-time.

> GCC never allows the registers of unsupported instructions in the clobber
> list. GCC with -mmmx (or -march=pentium-mmx or above), "unknown register
> mm" errors disappear.
> GCC with -msse (or -march=pentium3 or above), "unknown register xmm"
> errors disappear.
> GCC with -mtune doesn't help.

Yeah sure. This was already discussed a few weeks ago here.

That's why we have VLC_MMX and VLC_SSE nowadays.

> Here is the short test log on Mac OS X, native GCC vs cross mingw32 GCC.
> http://pastebin.com/ZWA2JpRv
> 
> 
> I found a discussion about the clobber.
> http://ffmpeg.org/pipermail/ffmpeg-devel/2010-September/102164.html
> http://ffmpeg.org/pipermail/ffmpeg-devel/2010-October/098782.html
> As a result of the discussion,
> - Check whether xmm clobbers are supported in configure like:
> int main()
> {
>    asm volatile( "" ::: "xmm0");
> }
> - Define a macro that put xmm registers to the clobber list only if
> xmm clobbers are supported.

That's a brittle hack. Please don't do this.

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From remi at remlab.net  Fri Dec  9 18:42:46 2011
From: remi at remlab.net (=?iso-8859-15?q?R=E9mi?= Denis-Courmont)
Date: Fri, 9 Dec 2011 19:42:46 +0200
Subject: [vlc-devel] [Patch] Adding an helper to convert iso8601
	durations to seconds
In-Reply-To: <CAK-C75T4x2x4cBjFwvz-FtSseLsV1EoHe=RvfsqUAqb70mf7Aw@mail.gmail.com>
References: <CAK-C75R5BPUud6z2QxrFNNnh1wMFktu48O59oXi2bAUqZ0mYMw@mail.gmail.com>
	<CAK-C75RQNA-+iYR8wKs2LfFAE54oDOswZ_e-PXcpBJwe8W2zqg@mail.gmail.com>
	<CAK-C75T4x2x4cBjFwvz-FtSseLsV1EoHe=RvfsqUAqb70mf7Aw@mail.gmail.com>
Message-ID: <201112091942.47134.remi@remlab.net>

Function name is too long. And time_t is thecorrect type for seconds IMHO.

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From aballier at gentoo.org  Fri Dec  9 20:11:55 2011
From: aballier at gentoo.org (Alexis Ballier)
Date: Fri, 9 Dec 2011 16:11:55 -0300
Subject: [vlc-devel] [PATCH?] xosd and xlib init
In-Reply-To: <201112091756.08711.remi@remlab.net>
References: <20111208200813.206d1142@gentoo.org>
	<201112091715.32782.remi@remlab.net>
	<20111209122717.6308ee87@gentoo.org>
	<201112091756.08711.remi@remlab.net>
Message-ID: <20111209161155.5873e81c@gentoo.org>

On Fri, 9 Dec 2011 17:56:08 +0200
"R?mi Denis-Courmont" <remi at remlab.net> wrote:

> Le vendredi 9 d?cembre 2011 17:27:17 Alexis Ballier, vous avez ?crit :
> > whats the correct/recommended way to do this ?
> > adding $X_CFLAGS to xosd CFLAGS, and ${X_LIBS} ${X_PRE_LIBS} -lX11
> > to xosd LIBS ? (like eg, done for libva), or checking for no_x
> > first and disabling xosd / bailing out like done for pulseaudio ?
> 
> Both?
> 

attached patch should do it then
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-xosd-require-and-link-against-Xlib.patch
Type: text/x-patch
Size: 916 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111209/f76e6a04/attachment.bin>

From jb at videolan.org  Sat Dec 10 00:04:48 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Sat, 10 Dec 2011 00:04:48 +0100
Subject: [vlc-devel] [vlc-commits] Detect AVX-enabled processors
In-Reply-To: <20111209230031.0C0581452E3@albiero.videolan.org>
References: <20111209230031.0C0581452E3@albiero.videolan.org>
Message-ID: <20111209230448.GA750@videolan.org>

On Sat, Dec 10, 2011 at 12:00:30AM +0100, Jean-Baptiste Kempf wrote :
> vlc | branch: master | Jean-Baptiste Kempf <jb at videolan.org> | Fri Dec  9 16:23:28 2011 +0100| [09215763d3fa819c516fb88419f01f37e58c7383] | committer: Jean-Baptiste Kempf
> 
> Detect AVX-enabled processors
> 
> > http://git.videolan.org/gitweb.cgi/vlc.git/?a=commit;h=09215763d3fa819c516fb88419f01f37e58c7383

Sorry. Shouldn't have pushed that. Git mistake. Removed.


-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From nkoriyama at gmail.com  Sat Dec 10 06:55:16 2011
From: nkoriyama at gmail.com (Naohiro KORIYAMA)
Date: Sat, 10 Dec 2011 14:55:16 +0900
Subject: [vlc-devel] [PATCH] yadif : Add SSSE3 and SSE2 support. porting
 from FFmpeg.
In-Reply-To: <201112091941.15730.remi@remlab.net>
References: <CAOa_K+vdRUzBj4cE9qoLyU19+NYnVuPx03UqdOEXgkT6fw60Gg@mail.gmail.com>
	<20111205214242.GA27741@videolan.org>
	<CAOa_K+vcob=0Af2vCLHH7jeQeBXhO5TnQtieo_LWNmC9_msgSQ@mail.gmail.com>
	<201112091941.15730.remi@remlab.net>
Message-ID: <CAOa_K+shj1Wy4pm3=sFZxw4_nM+k-VCZnRauiQujeV0+LS4CGw@mail.gmail.com>

2011/12/10 R?mi Denis-Courmont <remi at remlab.net>:
> Le mardi 6 d?cembre 2011 16:50:48 Naohiro KORIYAMA, vous avez ?crit :
>> GCC never allows the registers of unsupported instructions in the clobber
>> list. GCC with -mmmx (or -march=pentium-mmx or above), "unknown register
>> mm" errors disappear.
>> GCC with -msse (or -march=pentium3 or above), "unknown register xmm"
>> errors disappear.
>> GCC with -mtune doesn't help.
>
> Yeah sure. This was already discussed a few weeks ago here.

Oops, I missed the discussion.

> That's why we have VLC_MMX and VLC_SSE nowadays.

It's smart, easy, and better solution for the compilation issue.

>
> That's a brittle hack. Please don't do this.
>

OK. I reverted that part. and used VLC_MMX and VLC_SSE.
I confirmed the compilation issue is fixed!

Thanks for your advice.

Best regards,


-- 
KORIYAMA, Naohiro
nkoriyama at gmail.com
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-yadif-Add-SSSE3-and-SSE2-support.-porting-from-FFmpe.patch
Type: application/octet-stream
Size: 30720 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111210/7bc39d7d/attachment.obj>

From fenrir at elivagar.org  Sat Dec 10 23:43:22 2011
From: fenrir at elivagar.org (Laurent Aimar)
Date: Sat, 10 Dec 2011 23:43:22 +0100
Subject: [vlc-devel] [vlc-commits] avcodec: fix direct rendering with
	some codecs
In-Reply-To: <20111210220007.D76BA145379@albiero.videolan.org>
References: <20111210220007.D76BA145379@albiero.videolan.org>
Message-ID: <20111210224322.GA3494@elivagar.org>

On Sat, Dec 10, 2011 at 11:00:07PM +0100, Rafa?l Carr? wrote:
> vlc | branch: master | Rafa?l Carr? <funman at videolan.org> | Sat Dec 10 16:58:35 2011 -0500| [fcc84608b67f3c3567b96604577bfdcc0de45b53] | committer: Rafa?l Carr?
> 
> avcodec: fix direct rendering with some codecs
> 
> Provide edge buffers when CODEC_FLAG_EMU_EDGE is set
> Fix ProRes playback with direct rendering

 From what I understand of avcodec, this is wrong. We need edge only when
CODEC_FLAG_EMU_EDGE isn't set while we always set it.
Besides, I think (but unsure) that you need to allocated edge on the right
*and* left side of the buffers, which won't work as is.

Also, increasing the resolution won't correctly work, mostly due to filter
and vout modules (you will end up in some case with green pixels on the rigt).

So I think this commit should be reverted.

Regards,

-- 
fenrir



From funman at videolan.org  Sun Dec 11 00:07:21 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Sat, 10 Dec 2011 18:07:21 -0500
Subject: [vlc-devel] [vlc-commits] avcodec: fix direct rendering with
 some codecs
In-Reply-To: <20111210224322.GA3494@elivagar.org>
References: <20111210220007.D76BA145379@albiero.videolan.org>
	<20111210224322.GA3494@elivagar.org>
Message-ID: <20111210180721.44dc660a@tok>

Le Sat, 10 Dec 2011 23:43:22 +0100,
Laurent Aimar <fenrir at elivagar.org> a ?crit :

> On Sat, Dec 10, 2011 at 11:00:07PM +0100, Rafa?l Carr? wrote:
> > vlc | branch: master | Rafa?l Carr? <funman at videolan.org> | Sat Dec 10 16:58:35 2011 -0500| [fcc84608b67f3c3567b96604577bfdcc0de45b53] | committer: Rafa?l Carr?
> > 
> > avcodec: fix direct rendering with some codecs
> > 
> > Provide edge buffers when CODEC_FLAG_EMU_EDGE is set
> > Fix ProRes playback with direct rendering
> 
>  From what I understand of avcodec, this is wrong. We need edge only when
> CODEC_FLAG_EMU_EDGE isn't set while we always set it.

Hm right, EMU_EDGE shouldn't draw in this border.

I'm not sure why this fixes ProRes decoding when EMU_EDGE is set,
perhaps libav doesn't respect the flag, or perhaps video buffers
provide the edges already?

When breaking in avcodec functions I could see that it was assuming a
buffer 16 or 32 pixels larger than the width, and not setting EMU_EDGE
effectively fixed playback of ProRes.

> Besides, I think (but unsure) that you need to allocated edge on the right
> *and* left side of the buffers, which won't work as is.

Right, I think it also includes top and bottom sides.

> 
> Also, increasing the resolution won't correctly work, mostly due to filter
> and vout modules (you will end up in some case with green pixels on the rigt).

What would go wrong? We don't change i_visible_width

> So I think this commit should be reverted.

Ok.


Meanwhile I can look for a better fix but I don't understand what is
wrong with just increasing resolution.

For left/top edges we can use 
    unsigned int i_x_offset;               /**< start offset of visible area */
    unsigned int i_y_offset;               /**< start offset of visible area */

The only sample I have is AppleProRes422.mov (yuv422p10le) but IIUC any
vp8 stream would work.

-- 
Rafa?l Carr?


From fenrir at elivagar.org  Sun Dec 11 00:29:54 2011
From: fenrir at elivagar.org (Laurent Aimar)
Date: Sun, 11 Dec 2011 00:29:54 +0100
Subject: [vlc-devel] [vlc-commits] avcodec: fix direct rendering
	with	some codecs
In-Reply-To: <20111210180721.44dc660a@tok>
References: <20111210220007.D76BA145379@albiero.videolan.org>
	<20111210224322.GA3494@elivagar.org> <20111210180721.44dc660a@tok>
Message-ID: <20111210232954.GA5925@elivagar.org>

On Sat, Dec 10, 2011 at 06:07:21PM -0500, Rafa?l Carr? wrote:
> Le Sat, 10 Dec 2011 23:43:22 +0100,
> Laurent Aimar <fenrir at elivagar.org> a ?crit :
> 
> > On Sat, Dec 10, 2011 at 11:00:07PM +0100, Rafa?l Carr? wrote:
> > > vlc | branch: master | Rafa?l Carr? <funman at videolan.org> | Sat Dec 10 16:58:35 2011 -0500| [fcc84608b67f3c3567b96604577bfdcc0de45b53] | committer: Rafa?l Carr?
> > > 
> > > avcodec: fix direct rendering with some codecs
> > > 
> > > Provide edge buffers when CODEC_FLAG_EMU_EDGE is set
> > > Fix ProRes playback with direct rendering
> > 
> >  From what I understand of avcodec, this is wrong. We need edge only when
> > CODEC_FLAG_EMU_EDGE isn't set while we always set it.
> 
> Hm right, EMU_EDGE shouldn't draw in this border.
> 
> I'm not sure why this fixes ProRes decoding when EMU_EDGE is set,
> perhaps libav doesn't respect the flag, or perhaps video buffers
> provide the edges already?

 Either the prores decoder doesn't respect EMU_EDGE or it expects a greater
buffer/pitch alignment than we provide.

> > Besides, I think (but unsure) that you need to allocated edge on the right
> > *and* left side of the buffers, which won't work as is.
> 
> Right, I think it also includes top and bottom sides.
 That too.

> > Also, increasing the resolution won't correctly work, mostly due to filter
> > and vout modules (you will end up in some case with green pixels on the rigt).
> 
> What would go wrong? We don't change i_visible_width
 Most filters will use pixels beyong i_visible_width/height (up to
width/height). It's the same with some vout (ex: xvideo/opengl when they do
the resizing, it seems to depends on the driver though). If thoses pixels are
not set and end up in the visible part, you have artifacts.

 So basically, x/y offset, visible w/h are only safe to use if it's to crop
a picture and the left over part isn't garbage.

-- 
fenrir



From funman at videolan.org  Sun Dec 11 07:03:33 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Sun, 11 Dec 2011 01:03:33 -0500
Subject: [vlc-devel] [vlc-commits] avcodec: fix direct rendering with
 some codecs
In-Reply-To: <20111210232954.GA5925@elivagar.org>
References: <20111210220007.D76BA145379@albiero.videolan.org>
	<20111210224322.GA3494@elivagar.org> <20111210180721.44dc660a@tok>
	<20111210232954.GA5925@elivagar.org>
Message-ID: <20111211010333.2df5b35c@tok>

Le Sun, 11 Dec 2011 00:29:54 +0100,
Laurent Aimar <fenrir at elivagar.org> a ?crit :

> On Sat, Dec 10, 2011 at 06:07:21PM -0500, Rafa?l Carr? wrote:
> > Le Sat, 10 Dec 2011 23:43:22 +0100,
> > Laurent Aimar <fenrir at elivagar.org> a ?crit :
> > 
> > > On Sat, Dec 10, 2011 at 11:00:07PM +0100, Rafa?l Carr? wrote:
> > > > vlc | branch: master | Rafa?l Carr? <funman at videolan.org> | Sat Dec 10 16:58:35 2011 -0500| [fcc84608b67f3c3567b96604577bfdcc0de45b53] | committer: Rafa?l Carr?
> > > > 
> > > > avcodec: fix direct rendering with some codecs
> > > > 
> > > > Provide edge buffers when CODEC_FLAG_EMU_EDGE is set
> > > > Fix ProRes playback with direct rendering
> > > 
> > >  From what I understand of avcodec, this is wrong. We need edge only when
> > > CODEC_FLAG_EMU_EDGE isn't set while we always set it.

Reverted.

> > Hm right, EMU_EDGE shouldn't draw in this border.
> > 
> > I'm not sure why this fixes ProRes decoding when EMU_EDGE is set,
> > perhaps libav doesn't respect the flag, or perhaps video buffers
> > provide the edges already?
> 
>  Either the prores decoder doesn't respect EMU_EDGE or it expects a greater
> buffer/pitch alignment than we provide.

We use different picture sizes for vout and decoder when using direct
rendering.

vout pictures are aligned on requested width/height, but decoder
pictures width/height are modified by picture_Setup()

    /* We want V (width/height) to respect:
        (V * p_dsc->p[i].w.i_num) % p_dsc->p[i].w.i_den == 0
        (V * p_dsc->p[i].w.i_num/p_dsc->p[i].w.i_den * p_dsc->i_pixel_size) % 16 == 0
       Which is respected if you have
       V % lcm( p_dsc->p[0..planes].w.i_den * 16) == 0
    */

IIUC we modify picture width so all planes are modulo 16 (IIUC), so
when there is chroma subsampling, even if the width is modulo 16, the
chroma planes are not necessary.

The problem might be in swscale when converting from the 10bits FFmpeg
output to xvideo yuy2.

However when running in valgrind I see an invalid write in asm in the
decoding path, so it's using correctly sized picture. Perhaps another
bug though.

It's a bit fucked up but navigating between layers of decoder, libav,
and display is quite confusing :o

pointers welcome

-- 
Rafa?l Carr?


From fenrir at elivagar.org  Sun Dec 11 11:25:29 2011
From: fenrir at elivagar.org (Laurent Aimar)
Date: Sun, 11 Dec 2011 11:25:29 +0100
Subject: [vlc-devel] [vlc-commits] avcodec: fix direct rendering
	with	some codecs
In-Reply-To: <20111211010333.2df5b35c@tok>
References: <20111210220007.D76BA145379@albiero.videolan.org>
	<20111210224322.GA3494@elivagar.org> <20111210180721.44dc660a@tok>
	<20111210232954.GA5925@elivagar.org> <20111211010333.2df5b35c@tok>
Message-ID: <20111211102529.GA6144@elivagar.org>

On Sun, Dec 11, 2011 at 01:03:33AM -0500, Rafa?l Carr? wrote:
> Le Sun, 11 Dec 2011 00:29:54 +0100,
> Laurent Aimar <fenrir at elivagar.org> a ?crit :
> 
> > On Sat, Dec 10, 2011 at 06:07:21PM -0500, Rafa?l Carr? wrote:
> > > Le Sat, 10 Dec 2011 23:43:22 +0100,
> > > Laurent Aimar <fenrir at elivagar.org> a ?crit :
> > > 
> > > > On Sat, Dec 10, 2011 at 11:00:07PM +0100, Rafa?l Carr? wrote:
> > > > > vlc | branch: master | Rafa?l Carr? <funman at videolan.org> | Sat Dec 10 16:58:35 2011 -0500| [fcc84608b67f3c3567b96604577bfdcc0de45b53] | committer: Rafa?l Carr?
> > > > > 
> > > > > avcodec: fix direct rendering with some codecs
> > > > > 
> > > > > Provide edge buffers when CODEC_FLAG_EMU_EDGE is set
> > > > > Fix ProRes playback with direct rendering
> > > > 
> > > >  From what I understand of avcodec, this is wrong. We need edge only when
> > > > CODEC_FLAG_EMU_EDGE isn't set while we always set it.
> 
> Reverted.
> 
> > > Hm right, EMU_EDGE shouldn't draw in this border.
> > > 
> > > I'm not sure why this fixes ProRes decoding when EMU_EDGE is set,
> > > perhaps libav doesn't respect the flag, or perhaps video buffers
> > > provide the edges already?
> > 
> >  Either the prores decoder doesn't respect EMU_EDGE or it expects a greater
> > buffer/pitch alignment than we provide.
> 
> We use different picture sizes for vout and decoder when using direct
> rendering.
No. The pictures than a decoder in VLC receives can comes from 2 sources:
 1. the vout.
 2. picture_New() (through a picture_pool I think)

In the case 1., we don't always control the properties of the buffers except
that it will be large enough for holding width/height (ie we may not have
sufficient data pointer/pitch alignment).
That's partially why we have a compatibility check in ffmpeg_GetFrameBuf().

> vout pictures are aligned on requested width/height, but decoder
> pictures width/height are modified by picture_Setup()
> 
>     /* We want V (width/height) to respect:
>         (V * p_dsc->p[i].w.i_num) % p_dsc->p[i].w.i_den == 0
>         (V * p_dsc->p[i].w.i_num/p_dsc->p[i].w.i_den * p_dsc->i_pixel_size) % 16 == 0
>        Which is respected if you have
>        V % lcm( p_dsc->p[0..planes].w.i_den * 16) == 0
>     */
 Not always, picture_Setup() set what VLC wants, but the vout may not be using
it or it can always force other values (ex: xvideo does not use it, see
PoolAlloc() in xcb/xvideo.c).

> IIUC we modify picture width so all planes are modulo 16 (IIUC), so
> when there is chroma subsampling, even if the width is modulo 16, the
> chroma planes are not necessary.
 I don't understand this part/what you mean.

> pointers welcome
 I am a bit dubious about the fact that the proress decoder in ffmpeg doesn't
seem to check the value of CODEC_FLAG_EMU_EDGE. (h264 decoder does for
example).

-- 
fenrir



From var1able at var1able.ru  Sun Dec 11 12:18:27 2011
From: var1able at var1able.ru (=?utf-8?B?0JrQvtC90YHRgtCw0L3RgtC40L0g0JHQvtCz0LTQsNC90L7Qsg==?=)
Date: Sun, 11 Dec 2011 15:18:27 +0400
Subject: [vlc-devel] Web parsers from GCI
Message-ID: <76188B80-326F-47AE-ADBE-E3E6E8AC8CB6@var1able.ru>

http://www.google-melange.com/gci/task/view/google/gci2011/7133336
See attachment
-------------- next part --------------
A non-text attachment was scrubbed...
Name: liveleak1.lua
Type: application/octet-stream
Size: 1616 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111211/f0570e64/attachment.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: videobash1.lua
Type: application/octet-stream
Size: 1633 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111211/f0570e64/attachment-0001.obj>
-------------- next part --------------

_______________
Best regards,
Konstantin 

From benoit.tuduri at gmail.com  Sun Dec 11 16:19:57 2011
From: benoit.tuduri at gmail.com (=?ISO-8859-1?Q?beno=EEt_tuduri?=)
Date: Sun, 11 Dec 2011 16:19:57 +0100
Subject: [vlc-devel] patch for french translation on macosx
Message-ID: <CALKXa+-31pVS-aOUvk-OhXpOyG-dfME9w+M=2B-TtS-Z-uinRQ@mail.gmail.com>

Hello,

You can find a patch to fix a typo error.

Sincerly,
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-FIX-typo.patch
Type: application/octet-stream
Size: 651 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111211/4fe52017/attachment.obj>

From mehrotra.akash at gmail.com  Sun Dec 11 20:24:25 2011
From: mehrotra.akash at gmail.com (akash mehrotra)
Date: Mon, 12 Dec 2011 00:54:25 +0530
Subject: [vlc-devel] [PATCH] Fix EQ HTTP UI partially
Message-ID: <CAOgZoSd6MBJdT_DFKwWSD+AmNEBCA6EujhWyX6MH96hGur4AHg@mail.gmail.com>

Partial fix for the Lua Web Interface Equalizer
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111212/50e2b707/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-Fix-EQ-HTTP-UI-partially.patch
Type: application/octet-stream
Size: 7344 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111212/50e2b707/attachment.obj>

From funman at videolan.org  Sun Dec 11 20:47:55 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Sun, 11 Dec 2011 14:47:55 -0500
Subject: [vlc-devel] [vlc-commits] avcodec: fix direct rendering with
 some codecs
In-Reply-To: <20111211102529.GA6144@elivagar.org>
References: <20111210220007.D76BA145379@albiero.videolan.org>
	<20111210224322.GA3494@elivagar.org> <20111210180721.44dc660a@tok>
	<20111210232954.GA5925@elivagar.org> <20111211010333.2df5b35c@tok>
	<20111211102529.GA6144@elivagar.org>
Message-ID: <20111211144755.7aa57f9d@tok>

Le Sun, 11 Dec 2011 11:25:29 +0100,
Laurent Aimar <fenrir at elivagar.org> a ?crit :

> On Sun, Dec 11, 2011 at 01:03:33AM -0500, Rafa?l Carr? wrote:
> > Le Sun, 11 Dec 2011 00:29:54 +0100,
> > Laurent Aimar <fenrir at elivagar.org> a ?crit :

> > > > Hm right, EMU_EDGE shouldn't draw in this border.
> > > > 
> > > > I'm not sure why this fixes ProRes decoding when EMU_EDGE is set,
> > > > perhaps libav doesn't respect the flag, or perhaps video buffers
> > > > provide the edges already?
> > > 
> > >  Either the prores decoder doesn't respect EMU_EDGE or it expects a greater
> > > buffer/pitch alignment than we provide.
> > 
> > We use different picture sizes for vout and decoder when using direct
> > rendering.
> No. The pictures than a decoder in VLC receives can comes from 2 sources:
>  1. the vout.
>  2. picture_New() (through a picture_pool I think)
> 
> In the case 1., we don't always control the properties of the buffers except
> that it will be large enough for holding width/height (ie we may not have
> sufficient data pointer/pitch alignment).
> That's partially why we have a compatibility check in ffmpeg_GetFrameBuf().

ok

> > vout pictures are aligned on requested width/height, but decoder
> > pictures width/height are modified by picture_Setup()
> > 
> >     /* We want V (width/height) to respect:
> >         (V * p_dsc->p[i].w.i_num) % p_dsc->p[i].w.i_den == 0
> >         (V * p_dsc->p[i].w.i_num/p_dsc->p[i].w.i_den * p_dsc->i_pixel_size) % 16 == 0
> >        Which is respected if you have
> >        V % lcm( p_dsc->p[0..planes].w.i_den * 16) == 0
> >     */
>  Not always, picture_Setup() set what VLC wants, but the vout may not be using
> it or it can always force other values (ex: xvideo does not use it, see
> PoolAlloc() in xcb/xvideo.c).

Yes i noticed that.

For ProRes, vout and decoder chromas are not compatible so there is no
direct copy: swscale is used instead.

> > IIUC we modify picture width so all planes are modulo 16 (IIUC), so
> > when there is chroma subsampling, even if the width is modulo 16, the
> > chroma planes are not necessary.
>  I don't understand this part/what you mean.

Usually there is the requirement that width/height are aligned on 16
(pixels) as required by the encoder.

I do not fully understand the maths, I thought that we require the
stride of each plane to be aligned on 16 bytes, but for the sample I
have (720x486), the luma stride is 1472 (== 32*x) instead of 1440
(2 * 720 which is already aligned on 32).

The height is changed from 486 to 498, neither are aligned on 32; and i
don't understand the purpose anyway.

> > pointers welcome
>  I am a bit dubious about the fact that the proress decoder in ffmpeg doesn't
> seem to check the value of CODEC_FLAG_EMU_EDGE. (h264 decoder does for
> example).

I thought that perhaps edges are drawn by some helper function because
only a handful of decoders check this flag, or perhaps they are just
not drawn at all.

Why do we need libavcodec to draw the edges for us anyway?

-- 
Rafa?l Carr?


From fenrir at elivagar.org  Sun Dec 11 21:13:07 2011
From: fenrir at elivagar.org (Laurent Aimar)
Date: Sun, 11 Dec 2011 21:13:07 +0100
Subject: [vlc-devel] [vlc-commits] avcodec: fix direct rendering
	with	some codecs
In-Reply-To: <20111211144755.7aa57f9d@tok>
References: <20111210220007.D76BA145379@albiero.videolan.org>
	<20111210224322.GA3494@elivagar.org> <20111210180721.44dc660a@tok>
	<20111210232954.GA5925@elivagar.org> <20111211010333.2df5b35c@tok>
	<20111211102529.GA6144@elivagar.org> <20111211144755.7aa57f9d@tok>
Message-ID: <20111211201307.GA3912@elivagar.org>

On Sun, Dec 11, 2011 at 02:47:55PM -0500, Rafa?l Carr? wrote:
> > > vout pictures are aligned on requested width/height, but decoder
> > > pictures width/height are modified by picture_Setup()
> > > 
> > >     /* We want V (width/height) to respect:
> > >         (V * p_dsc->p[i].w.i_num) % p_dsc->p[i].w.i_den == 0
> > >         (V * p_dsc->p[i].w.i_num/p_dsc->p[i].w.i_den * p_dsc->i_pixel_size) % 16 == 0
> > >        Which is respected if you have
> > >        V % lcm( p_dsc->p[0..planes].w.i_den * 16) == 0
> > >     */
> >  Not always, picture_Setup() set what VLC wants, but the vout may not be using
> > it or it can always force other values (ex: xvideo does not use it, see
> > PoolAlloc() in xcb/xvideo.c).
> 
> Yes i noticed that.
> 
> For ProRes, vout and decoder chromas are not compatible so there is no
> direct copy: swscale is used instead.
> 
> > > IIUC we modify picture width so all planes are modulo 16 (IIUC), so
> > > when there is chroma subsampling, even if the width is modulo 16, the
> > > chroma planes are not necessary.
> >  I don't understand this part/what you mean.
> 
> Usually there is the requirement that width/height are aligned on 16
> (pixels) as required by the encoder.
Nope, no such things.
What avcodec requires:
 - each pitch/stride must at least be a multiple of 16
 - the base pointer of each plane must be aligned (16)
 - the stride/pitch of the luma has the right ratio against the chroma (eg: *2 for 420).
 - the buffer provided can at least store a picture of the wxh returned by
   avcodec_align_dimensions()
 - the pitch/stride of consecutives pictures are the sames.
and that suppose that we do request EMU_EDGE. If we don't, they are more
constraints that we cannot support at the moment.

> I do not fully understand the maths, I thought that we require the
> stride of each plane to be aligned on 16 bytes, but for the sample I
> have (720x486), the luma stride is 1472 (== 32*x) instead of 1440
> (2 * 720 which is already aligned on 32).
> The height is changed from 486 to 498, neither are aligned on 32; and i
> don't understand the purpose anyway.

 Which colorspace? And what do you mean by width/height? (width/height or i_pitch/i_lines).

> > > pointers welcome
> >  I am a bit dubious about the fact that the proress decoder in ffmpeg doesn't
> > seem to check the value of CODEC_FLAG_EMU_EDGE. (h264 decoder does for
> > example).
> 
> I thought that perhaps edges are drawn by some helper function because
> only a handful of decoders check this flag, or perhaps they are just
> not drawn at all.
> 
> Why do we need libavcodec to draw the edges for us anyway?
 It's not that we require the edges, it's because we cannot provide a
buffer suitable for avcodec if we don't ask for EMU_EDGE (or if we could,
we would probably end up doing a copy later anyway, at least for playback).

-- 
fenrir



From fenrir at elivagar.org  Sun Dec 11 21:15:34 2011
From: fenrir at elivagar.org (Laurent Aimar)
Date: Sun, 11 Dec 2011 21:15:34 +0100
Subject: [vlc-devel] [vlc-commits] avcodec: fix direct rendering
	with	some codecs
In-Reply-To: <20111211201307.GA3912@elivagar.org>
References: <20111210220007.D76BA145379@albiero.videolan.org>
	<20111210224322.GA3494@elivagar.org> <20111210180721.44dc660a@tok>
	<20111210232954.GA5925@elivagar.org> <20111211010333.2df5b35c@tok>
	<20111211102529.GA6144@elivagar.org> <20111211144755.7aa57f9d@tok>
	<20111211201307.GA3912@elivagar.org>
Message-ID: <20111211201534.GA4867@elivagar.org>

On Sun, Dec 11, 2011 at 09:13:07PM +0100, Laurent Aimar wrote:
> > Why do we need libavcodec to draw the edges for us anyway?
>  It's not that we require the edges, it's because we cannot provide a
> buffer suitable for avcodec if we don't ask for EMU_EDGE (or if we could,
> we would probably end up doing a copy later anyway, at least for playback).
 Btw, when we set EMU_EDGE it is to prevent avcodec from drawing/using the
edges space (and so from having to have enough space for them in the buffers). 

-- 
fenrir



From funman at videolan.org  Sun Dec 11 21:20:41 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Sun, 11 Dec 2011 15:20:41 -0500
Subject: [vlc-devel] [vlc-commits] avcodec: fix direct rendering with
 some codecs
In-Reply-To: <20111211201534.GA4867@elivagar.org>
References: <20111210220007.D76BA145379@albiero.videolan.org>
	<20111210224322.GA3494@elivagar.org> <20111210180721.44dc660a@tok>
	<20111210232954.GA5925@elivagar.org> <20111211010333.2df5b35c@tok>
	<20111211102529.GA6144@elivagar.org> <20111211144755.7aa57f9d@tok>
	<20111211201307.GA3912@elivagar.org>
	<20111211201534.GA4867@elivagar.org>
Message-ID: <20111211152041.3bbb72b0@tok>

Le Sun, 11 Dec 2011 21:15:34 +0100,
Laurent Aimar <fenrir at elivagar.org> a ?crit :

> On Sun, Dec 11, 2011 at 09:13:07PM +0100, Laurent Aimar wrote:
> > > Why do we need libavcodec to draw the edges for us anyway?
> >  It's not that we require the edges, it's because we cannot provide a
> > buffer suitable for avcodec if we don't ask for EMU_EDGE (or if we could,
> > we would probably end up doing a copy later anyway, at least for playback).
>  Btw, when we set EMU_EDGE it is to prevent avcodec from drawing/using the
> edges space (and so from having to have enough space for them in the buffers). 
> 

Yeah we should set this flag unconditionally I guess.

-- 
Rafa?l Carr?


From fenrir at elivagar.org  Sun Dec 11 21:25:53 2011
From: fenrir at elivagar.org (Laurent Aimar)
Date: Sun, 11 Dec 2011 21:25:53 +0100
Subject: [vlc-devel] [vlc-commits] avcodec: fix direct rendering
	with	some codecs
In-Reply-To: <20111211152041.3bbb72b0@tok>
References: <20111210220007.D76BA145379@albiero.videolan.org>
	<20111210224322.GA3494@elivagar.org> <20111210180721.44dc660a@tok>
	<20111210232954.GA5925@elivagar.org> <20111211010333.2df5b35c@tok>
	<20111211102529.GA6144@elivagar.org> <20111211144755.7aa57f9d@tok>
	<20111211201307.GA3912@elivagar.org>
	<20111211201534.GA4867@elivagar.org> <20111211152041.3bbb72b0@tok>
Message-ID: <20111211202553.GA5418@elivagar.org>

On Sun, Dec 11, 2011 at 03:20:41PM -0500, Rafa?l Carr? wrote:
> Le Sun, 11 Dec 2011 21:15:34 +0100,
> Laurent Aimar <fenrir at elivagar.org> a ?crit :
> 
> > On Sun, Dec 11, 2011 at 09:13:07PM +0100, Laurent Aimar wrote:
> > > > Why do we need libavcodec to draw the edges for us anyway?
> > >  It's not that we require the edges, it's because we cannot provide a
> > > buffer suitable for avcodec if we don't ask for EMU_EDGE (or if we could,
> > > we would probably end up doing a copy later anyway, at least for playback).
> >  Btw, when we set EMU_EDGE it is to prevent avcodec from drawing/using the
> > edges space (and so from having to have enough space for them in the buffers). 
> > 
> 
> Yeah we should set this flag unconditionally I guess.
 Wich we do when trying to use direct-rendering. If we don't use
direct-rendering then it's better not to ask for it (a bit faster en general)

-- 
fenrir


From funman at videolan.org  Sun Dec 11 21:50:33 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Sun, 11 Dec 2011 15:50:33 -0500
Subject: [vlc-devel] [vlc-commits] avcodec: fix direct rendering with
 some codecs
In-Reply-To: <20111211201307.GA3912@elivagar.org>
References: <20111210220007.D76BA145379@albiero.videolan.org>
	<20111210224322.GA3494@elivagar.org> <20111210180721.44dc660a@tok>
	<20111210232954.GA5925@elivagar.org> <20111211010333.2df5b35c@tok>
	<20111211102529.GA6144@elivagar.org> <20111211144755.7aa57f9d@tok>
	<20111211201307.GA3912@elivagar.org>
Message-ID: <20111211155033.4a89d8ef@tok>

Le Sun, 11 Dec 2011 21:13:07 +0100,
Laurent Aimar <fenrir at elivagar.org> a ?crit :

> > > > IIUC we modify picture width so all planes are modulo 16 (IIUC), so
> > > > when there is chroma subsampling, even if the width is modulo 16, the
> > > > chroma planes are not necessary.
> > >  I don't understand this part/what you mean.
> > 
> > Usually there is the requirement that width/height are aligned on 16
> > (pixels) as required by the encoder.
> Nope, no such things.
> What avcodec requires:
>  - each pitch/stride must at least be a multiple of 16
>  - the base pointer of each plane must be aligned (16)
>  - the stride/pitch of the luma has the right ratio against the chroma (eg: *2 for 420).
>  - the buffer provided can at least store a picture of the wxh returned by
>    avcodec_align_dimensions()
>  - the pitch/stride of consecutives pictures are the sames.
> and that suppose that we do request EMU_EDGE. If we don't, they are more
> constraints that we cannot support at the moment.
> 
> > I do not fully understand the maths, I thought that we require the
> > stride of each plane to be aligned on 16 bytes, but for the sample I
> > have (720x486), the luma stride is 1472 (== 32*x) instead of 1440
> > (2 * 720 which is already aligned on 32).
> > The height is changed from 486 to 498, neither are aligned on 32; and i
> > don't understand the purpose anyway.
> 
>  Which colorspace? And what do you mean by width/height? (width/height or i_pitch/i_lines).

[0xa44a828] swscale filter debug: 720x486 chroma: I2AL -> 720x486 chroma: YUY2 with scaling using Bicubic (good quality)

So chroma is I2AL

The picture is here:
	width/height = visible width/height = 720x486
For the 3 planes though:
	width/height = 736x498
	visible width/height = 720x486

#0  ffmpeg_GetFrameBuf (p_context=0x716b20, p_ff_pic=0x717e10) at ../../../../vlc/modules/codec/avcodec/video.c:1062
1062	    p_ff_pic->age = 256*256*256*64; // FIXME FIXME from ffmpeg
(gdb) print *p_pic
$16 = {format = {i_chroma = 1279341129, i_width = 720, i_height = 486, i_x_offset = 0, i_y_offset = 0, i_visible_width = 720, 
    i_visible_height = 486, i_bits_per_pixel = 0, i_sar_num = 1, i_sar_den = 1, i_frame_rate = 2997, i_frame_rate_base = 100, 
    i_rmask = 0, i_gmask = 0, i_bmask = 0, i_rrshift = 0, i_lrshift = 0, i_rgshift = 0, i_lgshift = 0, i_rbshift = 0, 
    i_lbshift = 0, p_palette = 0x0}, p_data_orig = 0x17bc0c0, p = {{p_pixels = 0x17bc0c0 "", i_lines = 498, i_pitch = 1472, 
      i_pixel_pitch = 2, i_visible_lines = 486, i_visible_pitch = 1440}, {p_pixels = 0x186f040 "", i_lines = 498, 
      i_pitch = 736, i_pixel_pitch = 2, i_visible_lines = 486, i_visible_pitch = 720}, {p_pixels = 0x18c8800 "", 
      i_lines = 498, i_pitch = 736, i_pixel_pitch = 2, i_visible_lines = 486, i_visible_pitch = 720}, {p_pixels = 0x0, 
      i_lines = 0, i_pitch = 0, i_pixel_pitch = 0, i_visible_lines = 0, i_visible_pitch = 0}, {p_pixels = 0x0, i_lines = 0, 
      i_pitch = 0, i_pixel_pitch = 0, i_visible_lines = 0, i_visible_pitch = 0}}, i_planes = 3, i_refcount = 1, date = 0, 
  b_force = false, b_progressive = false, b_top_field_first = false, i_nb_fields = 2, p_q = 0x0, i_qstride = 0, i_qtype = 0, 
  p_sys = 0x0, pf_release = 0x7ffff7961030 <Release>, p_release_sys = 0x8e6050, p_next = 0x0}


Here is valgrind/gdb traces at invalid write


valgrind --db-attach=yes vlc --no-spu --no-osd --no-audio --no-media-library --ignore-config -I dummy -vv --ffmpeg-dr --ffmpeg-threads 1 AppleProRes422.mov 

...

[0x6a18628] avcodec decoder debug: using direct rendering
==2248== Invalid write of size 8
==2248==    at 0x145B00A8: ??? (proresdsp.asm:428)
==2248==    by 0x14453BB0: decode_slice_plane (proresdec.c:492)
==2248==    by 0x1445403F: decode_slice (proresdec.c:587)
==2248==    by 0x144D690E: avcodec_default_execute (utils.c:568)
==2248==    by 0x1445315C: decode_frame (proresdec.c:624)
==2248==    by 0x144D7277: avcodec_decode_video2 (utils.c:866)
==2248==    by 0x13EAEF70: DecodeVideo (video.c:590)
==2248==    by 0x5085966: DecoderDecodeVideo (decoder.c:1512)
==2248==    by 0x508567F: DecoderProcess (decoder.c:1878)
==2248==    by 0x508577A: DecoderThread (decoder.c:938)
==2248==    by 0x532AEFB: start_thread (pthread_create.c:304)
==2248==    by 0x582585C: clone (clone.S:112)
==2248==  Address 0x96c5650 is 0 bytes after a block of size 1,466,112 alloc'd
==2248==    at 0x4C278E8: memalign (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
==2248==    by 0x4C27997: posix_memalign (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
==2248==    by 0x50C05AF: picture_NewFromResource (vlc_common.h:935)
==2248==    by 0x50C1267: picture_pool_NewFromFormat (picture_pool.c:138)
==2248==    by 0x50AF3DE: vout_InitWrapper (vout_wrapper.c:153)
==2248==    by 0x50A78FB: ThreadStart (video_output.c:1360)
==2248==    by 0x50A8085: Thread (video_output.c:1499)
==2248==    by 0x532AEFB: start_thread (pthread_create.c:304)
==2248==    by 0x582585C: clone (clone.S:112)
==2248== 
==2248== 


ff_prores_idct_put_10_sse4 () at libavcodec/x86/proresdsp.asm:428
428	idct_put_fn sse4, 16
(gdb) bt
#0  ff_prores_idct_put_10_sse4 () at libavcodec/x86/proresdsp.asm:428
#1  0x0000000014453bb1 in decode_slice_plane (ctx=<optimized out>, td=0x2e00, buf=<optimized out>, data_size=<optimized out>, 
    out_ptr=0x96c2290, linesize=1472, mbs_per_slice=8, blocks_per_mb=2, plane_size_factor=4, qmat=0x66e1270)
    at libavcodec/proresdec.c:492
#2  0x0000000014454040 in decode_slice (avctx=<optimized out>, td=0xc328f20) at libavcodec/proresdec.c:587
#3  0x00000000144d690f in avcodec_default_execute (c=0x66df560, func=0x14453d10 <decode_slice>, arg=<optimized out>, 
    ret=<optimized out>, count=112, size=4128) at libavcodec/utils.c:568
#4  0x000000001445315d in decode_picture (avctx=0x66df560, pic_num=1, ctx=0x66e0f20) at libavcodec/proresdec.c:624
#5  decode_frame (avctx=0x66df560, data=0x66dff60, data_size=0x8383cec, avpkt=0x8383c90) at libavcodec/proresdec.c:669
#6  0x00000000144d7278 in avcodec_decode_video2 (avctx=0x66df560, picture=0x66dff60, got_picture_ptr=0x8383cec, 
    avpkt=0x8383c90) at libavcodec/utils.c:866
#7  0x0000000013eaef71 in DecodeVideo (p_dec=0x6a18628, pp_block=<optimized out>)
    at ../../../../vlc/modules/codec/avcodec/video.c:590
#8  0x0000000005085967 in DecoderDecodeVideo (p_dec=0x6a18628, p_block=0x7291190) at ../../vlc/src/input/decoder.c:1512
#9  0x0000000005085680 in DecoderProcessVideo (b_flush=false, p_block=0x7291190, p_dec=0x6a18628)
    at ../../vlc/src/input/decoder.c:1878
#10 DecoderProcess (p_dec=0x6a18628, p_block=<optimized out>) at ../../vlc/src/input/decoder.c:2052
#11 0x000000000508577b in DecoderThread (p_data=0x6a18628) at ../../vlc/src/input/decoder.c:938
#12 0x000000000532aefc in start_thread (arg=0x8384700) at pthread_create.c:304
#13 0x000000000582585d in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:112
#14 0x0000000000000000 in ?? ()
(gdb) 

The invalid write happens when decoding V chroma plane, where we supposedly overflows the picture buffer.

> 
> > > > pointers welcome
> > >  I am a bit dubious about the fact that the proress decoder in ffmpeg doesn't
> > > seem to check the value of CODEC_FLAG_EMU_EDGE. (h264 decoder does for
> > > example).
> > 
> > I thought that perhaps edges are drawn by some helper function because
> > only a handful of decoders check this flag, or perhaps they are just
> > not drawn at all.
> > 
> > Why do we need libavcodec to draw the edges for us anyway?
>  It's not that we require the edges, it's because we cannot provide a
> buffer suitable for avcodec if we don't ask for EMU_EDGE (or if we could,
> we would probably end up doing a copy later anyway, at least for playback).
> 



-- 
Rafa?l Carr?


From edward.c.wang at compdigitec.com  Sun Dec 11 22:34:21 2011
From: edward.c.wang at compdigitec.com (Edward Wang)
Date: Sun, 11 Dec 2011 16:34:21 -0500
Subject: [vlc-devel] [PATCH] Qt4: use QWizard for sout (stream out) dialog
	(#5651)
Message-ID: <CAHqHa+JPacAOkvUFnD_=mUxseqJp9-kLAYmZBhj9u3LDHamFYw@mail.gmail.com>

Hi all:

This patch will add the new feature in bug #5651 (
http://trac.videolan.org/vlc/ticket/5651 - Qt4: use QWizard for sout
(stream out) dialog). Attached is the patch file.

Thanks,
Edward Wang
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111211/511f48f4/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: patch5651.diff
Type: text/x-diff
Size: 24062 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111211/511f48f4/attachment.diff>

From nkoriyama at gmail.com  Sun Dec 11 22:57:43 2011
From: nkoriyama at gmail.com (Naohiro KORIYAMA)
Date: Mon, 12 Dec 2011 06:57:43 +0900
Subject: [vlc-devel] [PATCH] Fix for cue sheet problem on Windows
Message-ID: <CAOa_K+urATGuH-ExQDaXZAWR_n5_C1huTnUMCJ4CYMdTdysxfA@mail.gmail.com>

Here is a patch for cue sheet problem on Windows.

On windows nightly build, cue sheet doesn't work at all.
I tested the cue sheets placed on local path
(D:\tmp\cue_problem\CDImageApe.cue) and UNC Path
(\\NADIA\backup\backup\cue_problem\CDImageApe.cue).
VLC can't process properly both of them.

Detail log is pasted at:
http://pastebin.com/LeSk9prR

-- 
KORIYAMA, Naohiro
nkoriyama at gmail.com
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-Fix-for-CUE-Sheet-problem-on-Windows.patch
Type: application/octet-stream
Size: 1422 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111212/1865cc5a/attachment.obj>

From fenrir at elivagar.org  Sun Dec 11 23:44:33 2011
From: fenrir at elivagar.org (Laurent Aimar)
Date: Sun, 11 Dec 2011 23:44:33 +0100
Subject: [vlc-devel] [vlc-commits] avcodec: fix direct rendering
	with	some codecs
In-Reply-To: <20111211155033.4a89d8ef@tok>
References: <20111210220007.D76BA145379@albiero.videolan.org>
	<20111210224322.GA3494@elivagar.org> <20111210180721.44dc660a@tok>
	<20111210232954.GA5925@elivagar.org> <20111211010333.2df5b35c@tok>
	<20111211102529.GA6144@elivagar.org> <20111211144755.7aa57f9d@tok>
	<20111211201307.GA3912@elivagar.org> <20111211155033.4a89d8ef@tok>
Message-ID: <20111211224433.GA11657@elivagar.org>

On Sun, Dec 11, 2011 at 03:50:33PM -0500, Rafa??l Carr?? wrote:
> So chroma is I2AL
> 
> The picture is here:
> 	width/height = visible width/height = 720x486
> For the 3 planes though:
> 	width/height = 736x498
> 	visible width/height = 720x486
That's normal.
For the width:
 U/V -> 720/2 -> 360 -> 368 (mod 16)
 Y -> 2 * U/V (I422? -> 2*368 -> 736 (and already mod 16)

For the height:
 Y/U/V -> 486 -> 496 (mod 16)
and there is 2 extra hardcoded lines to ensure that the pictures will
be compatible with avcodec_align_dimensions() for H264 (otherwise
dr will be disabled by ffmpeg_GetFrameBuf()).

> 
> [0x6a18628] avcodec decoder debug: using direct rendering
> ==2248== Invalid write of size 8
> ==2248==    at 0x145B00A8: ??? (proresdsp.asm:428)
> ==2248==    by 0x14453BB0: decode_slice_plane (proresdec.c:492)
> ==2248==    by 0x1445403F: decode_slice (proresdec.c:587)
> ==2248==    by 0x144D690E: avcodec_default_execute (utils.c:568)
> ==2248==    by 0x1445315C: decode_frame (proresdec.c:624)
> ==2248==    by 0x144D7277: avcodec_decode_video2 (utils.c:866)
> ==2248==    by 0x13EAEF70: DecodeVideo (video.c:590)
> ==2248==    by 0x5085966: DecoderDecodeVideo (decoder.c:1512)
> ==2248==    by 0x508567F: DecoderProcess (decoder.c:1878)
> ==2248==    by 0x508577A: DecoderThread (decoder.c:938)
> ==2248==    by 0x532AEFB: start_thread (pthread_create.c:304)
> ==2248==    by 0x582585C: clone (clone.S:112)
> ==2248==  Address 0x96c5650 is 0 bytes after a block of size 1,466,112 alloc'd
> ==2248==    at 0x4C278E8: memalign (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
> ==2248==    by 0x4C27997: posix_memalign (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
> ==2248==    by 0x50C05AF: picture_NewFromResource (vlc_common.h:935)
> ==2248==    by 0x50C1267: picture_pool_NewFromFormat (picture_pool.c:138)
> ==2248==    by 0x50AF3DE: vout_InitWrapper (vout_wrapper.c:153)
> ==2248==    by 0x50A78FB: ThreadStart (video_output.c:1360)
> ==2248==    by 0x50A8085: Thread (video_output.c:1499)
> ==2248==    by 0x532AEFB: start_thread (pthread_create.c:304)
> ==2248==    by 0x582585C: clone (clone.S:112)
> ==2248== 
> ==2248== 
> 
> 
> ff_prores_idct_put_10_sse4 () at libavcodec/x86/proresdsp.asm:428
> 428	idct_put_fn sse4, 16
> (gdb) bt
> #0  ff_prores_idct_put_10_sse4 () at libavcodec/x86/proresdsp.asm:428
> #1  0x0000000014453bb1 in decode_slice_plane (ctx=<optimized out>, td=0x2e00, buf=<optimized out>, data_size=<optimized out>, 
>     out_ptr=0x96c2290, linesize=1472, mbs_per_slice=8, blocks_per_mb=2, plane_size_factor=4, qmat=0x66e1270)
>     at libavcodec/proresdec.c:492
> #2  0x0000000014454040 in decode_slice (avctx=<optimized out>, td=0xc328f20) at libavcodec/proresdec.c:587
> #3  0x00000000144d690f in avcodec_default_execute (c=0x66df560, func=0x14453d10 <decode_slice>, arg=<optimized out>, 
>     ret=<optimized out>, count=112, size=4128) at libavcodec/utils.c:568
> #4  0x000000001445315d in decode_picture (avctx=0x66df560, pic_num=1, ctx=0x66e0f20) at libavcodec/proresdec.c:624
> #5  decode_frame (avctx=0x66df560, data=0x66dff60, data_size=0x8383cec, avpkt=0x8383c90) at libavcodec/proresdec.c:669
> #6  0x00000000144d7278 in avcodec_decode_video2 (avctx=0x66df560, picture=0x66dff60, got_picture_ptr=0x8383cec, 
>     avpkt=0x8383c90) at libavcodec/utils.c:866
> #7  0x0000000013eaef71 in DecodeVideo (p_dec=0x6a18628, pp_block=<optimized out>)
>     at ../../../../vlc/modules/codec/avcodec/video.c:590
> #8  0x0000000005085967 in DecoderDecodeVideo (p_dec=0x6a18628, p_block=0x7291190) at ../../vlc/src/input/decoder.c:1512
> #9  0x0000000005085680 in DecoderProcessVideo (b_flush=false, p_block=0x7291190, p_dec=0x6a18628)
>     at ../../vlc/src/input/decoder.c:1878
> #10 DecoderProcess (p_dec=0x6a18628, p_block=<optimized out>) at ../../vlc/src/input/decoder.c:2052
> #11 0x000000000508577b in DecoderThread (p_data=0x6a18628) at ../../vlc/src/input/decoder.c:938
> #12 0x000000000532aefc in start_thread (arg=0x8384700) at pthread_create.c:304
> #13 0x000000000582585d in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:112
> #14 0x0000000000000000 in ?? ()
> (gdb) 

 Did yoy try the decoder from ffmpeg iof libav? Just in case. Also, did you
try disabling sse4 and see if the invalid write still happen?

-- 
fenrir



From jb at videolan.org  Sun Dec 11 23:52:08 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Sun, 11 Dec 2011 23:52:08 +0100
Subject: [vlc-devel] [PATCH] Fix for cue sheet problem on Windows
In-Reply-To: <CAOa_K+urATGuH-ExQDaXZAWR_n5_C1huTnUMCJ4CYMdTdysxfA@mail.gmail.com>
References: <CAOa_K+urATGuH-ExQDaXZAWR_n5_C1huTnUMCJ4CYMdTdysxfA@mail.gmail.com>
Message-ID: <20111211225208.GA28616@videolan.org>

On Mon, Dec 12, 2011 at 06:57:43AM +0900, Naohiro KORIYAMA wrote :
> Here is a patch for cue sheet problem on Windows.

Does it work on normal OSes?

Why the patch on src/ at the same time?

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From nkoriyama at gmail.com  Mon Dec 12 00:04:31 2011
From: nkoriyama at gmail.com (Naohiro KORIYAMA)
Date: Mon, 12 Dec 2011 08:04:31 +0900
Subject: [vlc-devel] [PATCH] Fix for cue sheet problem on Windows
In-Reply-To: <20111211225208.GA28616@videolan.org>
References: <CAOa_K+urATGuH-ExQDaXZAWR_n5_C1huTnUMCJ4CYMdTdysxfA@mail.gmail.com>
	<20111211225208.GA28616@videolan.org>
Message-ID: <CAOa_K+uB0bB5URT=LtoodFfHLX5Ga6qp2PZ5tb3e3qUK4SXjiA@mail.gmail.com>

2011/12/12 Jean-Baptiste Kempf <jb at videolan.org>:
> On Mon, Dec 12, 2011 at 06:57:43AM +0900, Naohiro KORIYAMA wrote :
>> Here is a patch for cue sheet problem on Windows.
>
> Does it work on normal OSes?

I check it on Windows and Mac OS X.
I don't have Linux environment.

> Why the patch on src/ at the same time?

A function vlc.string.make_uri  in lua scripts is a function
make_URI() in src/strings.c
And need to convert slash('/') to backslash('\').
If don't change this, VLC's executable path is added to the correct
path. (at line1140 in strings.c)

-- 
KORIYAMA, Naohiro
nkoriyama at gmail.com


From jb at videolan.org  Mon Dec 12 00:23:46 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Mon, 12 Dec 2011 00:23:46 +0100
Subject: [vlc-devel] [vlc-commits] Qt: do not explode when using OpenGL
	vout
In-Reply-To: <20111211232241.5CFA613F7DA@albiero.videolan.org>
References: <20111211232241.5CFA613F7DA@albiero.videolan.org>
Message-ID: <20111211232346.GB28616@videolan.org>

On Mon, Dec 12, 2011 at 12:22:41AM +0100, Jean-Baptiste Kempf wrote :
> vlc | branch: master | Jean-Baptiste Kempf <jb at videolan.org> | Mon Dec 12 00:22:04 2011 +0100| [f4c037efd2d3cde44daa1a58c7042fecc32b2177] | committer: Jean-Baptiste Kempf
> 
> Qt: do not explode when using OpenGL vout

DO NOT BACKPORT this, unless you have thoroughly tested this on Vista, 7
AND XP.

kthxbye

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From madmalibu at gmail.com  Mon Dec 12 01:13:44 2011
From: madmalibu at gmail.com (Mal Graty)
Date: Mon, 12 Dec 2011 00:13:44 +0000
Subject: [vlc-devel] [PATCH] Playlist status not being updated on play/pause
Message-ID: <1323648824-31198-1-git-send-email-madmalibu@gmail.com>

A fix for issue #4899

The problem was that the playing/paused state was being misreported
to the luahttp interface, turns out this was due to the way the playing
state was being updated in the Qt and hotkey logic.

The p_input state was being updated, but the playlist wasn't being
informed, this patch switches out the p_state calls to use the
equivalent playlist aware functions.
---
 modules/control/hotkeys.c         |   13 ++-----------
 modules/gui/qt4/input_manager.cpp |   17 +++--------------
 modules/gui/qt4/input_manager.hpp |    1 -
 3 files changed, 5 insertions(+), 26 deletions(-)

-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-Playlist-status-not-being-updated-on-play-pause.patch
Type: text/x-patch
Size: 2647 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111212/798e0429/attachment.bin>

From funman at videolan.org  Mon Dec 12 01:21:01 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Sun, 11 Dec 2011 19:21:01 -0500
Subject: [vlc-devel] [vlc-commits] avcodec: fix direct rendering with
 some codecs
In-Reply-To: <20111211224433.GA11657@elivagar.org>
References: <20111210220007.D76BA145379@albiero.videolan.org>
	<20111210224322.GA3494@elivagar.org> <20111210180721.44dc660a@tok>
	<20111210232954.GA5925@elivagar.org> <20111211010333.2df5b35c@tok>
	<20111211102529.GA6144@elivagar.org> <20111211144755.7aa57f9d@tok>
	<20111211201307.GA3912@elivagar.org> <20111211155033.4a89d8ef@tok>
	<20111211224433.GA11657@elivagar.org>
Message-ID: <20111211192101.1514e7e5@tok>

Le Sun, 11 Dec 2011 23:44:33 +0100,
Laurent Aimar <fenrir at elivagar.org> a ??crit :

> On Sun, Dec 11, 2011 at 03:50:33PM -0500, Rafa??l Carr?? wrote:
> > So chroma is I2AL
> > 
> > The picture is here:
> > 	width/height = visible width/height = 720x486
> > For the 3 planes though:
> > 	width/height = 736x498
> > 	visible width/height = 720x486
> That's normal.
> For the width:
>  U/V -> 720/2 -> 360 -> 368 (mod 16)
>  Y -> 2 * U/V (I422? -> 2*368 -> 736 (and already mod 16)
> 
> For the height:
>  Y/U/V -> 486 -> 496 (mod 16)
> and there is 2 extra hardcoded lines to ensure that the pictures will
> be compatible with avcodec_align_dimensions() for H264 (otherwise
> dr will be disabled by ffmpeg_GetFrameBuf()).
> 
> > 
> > [0x6a18628] avcodec decoder debug: using direct rendering
> > ==2248== Invalid write of size 8
> > ==2248==    at 0x145B00A8: ??? (proresdsp.asm:428)
> > ==2248==    by 0x14453BB0: decode_slice_plane (proresdec.c:492)
> > ==2248==    by 0x1445403F: decode_slice (proresdec.c:587)
> > ==2248==    by 0x144D690E: avcodec_default_execute (utils.c:568)
> > ==2248==    by 0x1445315C: decode_frame (proresdec.c:624)
> > ==2248==    by 0x144D7277: avcodec_decode_video2 (utils.c:866)
> > ==2248==    by 0x13EAEF70: DecodeVideo (video.c:590)
> > ==2248==    by 0x5085966: DecoderDecodeVideo (decoder.c:1512)
> > ==2248==    by 0x508567F: DecoderProcess (decoder.c:1878)
> > ==2248==    by 0x508577A: DecoderThread (decoder.c:938)
> > ==2248==    by 0x532AEFB: start_thread (pthread_create.c:304)
> > ==2248==    by 0x582585C: clone (clone.S:112)
> > ==2248==  Address 0x96c5650 is 0 bytes after a block of size 1,466,112 alloc'd
> > ==2248==    at 0x4C278E8: memalign (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
> > ==2248==    by 0x4C27997: posix_memalign (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
> > ==2248==    by 0x50C05AF: picture_NewFromResource (vlc_common.h:935)
> > ==2248==    by 0x50C1267: picture_pool_NewFromFormat (picture_pool.c:138)
> > ==2248==    by 0x50AF3DE: vout_InitWrapper (vout_wrapper.c:153)
> > ==2248==    by 0x50A78FB: ThreadStart (video_output.c:1360)
> > ==2248==    by 0x50A8085: Thread (video_output.c:1499)
> > ==2248==    by 0x532AEFB: start_thread (pthread_create.c:304)
> > ==2248==    by 0x582585C: clone (clone.S:112)
> > ==2248== 
> > ==2248== 
> > 
> > 
> > ff_prores_idct_put_10_sse4 () at libavcodec/x86/proresdsp.asm:428
> > 428	idct_put_fn sse4, 16
> > (gdb) bt
> > #0  ff_prores_idct_put_10_sse4 () at libavcodec/x86/proresdsp.asm:428
> > #1  0x0000000014453bb1 in decode_slice_plane (ctx=<optimized out>, td=0x2e00, buf=<optimized out>, data_size=<optimized out>, 
> >     out_ptr=0x96c2290, linesize=1472, mbs_per_slice=8, blocks_per_mb=2, plane_size_factor=4, qmat=0x66e1270)
> >     at libavcodec/proresdec.c:492
> > #2  0x0000000014454040 in decode_slice (avctx=<optimized out>, td=0xc328f20) at libavcodec/proresdec.c:587
> > #3  0x00000000144d690f in avcodec_default_execute (c=0x66df560, func=0x14453d10 <decode_slice>, arg=<optimized out>, 
> >     ret=<optimized out>, count=112, size=4128) at libavcodec/utils.c:568
> > #4  0x000000001445315d in decode_picture (avctx=0x66df560, pic_num=1, ctx=0x66e0f20) at libavcodec/proresdec.c:624
> > #5  decode_frame (avctx=0x66df560, data=0x66dff60, data_size=0x8383cec, avpkt=0x8383c90) at libavcodec/proresdec.c:669
> > #6  0x00000000144d7278 in avcodec_decode_video2 (avctx=0x66df560, picture=0x66dff60, got_picture_ptr=0x8383cec, 
> >     avpkt=0x8383c90) at libavcodec/utils.c:866
> > #7  0x0000000013eaef71 in DecodeVideo (p_dec=0x6a18628, pp_block=<optimized out>)
> >     at ../../../../vlc/modules/codec/avcodec/video.c:590
> > #8  0x0000000005085967 in DecoderDecodeVideo (p_dec=0x6a18628, p_block=0x7291190) at ../../vlc/src/input/decoder.c:1512
> > #9  0x0000000005085680 in DecoderProcessVideo (b_flush=false, p_block=0x7291190, p_dec=0x6a18628)
> >     at ../../vlc/src/input/decoder.c:1878
> > #10 DecoderProcess (p_dec=0x6a18628, p_block=<optimized out>) at ../../vlc/src/input/decoder.c:2052
> > #11 0x000000000508577b in DecoderThread (p_data=0x6a18628) at ../../vlc/src/input/decoder.c:938
> > #12 0x000000000532aefc in start_thread (arg=0x8384700) at pthread_create.c:304
> > #13 0x000000000582585d in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:112
> > #14 0x0000000000000000 in ?? ()
> > (gdb) 
> 
>  Did yoy try the decoder from ffmpeg iof libav? Just in case. 

> Also, did you
> try disabling sse4 and see if the invalid write still happen?


==25067== Invalid write of size 2
==25067==    at 0x15FC7790: put_pixels (proresdsp.c:43)
==25067==    by 0x15FC77F6: prores_idct_put_c (proresdsp.c:51)
==25067==    by 0x15FC6AE4: decode_slice_plane (proresdec.c:492)
==25067==    by 0x15FC7054: decode_slice (proresdec.c:587)
==25067==    by 0x1605E399: avcodec_default_execute (utils.c:569)
==25067==    by 0x15FC71E3: decode_picture (proresdec.c:624)
==25067==    by 0x15FC7372: decode_frame (proresdec.c:669)
==25067==    by 0x1605F198: avcodec_decode_video2 (utils.c:867)
==25067==    by 0x15AAEF70: DecodeVideo (video.c:590)
==25067==    by 0x5085966: DecoderDecodeVideo (decoder.c:1512)
==25067==    by 0x508567F: DecoderProcess (decoder.c:1878)
==25067==    by 0x508577A: DecoderThread (decoder.c:938)
==25067==  Address 0xfd50460 is not stack'd, malloc'd or (recently)
free'd ==25067== 
==25067== 
==25067== ---- Attach to debugger ? --- [Return/N/n/Y/y/C/c] ---- 

Same bug with C instead of asm.


0x0000000015fc7790 in put_pixels (dst=0xfd4fea0, stride=736,
in=0xc8665a0) at libavcodec/proresdsp.c:43
43	            dst[dst_offset + x] =
CLIP_AND_BIAS(in[src_offset]); (gdb) print dst
$1 = (uint16_t *) 0xfd4fea0
(gdb) print dst_offset
$2 = 736
(gdb) print x
$3 = 0
(gdb) 
(gdb) frame 1
#1  0x0000000015fc77f7 in prores_idct_put_c (out=0xfd4fea0,
linesize=1472, block=0xc8665a0, qmat=0x69db6d0) at
libavcodec/proresdsp.c:51 51	    put_pixels(out, linesize >> 1,
block); (gdb) print out
$4 = (uint16_t *) 0xfd4fea0
(gdb) print linesize
$5 = 1472
(gdb) frame 2
#2  0x0000000015fc6ae5 in decode_slice_plane (ctx=0x69db380,
td=0xc866500, buf=0xca7620b
"\212_D\261,S\024\306?_\027\305\321te\031rH\n\210?R<#\242?8\212\222'I5$GZ%\254\212\243$\201\322&#\",\"\262/\n\220\342P\231%\311r`\322t\221:I\250Mbjvd\205&)jH)\022?3?o\026r\205-\231l\301>\026l\361b\343~qg\036au\364\245#_H\331\r\363\002\242_3\204O0\352J\230\245):R?ki\032\352\352\373jg\b\274X\250\063\023?\363\006\213\226\023\b\304\325\301\213\302G\374\016w\006/\006B=\300\066a\317\060\004",
<incomplete sequence \333>, data_size=167, out_ptr=0xfd4d0a0,
linesize=1472, mbs_per_slice=8, blocks_per_mb=2, plane_size_factor=4,
qmat=0x69db6d0) at libavcodec/proresdec.c:492 492
ctx->dsp.idct_put(out_ptr + linesize * 4,     linesize, block_ptr,
qmat); (gdb) print linesize $6 = 1472 (gdb) print mb_num
$7 = 0
(gdb) print mbs_per_slice
$8 = 8
(gdb) print blocks_per_mb
$9 = 2
(gdb) 
(gdb) frame 3
#3  0x0000000015fc7055 in decode_slice (avctx=0x699ebe0, td=0xc866500)
    at libavcodec/proresdec.c:587
587	    decode_slice_plane(ctx, td, buf + hdr_size + y_data_size
+ u_data_size, (gdb) print pic
$10 = (AVFrame *) 0x69db3d0
(gdb) print *pic
$11 = {data = {0xfbea280 "", 0xfc9d200 "", 0xfcf69c0 "", 0x0}, linesize
= { 1472, 736, 736, 0}, base = {0x0, 0x0, 0x0, 0x0}, key_frame = 1, 
  pict_type = 0, pts = -9223372036854775808, coded_picture_number = 0, 
  display_picture_number = 0, quality = 0, age = 1073741824, reference
= 0, qscale_table = 0x0, qstride = 0, mbskip_table = 0x0, motion_val =
{0x0, 0x0}, mb_type = 0x0, motion_subsample_log2 = 0 '\000', 
  opaque = 0xab67880, error = {0, 0, 0, 0}, type = 2, repeat_pict = 0, 
  qscale_type = 0, interlaced_frame = 1, top_field_first = 0, pan_scan
= 0x0, palette_has_changed = 0, buffer_hints = 0, dct_coeff = 0x0,
ref_index = { 0x0, 0x0}, reordered_opaque = 3, hwaccel_picture_private
= 0x0, pkt_pts = 0, pkt_dts = 0, owner = 0x0, thread_opaque = 0x0,
nb_samples = 0, extended_data = 0x0}
(gdb) 


V plane is at 0xfcf69c0
(gdb) frame 0
(gdb) print &dst[dst_offset+x]
$13 = (uint16_t *) 0xfd50460

invalid access is at V plane + 1472

BTW, pic is interlaced.

And FFmpeg doesn't have the bug.

avcodec_align_dimensions2() was changed when ProRes decoder was
committed in FFmpeg.
height is aligned on 32 pixels (= 2 blocks) for a bunch of codecs (to
account for interlacing), and ProRes is missing from the list in libav.

The side effect is that it disables direct rendering because the VLC
picture is too small, unless we align on 32 pixels too.

However I think in picture_Setup() we don't know the video codec, nor
if the stream is interlaced.

-- 
Rafa?l Carr?


From jb at videolan.org  Mon Dec 12 02:08:10 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Mon, 12 Dec 2011 02:08:10 +0100
Subject: [vlc-devel] [PATCH] Fix for cue sheet problem on Windows
In-Reply-To: <CAOa_K+uB0bB5URT=LtoodFfHLX5Ga6qp2PZ5tb3e3qUK4SXjiA@mail.gmail.com>
References: <CAOa_K+urATGuH-ExQDaXZAWR_n5_C1huTnUMCJ4CYMdTdysxfA@mail.gmail.com>
	<20111211225208.GA28616@videolan.org>
	<CAOa_K+uB0bB5URT=LtoodFfHLX5Ga6qp2PZ5tb3e3qUK4SXjiA@mail.gmail.com>
Message-ID: <20111212010809.GA22401@videolan.org>

On Mon, Dec 12, 2011 at 08:04:31AM +0900, Naohiro KORIYAMA wrote :
> 2011/12/12 Jean-Baptiste Kempf <jb at videolan.org>:
> > On Mon, Dec 12, 2011 at 06:57:43AM +0900, Naohiro KORIYAMA wrote :
> >> Here is a patch for cue sheet problem on Windows.
> >
> > Does it work on normal OSes?
> 
> I check it on Windows and Mac OS X.
> I don't have Linux environment.
> 
> > Why the patch on src/ at the same time?
> 
> A function vlc.string.make_uri  in lua scripts is a function
> make_URI() in src/strings.c
> And need to convert slash('/') to backslash('\').
> If don't change this, VLC's executable path is added to the correct
> path. (at line1140 in strings.c)

Changing make_URI now, probably won't be backported.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Mon Dec 12 02:28:46 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Mon, 12 Dec 2011 02:28:46 +0100
Subject: [vlc-devel] [PATCH] Playlist status not being updated on
 play/pause
In-Reply-To: <1323648824-31198-1-git-send-email-madmalibu@gmail.com>
References: <1323648824-31198-1-git-send-email-madmalibu@gmail.com>
Message-ID: <20111212012846.GA26809@videolan.org>

On Mon, Dec 12, 2011 at 12:13:44AM +0000, Mal Graty wrote :
> A fix for issue #4899

Thx.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From nkoriyama at gmail.com  Mon Dec 12 06:38:28 2011
From: nkoriyama at gmail.com (Naohiro KORIYAMA)
Date: Mon, 12 Dec 2011 14:38:28 +0900
Subject: [vlc-devel] [PATCH] Fix for cue sheet problem on Windows
In-Reply-To: <20111212010809.GA22401@videolan.org>
References: <CAOa_K+urATGuH-ExQDaXZAWR_n5_C1huTnUMCJ4CYMdTdysxfA@mail.gmail.com>
	<20111211225208.GA28616@videolan.org>
	<CAOa_K+uB0bB5URT=LtoodFfHLX5Ga6qp2PZ5tb3e3qUK4SXjiA@mail.gmail.com>
	<20111212010809.GA22401@videolan.org>
Message-ID: <CAOa_K+syW=G8gz=MGKBzzE-htDxB+YdN0uD03FOiMMJGF_VNAA@mail.gmail.com>

2011/12/12 Jean-Baptiste Kempf <jb at videolan.org>:
> On Mon, Dec 12, 2011 at 08:04:31AM +0900, Naohiro KORIYAMA wrote :
>> 2011/12/12 Jean-Baptiste Kempf <jb at videolan.org>:
>> > On Mon, Dec 12, 2011 at 06:57:43AM +0900, Naohiro KORIYAMA wrote :
>> >> Here is a patch for cue sheet problem on Windows.
>> >
>> > Does it work on normal OSes?
>>
>> I check it on Windows and Mac OS X.
>> I don't have Linux environment.
>>
>> > Why the patch on src/ at the same time?
>>
>> A function vlc.string.make_uri ?in lua scripts is a function
>> make_URI() in src/strings.c
>> And need to convert slash('/') to backslash('\').
>> If don't change this, VLC's executable path is added to the correct
>> path. (at line1140 in strings.c)
>
> Changing make_URI now, probably won't be backported.

Certainly, I'm worried that changing make_URI affect another functions.
If there is a way that not changing make_URI, it's better.

-- 
KORIYAMA, Naohiro
nkoriyama at gmail.com


From remi at remlab.net  Mon Dec 12 07:42:58 2011
From: remi at remlab.net (=?UTF-8?Q?R=C3=A9mi_Denis-Courmont?=)
Date: Mon, 12 Dec 2011 07:42:58 +0100
Subject: [vlc-devel] [PATCH] Fix for cue sheet problem on Windows
In-Reply-To: <CAOa_K+urATGuH-ExQDaXZAWR_n5_C1huTnUMCJ4CYMdTdysxfA@mail.gmail.com>
References: <CAOa_K+urATGuH-ExQDaXZAWR_n5_C1huTnUMCJ4CYMdTdysxfA@mail.gmail.com>
Message-ID: <b54c66065725be50adfad291d0d92e95@chewa.net>

On Mon, 12 Dec 2011 06:57:43 +0900, Naohiro KORIYAMA <nkoriyama at gmail.com>
wrote:
> Here is a patch for cue sheet problem on Windows.
> 
> On windows nightly build, cue sheet doesn't work at all.
> I tested the cue sheets placed on local path
> (D:\tmp\cue_problem\CDImageApe.cue) and UNC Path
> (\\NADIA\backup\backup\cue_problem\CDImageApe.cue).
> VLC can't process properly both of them.

Not only I don't understand why you need this change, but this patch
reintroduces a security problem whereby long URI cause VLC to crash.

Rejected.

-- 
R?mi Denis-Courmont
http://www.remlab.net/


From nkoriyama at gmail.com  Mon Dec 12 07:48:46 2011
From: nkoriyama at gmail.com (Naohiro KORIYAMA)
Date: Mon, 12 Dec 2011 15:48:46 +0900
Subject: [vlc-devel] [PATCH] Fix for cue sheet problem on Windows
In-Reply-To: <b54c66065725be50adfad291d0d92e95@chewa.net>
References: <CAOa_K+urATGuH-ExQDaXZAWR_n5_C1huTnUMCJ4CYMdTdysxfA@mail.gmail.com>
	<b54c66065725be50adfad291d0d92e95@chewa.net>
Message-ID: <CAOa_K+vuXJuXqdeW_ODQH7qVds+FKuY1-usLXC_KE-K0Sf5SXA@mail.gmail.com>

2011/12/12 R?mi Denis-Courmont <remi at remlab.net>:
> On Mon, 12 Dec 2011 06:57:43 +0900, Naohiro KORIYAMA <nkoriyama at gmail.com>
> wrote:
>> Here is a patch for cue sheet problem on Windows.
>>
>> On windows nightly build, cue sheet doesn't work at all.
>> I tested the cue sheets placed on local path
>> (D:\tmp\cue_problem\CDImageApe.cue) and UNC Path
>> (\\NADIA\backup\backup\cue_problem\CDImageApe.cue).
>> VLC can't process properly both of them.
>
> Not only I don't understand why you need this change, but this patch
> reintroduces a security problem whereby long URI cause VLC to crash.
>
> Rejected.

OK, I agreed.
I also think that's a not good idea.
-- 
KORIYAMA, Naohiro
nkoriyama at gmail.com


From remi at remlab.net  Mon Dec 12 07:51:56 2011
From: remi at remlab.net (=?UTF-8?Q?R=C3=A9mi_Denis-Courmont?=)
Date: Mon, 12 Dec 2011 07:51:56 +0100
Subject: [vlc-devel] [PATCH] Fix for cue sheet problem on Windows
In-Reply-To: <b54c66065725be50adfad291d0d92e95@chewa.net>
References: <CAOa_K+urATGuH-ExQDaXZAWR_n5_C1huTnUMCJ4CYMdTdysxfA@mail.gmail.com>
	<b54c66065725be50adfad291d0d92e95@chewa.net>
Message-ID: <cec7ec33035c5df6cfbf85fc70b96105@chewa.net>

On Mon, 12 Dec 2011 07:42:58 +0100, R?mi Denis-Courmont <remi at remlab.net>
wrote:
> On Mon, 12 Dec 2011 06:57:43 +0900, Naohiro KORIYAMA
<nkoriyama at gmail.com>
> wrote:
>> Here is a patch for cue sheet problem on Windows.
>> 
>> On windows nightly build, cue sheet doesn't work at all.
>> I tested the cue sheets placed on local path
>> (D:\tmp\cue_problem\CDImageApe.cue) and UNC Path
>> (\\NADIA\backup\backup\cue_problem\CDImageApe.cue).
>> VLC can't process properly both of them.
> 
> Not only I don't understand why you need this change, but this patch
> reintroduces a security problem whereby long URI cause VLC to crash.

Err... long paths, not long URIs.

-- 
R?mi Denis-Courmont
http://www.remlab.net/


From quyadong at hotmail.com  Mon Dec 12 08:10:02 2011
From: quyadong at hotmail.com (=?gb2312?B?x/rRx7ar?=)
Date: Mon, 12 Dec 2011 15:10:02 +0800
Subject: [vlc-devel] How to set PTS
Message-ID: <BLU163-W22E4BFDECC200FBA7C5017D1BC0@phx.gbl>


hi all,
 
Recently I try to write an access-demux module to parse h264 stream from a local camera. However, Runthread function can't get frame. I stepped into function decodeVideo in modules/codec/avcodec/video.c and found that picture have been dropped because p_sys->i_pts is 0. So I set 400000 to the first frame,  800000 to the second frame, ... and so on, but vlc just display the first frame and don't enter function demux and decodevideo again. I had read ts.c and found PCRHandle function to deal with the similar thing, but I can't understand how to transfer pts into decode or display. Would you like to give me some hint to deal with this issue? 
  		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111212/050e3eb3/attachment.html>

From fenrir at elivagar.org  Mon Dec 12 10:48:54 2011
From: fenrir at elivagar.org (Laurent Aimar)
Date: Mon, 12 Dec 2011 10:48:54 +0100
Subject: [vlc-devel] [vlc-commits] avcodec: fix direct rendering
	with	some codecs
In-Reply-To: <20111211192101.1514e7e5@tok>
References: <20111210224322.GA3494@elivagar.org> <20111210180721.44dc660a@tok>
	<20111210232954.GA5925@elivagar.org> <20111211010333.2df5b35c@tok>
	<20111211102529.GA6144@elivagar.org> <20111211144755.7aa57f9d@tok>
	<20111211201307.GA3912@elivagar.org> <20111211155033.4a89d8ef@tok>
	<20111211224433.GA11657@elivagar.org> <20111211192101.1514e7e5@tok>
Message-ID: <20111212094854.GA13866@elivagar.org>

On Sun, Dec 11, 2011 at 07:21:01PM -0500, Rafa?l Carr? wrote:
> invalid access is at V plane + 1472
> 
> BTW, pic is interlaced.
> 
> And FFmpeg doesn't have the bug.
> avcodec_align_dimensions2() was changed when ProRes decoder was
> committed in FFmpeg.
> height is aligned on 32 pixels (= 2 blocks) for a bunch of codecs (to
> account for interlacing), and ProRes is missing from the list in libav.
 Ok, that's explain it.

> The side effect is that it disables direct rendering because the VLC
> picture is too small, unless we align on 32 pixels too.
 I will take care of that.

-- 
fenrir


From hoang.qc at anlab.biz  Mon Dec 12 11:49:10 2011
From: hoang.qc at anlab.biz (Hoang QC)
Date: Mon, 12 Dec 2011 17:49:10 +0700
Subject: [vlc-devel] VLC play video error question
Message-ID: <CAEG_UaiE58vRRzP1GChe1ppe3LjMUEdB+VpgayLKrn2U+oOQHw@mail.gmail.com>

Hi everyone!

I am not sure this is the right mail-list to post my question. If not
please forgive me.

I am writing a program to merge rtp packet into a mp4 file. The result is
uploaded at this link: *http://mediafire.com/?2aq53f9yxdqobtz*

If I play it by "Media Player Classic" 1.5.3.3752, everything is OK, but
when I play it with VLC 1.1.11, there is something wrong with the video.
There is a transparent blue above the image. If I used the deinterlacing
options, the video looks better (the blue disappeared) but still not ok.

Could somebody point me what is wrong with my video file?

Thanks!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111212/cb2a353b/attachment.html>

From funman at videolan.org  Mon Dec 12 18:20:14 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Mon, 12 Dec 2011 12:20:14 -0500
Subject: [vlc-devel] How to set PTS
In-Reply-To: <BLU163-W22E4BFDECC200FBA7C5017D1BC0@phx.gbl>
References: <BLU163-W22E4BFDECC200FBA7C5017D1BC0@phx.gbl>
Message-ID: <20111212122014.667333a9@tok>

Le Mon, 12 Dec 2011 15:10:02 +0800,
??? <quyadong at hotmail.com> a ?crit :

> 
> hi all,
>  
> Recently I try to write an access-demux module to parse h264 stream from a local camera. However, Runthread function can't get frame. I stepped into function decodeVideo in modules/codec/avcodec/video.c and found that picture have been dropped because p_sys->i_pts is 0. So I set 400000 to the first frame,  800000 to the second frame, ... and so on, but vlc just display the first frame and don't enter function demux and decodevideo again. I had read ts.c and found PCRHandle function to deal with the similar thing, but I can't understand how to transfer pts into decode or display. Would you like to give me some hint to deal with this issue? 

try --demux avformat

-- 
Rafa?l Carr?


From kaarlo.raiha at gmail.com  Mon Dec 12 20:54:18 2011
From: kaarlo.raiha at gmail.com (=?ISO-8859-1?B?S2FhcmxvIFLkaWjk?=)
Date: Mon, 12 Dec 2011 21:54:18 +0200
Subject: [vlc-devel] VLC play video error question
In-Reply-To: <CAEG_UaiE58vRRzP1GChe1ppe3LjMUEdB+VpgayLKrn2U+oOQHw@mail.gmail.com>
References: <CAEG_UaiE58vRRzP1GChe1ppe3LjMUEdB+VpgayLKrn2U+oOQHw@mail.gmail.com>
Message-ID: <CAMKrNty3QjPwM_585k6Xtz58PGYS=76NDAbyujOzThPhhgMbsQ@mail.gmail.com>

2011/12/12 Hoang QC <hoang.qc at anlab.biz>

> Hi everyone!
>
> I am not sure this is the right mail-list to post my question. If not
> please forgive me.
>
> I am writing a program to merge rtp packet into a mp4 file. The result is
> uploaded at this link: *http://mediafire.com/?2aq53f9yxdqobtz*
>
> If I play it by "Media Player Classic" 1.5.3.3752, everything is OK, but
> when I play it with VLC 1.1.11, there is something wrong with the video.
> There is a transparent blue above the image. If I used the deinterlacing
> options, the video looks better (the blue disappeared) but still not ok.
>

You should test with VLC 1.2.0 nightly builds, since VLC 1.1.x series will
be out of support soon.


>
> Could somebody point me what is wrong with my video file?
>
> Thanks!
> _______________________________________________
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
> http://mailman.videolan.org/listinfo/vlc-devel
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111212/454533c4/attachment.html>

From fenrir at elivagar.org  Mon Dec 12 21:30:42 2011
From: fenrir at elivagar.org (Laurent Aimar)
Date: Mon, 12 Dec 2011 21:30:42 +0100
Subject: [vlc-devel] [vlc-commits] Revert "Disable direct-rendering on
	ProRes"
In-Reply-To: <20111212201743.1995B1419A9@albiero.videolan.org>
References: <20111212201743.1995B1419A9@albiero.videolan.org>
Message-ID: <20111212203042.GA14726@elivagar.org>

On Mon, Dec 12, 2011 at 09:17:43PM +0100, Rafa?l Carr? wrote:
> vlc | branch: master | Rafa?l Carr? <funman at videolan.org> | Mon Dec 12 15:17:06 2011 -0500| [0d1ba56a9adabefa36dab7a107a708dec5c459da] | committer: Rafa?l Carr?
> 
> Revert "Disable direct-rendering on ProRes"
> 
> This reverts commit 8b914fcd30153d05b04921868166e5bd43370d61.
> 
> The problem was fixed in bc3b85c6bd6dc0094562561b084dcea3f0364743.
That's not exactly true. It's more that a libav issue was indirectly
workaround by bc3b85c6bd6dc0094562561b084dcea3f0364743 (this commit
was to increase the number of time that direct rendering was allowed).

-- 
fenrir


From jb at videolan.org  Mon Dec 12 22:07:19 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Mon, 12 Dec 2011 22:07:19 +0100
Subject: [vlc-devel] [vlc-commits] Revert "Disable direct-rendering on
 ProRes"
In-Reply-To: <20111212203042.GA14726@elivagar.org>
References: <20111212201743.1995B1419A9@albiero.videolan.org>
	<20111212203042.GA14726@elivagar.org>
Message-ID: <20111212210718.GB16377@videolan.org>

On Mon, Dec 12, 2011 at 09:30:42PM +0100, Laurent Aimar wrote :
> On Mon, Dec 12, 2011 at 09:17:43PM +0100, Rafa?l Carr? wrote:
> > vlc | branch: master | Rafa?l Carr? <funman at videolan.org> | Mon Dec 12 15:17:06 2011 -0500| [0d1ba56a9adabefa36dab7a107a708dec5c459da] | committer: Rafa?l Carr?
> > 
> > Revert "Disable direct-rendering on ProRes"
> > 
> > This reverts commit 8b914fcd30153d05b04921868166e5bd43370d61.
> > 
> > The problem was fixed in bc3b85c6bd6dc0094562561b084dcea3f0364743.
> That's not exactly true. It's more that a libav issue was indirectly
> workaround by bc3b85c6bd6dc0094562561b084dcea3f0364743 (this commit
> was to increase the number of time that direct rendering was allowed).

What should I do for 1.2? Let the --no-dr for PRORES or BP the 2
commits?

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From typx at dinauz.org  Mon Dec 12 22:08:43 2011
From: typx at dinauz.org (Denis Charmet)
Date: Mon, 12 Dec 2011 22:08:43 +0100
Subject: [vlc-devel] [PATCH 2/2] Fix subtitle loss at segment change.
In-Reply-To: <20110907231704.GA26659@dinauz.org>
References: <1315261489-22003-1-git-send-email-typx@dinauz.org>
	<1315261489-22003-3-git-send-email-typx@dinauz.org>
	<20110906192659.GA30525@elivagar.org>
	<20110907231704.GA26659@dinauz.org>
Message-ID: <20111212210843.GA30623@dinauz.org>

A little bump now that it's officialy a bug (#5625).

See :
http://mailman.videolan.org/pipermail/vlc-devel/2011-September/081731.html

Le jeudi 08 septembre 2011 ? 01:17:04, Denis Charmet a ?crit :
> Le mardi 06 septembre 2011 ? 09:26:59, Laurent Aimar a ?crit :
> >  I don't think that is valid sadly. The ES_OUT_SET_ES_DEFAULT is just here to
> > tell which track should be selected by default but should not override any user
> > choice.
> > 
> >  Could you describe why you need this change ?
> >
> 
> Afaik EsOutSelect doesn't override any user choice. The default is the
> last condition applied.
> The fact is when at segment change we add new ES and the default is
> totally ignored wether the user chose to change it or not.
> EsOutSelect is call at es_out_Add so before we can set the track as
> default.
> 
> -- 
> TypX
> Le mauvais esprit est un art de vivre
> _______________________________________________
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
> http://mailman.videolan.org/listinfo/vlc-devel

-- 
TypX
Le mauvais esprit est un art de vivre


From fenrir at elivagar.org  Mon Dec 12 22:20:09 2011
From: fenrir at elivagar.org (Laurent Aimar)
Date: Mon, 12 Dec 2011 22:20:09 +0100
Subject: [vlc-devel] [vlc-commits] Revert "Disable direct-rendering
	on	ProRes"
In-Reply-To: <20111212210718.GB16377@videolan.org>
References: <20111212201743.1995B1419A9@albiero.videolan.org>
	<20111212203042.GA14726@elivagar.org>
	<20111212210718.GB16377@videolan.org>
Message-ID: <20111212212009.GA17425@elivagar.org>

On Mon, Dec 12, 2011 at 10:07:19PM +0100, Jean-Baptiste Kempf wrote:
> On Mon, Dec 12, 2011 at 09:30:42PM +0100, Laurent Aimar wrote :
> > On Mon, Dec 12, 2011 at 09:17:43PM +0100, Rafa?l Carr? wrote:
> > > vlc | branch: master | Rafa?l Carr? <funman at videolan.org> | Mon Dec 12 15:17:06 2011 -0500| [0d1ba56a9adabefa36dab7a107a708dec5c459da] | committer: Rafa?l Carr?
> > > 
> > > Revert "Disable direct-rendering on ProRes"
> > > 
> > > This reverts commit 8b914fcd30153d05b04921868166e5bd43370d61.
> > > 
> > > The problem was fixed in bc3b85c6bd6dc0094562561b084dcea3f0364743.
> > That's not exactly true. It's more that a libav issue was indirectly
> > workaround by bc3b85c6bd6dc0094562561b084dcea3f0364743 (this commit
> > was to increase the number of time that direct rendering was allowed).
> 
> What should I do for 1.2? Let the --no-dr for PRORES or BP the 2
> commits?
 I think backporting the 2 is the best solution.

-- 
fenrir


From jb at videolan.org  Mon Dec 12 23:05:47 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Mon, 12 Dec 2011 23:05:47 +0100
Subject: [vlc-devel] [vlc-commits] Revert "Disable direct-rendering on
 ProRes"
In-Reply-To: <20111212212009.GA17425@elivagar.org>
References: <20111212201743.1995B1419A9@albiero.videolan.org>
	<20111212203042.GA14726@elivagar.org>
	<20111212210718.GB16377@videolan.org>
	<20111212212009.GA17425@elivagar.org>
Message-ID: <20111212220547.GA4484@videolan.org>

On Mon, Dec 12, 2011 at 10:20:09PM +0100, Laurent Aimar wrote :
>  I think backporting the 2 is the best solution.

Sir, yessir.


-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From quyadong at hotmail.com  Tue Dec 13 01:23:41 2011
From: quyadong at hotmail.com (Sheldon)
Date: Tue, 13 Dec 2011 00:23:41 +0000 (UTC)
Subject: [vlc-devel] How to set PTS
References: <BLU163-W22E4BFDECC200FBA7C5017D1BC0@phx.gbl>
	<20111212122014.667333a9@tok>
Message-ID: <loom.20111213T005440-703@post.gmane.org>

Rafa?l Carr? <funman <at> videolan.org> writes:

> 
> Le Mon, 12 Dec 2011 15:10:02 +0800,
> ??? <quyadong <at> hotmail.com> a ?crit :
> 
> > 
> > hi all,
> >  
> > Recently I try to write an access-demux module to parse h264 stream from a 
local camera. However,
> Runthread function can't get frame. I stepped into function decodeVideo in
> modules/codec/avcodec/video.c and found that picture have been dropped 
because p_sys->i_pts is 0. So I
> set 400000 to the first frame,  800000 to the second frame, ... and so on, 
but vlc just display the first frame
> and don't enter function demux and decodevideo again. I had read ts.c and 
found PCRHandle function to deal
> with the similar thing, but I can't understand how to transfer pts into 
decode or display. Would you like to
> give me some hint to deal with this issue? 
> 
> try --demux avformat
> 
In my case, if set p_block->i_pts in function Demux, vlc can display an frame 
correctly, Later RunThread always have picture, don't enter function 
DecodeVideo. If don't set this values, can enter DecodeVideo but drop all 
frames. Demux codes just like this:

static int Demux( demux_t *p_demux )
{
    demux_sys_t  *p_sys = p_demux->p_sys;
    block_t      *p_block;
    ssize_t       i_read = 0;

    p_block = block_New( p_demux, MAX_FRAME_BYTES );
    if( p_block == NULL )
        return 0;
    p_block->i_flags = BLOCK_FLAG_TYPE_P;

    // if no data, just wait.
    while( gi_framelength == 0 )
    {
        usleep( 5* 1000 );
    }

    /* copy data from an buffer. */
    if( pthread_mutex_lock( &g_mutex ) != 0 )
        return 0;
    memcpy( p_block->p_buffer, g_framebuffer, gi_framelength );
    p_block->i_buffer = i_read = gi_framelength;
    memset( g_framebuffer, 0, gi_framelength );
    gi_framelength = 0;
    pthread_mutex_unlock( &g_mutex );

   
#if 1
    if (p_sys->i_pts <= 0)
    {
        p_sys->i_pts = 400000;
    }
    else
    {
        p_sys->i_pts += 400000;
    }
    p_block->i_pts = p_sys->i_pts;
#endif
    msg_Err( p_demux, "%s read %d bytes pts %lld", __func__, i_read, p_block-
>i_pts );
    es_out_Send( p_demux->out, p_sys->p_es, p_block );
    if( i_read <= 0 )
    {
        block_Release( p_block );
        return 0;
    }
    return 1;
}






From quyadong at hotmail.com  Tue Dec 13 02:04:38 2011
From: quyadong at hotmail.com (Sheldon)
Date: Tue, 13 Dec 2011 01:04:38 +0000 (UTC)
Subject: [vlc-devel] How to set PTS
References: <BLU163-W22E4BFDECC200FBA7C5017D1BC0@phx.gbl>
	<20111212122014.667333a9@tok>
Message-ID: <loom.20111213T020414-536@post.gmane.org>

Rafa?l Carr? <funman <at> videolan.org> writes:

> 
> Le Mon, 12 Dec 2011 15:10:02 +0800,
> ??? <quyadong <at> hotmail.com> a ?crit :
> 
> > 
> > hi all,
> >  
> > Recently I try to write an access-demux module to parse h264 stream from a 
local camera. However,
> Runthread function can't get frame. I stepped into function decodeVideo in
> modules/codec/avcodec/video.c and found that picture have been dropped 
because p_sys->i_pts is 0. So I
> set 400000 to the first frame,  800000 to the second frame, ... and so on, 
but vlc just display the first frame
> and don't enter function demux and decodevideo again. I had read ts.c and 
found PCRHandle function to deal
> with the similar thing, but I can't understand how to transfer pts into 
decode or display. Would you like to
> give me some hint to deal with this issue? 
> 
> try --demux avformat
> 


It works. Thanks a lot.



From funman at videolan.org  Tue Dec 13 02:33:03 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Mon, 12 Dec 2011 20:33:03 -0500
Subject: [vlc-devel] How to set PTS
In-Reply-To: <loom.20111213T005440-703@post.gmane.org>
References: <BLU163-W22E4BFDECC200FBA7C5017D1BC0@phx.gbl>
	<20111212122014.667333a9@tok>
	<loom.20111213T005440-703@post.gmane.org>
Message-ID: <20111212203303.3c6f06a2@tok>

Le Tue, 13 Dec 2011 00:23:41 +0000 (UTC),
Sheldon <quyadong at hotmail.com> a ?crit :

> Rafa?l Carr? <funman <at> videolan.org> writes:

> > give me some hint to deal with this issue? 
> > 
> > try --demux avformat

Does --demux=avformat give the same result?

> In my case, if set p_block->i_pts in function Demux, vlc can display an frame 
> correctly

-- 
Rafa?l Carr?


From supermaster.ming at gmail.com  Tue Dec 13 07:52:47 2011
From: supermaster.ming at gmail.com (Ming@Supermaster)
Date: Tue, 13 Dec 2011 14:52:47 +0800
Subject: [vlc-devel] VLC play video error question
In-Reply-To: <CAMKrNty3QjPwM_585k6Xtz58PGYS=76NDAbyujOzThPhhgMbsQ@mail.gmail.com>
References: <CAEG_UaiE58vRRzP1GChe1ppe3LjMUEdB+VpgayLKrn2U+oOQHw@mail.gmail.com>
	<CAMKrNty3QjPwM_585k6Xtz58PGYS=76NDAbyujOzThPhhgMbsQ@mail.gmail.com>
Message-ID: <CAMLrna5TYHPAxowe-+8DysnSTdoY5x+VALmw=9Edhqwy2QzMrQ@mail.gmail.com>

I test it by libvlc.dll(1.3.0.0), it gives the following error in console
windows:

mmco: unref short failure
mmco: unref short failure
no frame!

I try to use ffmpeg to convert the file, same error message.
I hope this can help you~~

2011/12/13 Kaarlo R?ih? <kaarlo.raiha at gmail.com>

> 2011/12/12 Hoang QC <hoang.qc at anlab.biz>
>
>> Hi everyone!
>>
>> I am not sure this is the right mail-list to post my question. If not
>> please forgive me.
>>
>> I am writing a program to merge rtp packet into a mp4 file. The result is
>> uploaded at this link: *http://mediafire.com/?2aq53f9yxdqobtz*
>>
>> If I play it by "Media Player Classic" 1.5.3.3752, everything is OK, but
>> when I play it with VLC 1.1.11, there is something wrong with the video.
>> There is a transparent blue above the image. If I used the deinterlacing
>> options, the video looks better (the blue disappeared) but still not ok.
>>
>
> You should test with VLC 1.2.0 nightly builds, since VLC 1.1.x series will
> be out of support soon.
>
>
>>
>> Could somebody point me what is wrong with my video file?
>>
>> Thanks!
>> _______________________________________________
>> vlc-devel mailing list
>> To unsubscribe or modify your subscription options:
>> http://mailman.videolan.org/listinfo/vlc-devel
>>
>>
>
> _______________________________________________
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
> http://mailman.videolan.org/listinfo/vlc-devel
>
>


-- 
*????????*
http://www.supermaster.com.hk
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111213/e114c235/attachment.html>

From kaarlo.raiha at gmail.com  Tue Dec 13 10:55:47 2011
From: kaarlo.raiha at gmail.com (=?ISO-8859-1?B?S2FhcmxvIFLkaWjk?=)
Date: Tue, 13 Dec 2011 11:55:47 +0200
Subject: [vlc-devel] VLC play video error question
In-Reply-To: <CAMLrna5TYHPAxowe-+8DysnSTdoY5x+VALmw=9Edhqwy2QzMrQ@mail.gmail.com>
References: <CAEG_UaiE58vRRzP1GChe1ppe3LjMUEdB+VpgayLKrn2U+oOQHw@mail.gmail.com>
	<CAMKrNty3QjPwM_585k6Xtz58PGYS=76NDAbyujOzThPhhgMbsQ@mail.gmail.com>
	<CAMLrna5TYHPAxowe-+8DysnSTdoY5x+VALmw=9Edhqwy2QzMrQ@mail.gmail.com>
Message-ID: <CAMKrNtzSfwh36EyPDS_3+-Qpc5jzgu2kmHug1BSJNtDmjto_gA@mail.gmail.com>

2011/12/13 Ming at Supermaster <supermaster.ming at gmail.com>

> I test it by libvlc.dll(1.3.0.0), it gives the following error in console
> windows:
>
> mmco: unref short failure
> mmco: unref short failure
> no frame!
>
> I try to use ffmpeg to convert the file, same error message.
> I hope this can help you~~
>

http://ffmpeg.org/trac/ffmpeg/ticket/354 that should be fixed in newish
ffmpeg releases.


>
>
> 2011/12/13 Kaarlo R?ih? <kaarlo.raiha at gmail.com>
>
>> 2011/12/12 Hoang QC <hoang.qc at anlab.biz>
>>
>>> Hi everyone!
>>>
>>> I am not sure this is the right mail-list to post my question. If not
>>> please forgive me.
>>>
>>> I am writing a program to merge rtp packet into a mp4 file. The result
>>> is uploaded at this link: *http://mediafire.com/?2aq53f9yxdqobtz*
>>>
>>> If I play it by "Media Player Classic" 1.5.3.3752, everything is OK, but
>>> when I play it with VLC 1.1.11, there is something wrong with the video.
>>> There is a transparent blue above the image. If I used the deinterlacing
>>> options, the video looks better (the blue disappeared) but still not ok.
>>>
>>
>> You should test with VLC 1.2.0 nightly builds, since VLC 1.1.x series
>> will be out of support soon.
>>
>>
>>>
>>> Could somebody point me what is wrong with my video file?
>>>
>>> Thanks!
>>> _______________________________________________
>>> vlc-devel mailing list
>>> To unsubscribe or modify your subscription options:
>>> http://mailman.videolan.org/listinfo/vlc-devel
>>>
>>>
>>
>> _______________________________________________
>> vlc-devel mailing list
>> To unsubscribe or modify your subscription options:
>> http://mailman.videolan.org/listinfo/vlc-devel
>>
>>
>
>
> --
> *????????*
> http://www.supermaster.com.hk
>
> _______________________________________________
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
> http://mailman.videolan.org/listinfo/vlc-devel
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111213/34340224/attachment.html>

From nkoriyama at gmail.com  Tue Dec 13 16:06:59 2011
From: nkoriyama at gmail.com (Naohiro KORIYAMA)
Date: Wed, 14 Dec 2011 00:06:59 +0900
Subject: [vlc-devel] [PATCH] faad : Reinitializing decoder when we got
 Unexpected channel configuration change.
Message-ID: <CAOa_K+uWP0EoAr0Fschr-htnBjGF+BoE9a2PzgztptJeEG-Nng@mail.gmail.com>

Duraing playing some video including aac audio, once we got an error
"Unexpected channel configuration change",  same error occurs
afterwards, and we got no sound.

This patch fixes the aac audio issue.
There are 2 tickets about this issue. #5448, #3714

sample file:
http://dl.dropbox.com/u/1000972/Tim%27s%20No%20sound.mov
-- 
KORIYAMA, Naohiro
nkoriyama at gmail.com
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-Reinitialize-decoder-when-we-got-Unexpected-channel-.patch
Type: application/octet-stream
Size: 2672 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111214/56c13e56/attachment.obj>

From jb at videolan.org  Tue Dec 13 16:13:12 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Tue, 13 Dec 2011 16:13:12 +0100
Subject: [vlc-devel] [PATCH] faad : Reinitializing decoder when we got
 Unexpected channel configuration change.
In-Reply-To: <CAOa_K+uWP0EoAr0Fschr-htnBjGF+BoE9a2PzgztptJeEG-Nng@mail.gmail.com>
References: <CAOa_K+uWP0EoAr0Fschr-htnBjGF+BoE9a2PzgztptJeEG-Nng@mail.gmail.com>
Message-ID: <20111213151312.GA4595@videolan.org>

On Wed, Dec 14, 2011 at 12:06:59AM +0900, Naohiro KORIYAMA wrote :
> Duraing playing some video including aac audio, once we got an error
> "Unexpected channel configuration change",  same error occurs
> afterwards, and we got no sound.
> 
> This patch fixes the aac audio issue.
> There are 2 tickets about this issue. #5448, #3714

Cool.

Is this configuration retrieval/set thread-safe ?


Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From nkoriyama at gmail.com  Tue Dec 13 16:39:11 2011
From: nkoriyama at gmail.com (Naohiro KORIYAMA)
Date: Wed, 14 Dec 2011 00:39:11 +0900
Subject: [vlc-devel] [PATCH] faad : Reinitializing decoder when we got
 Unexpected channel configuration change.
In-Reply-To: <20111213151312.GA4595@videolan.org>
References: <CAOa_K+uWP0EoAr0Fschr-htnBjGF+BoE9a2PzgztptJeEG-Nng@mail.gmail.com>
	<20111213151312.GA4595@videolan.org>
Message-ID: <CAOa_K+uv5fLVXm6Duz-9wPcfvehB7bAdCv9uDTd1QT8AVPGCQQ@mail.gmail.com>

2011/12/14 Jean-Baptiste Kempf <jb at videolan.org>:
> On Wed, Dec 14, 2011 at 12:06:59AM +0900, Naohiro KORIYAMA wrote :
>> Duraing playing some video including aac audio, once we got an error
>> "Unexpected channel configuration change", ?same error occurs
>> afterwards, and we got no sound.
>>
>> This patch fixes the aac audio issue.
>> There are 2 tickets about this issue. #5448, #3714
>
> Cool.
>
> Is this configuration retrieval/set thread-safe ?


At least, I look at the faad2 source, Get/SetCurrentConfiguration:
- does not access global variables or the heap
- does not allocate/reallocate/free resources that have global scope
- does not use indirect accesses through handles or pointers
- does not have any visible side-effect

So,  I think they are thread safe.

Best regards,

-- 
KORIYAMA, Naohiro
nkoriyama at gmail.com


From nkoriyama at gmail.com  Tue Dec 13 16:44:29 2011
From: nkoriyama at gmail.com (Naohiro KORIYAMA)
Date: Wed, 14 Dec 2011 00:44:29 +0900
Subject: [vlc-devel] [PATCH] faad : Reinitializing decoder when we got
 Unexpected channel configuration change.
In-Reply-To: <CAOa_K+uv5fLVXm6Duz-9wPcfvehB7bAdCv9uDTd1QT8AVPGCQQ@mail.gmail.com>
References: <CAOa_K+uWP0EoAr0Fschr-htnBjGF+BoE9a2PzgztptJeEG-Nng@mail.gmail.com>
	<20111213151312.GA4595@videolan.org>
	<CAOa_K+uv5fLVXm6Duz-9wPcfvehB7bAdCv9uDTd1QT8AVPGCQQ@mail.gmail.com>
Message-ID: <CAOa_K+t5pxFBzJVPc9PBPNvw+GF42Q5c1VkeN0bC7UU_GKZMcA@mail.gmail.com>

2011/12/14 Naohiro KORIYAMA <nkoriyama at gmail.com>:
> 2011/12/14 Jean-Baptiste Kempf <jb at videolan.org>:
>> On Wed, Dec 14, 2011 at 12:06:59AM +0900, Naohiro KORIYAMA wrote :
>>> Duraing playing some video including aac audio, once we got an error
>>> "Unexpected channel configuration change", ?same error occurs
>>> afterwards, and we got no sound.
>>>
>>> This patch fixes the aac audio issue.
>>> There are 2 tickets about this issue. #5448, #3714
>>
>> Cool.
>>
>> Is this configuration retrieval/set thread-safe ?
>
>
> At least, I look at the faad2 source, Get/SetCurrentConfiguration:
> - does not access global variables or the heap
> - does not allocate/reallocate/free resources that have global scope
> - does not use indirect accesses through handles or pointers
> - does not have any visible side-effect
>
> So, ?I think they are thread safe.
>

I misunderstood what you said.
DecodeBlock() is threaded?


-- 
KORIYAMA, Naohiro
nkoriyama at gmail.com


From evgrafov.vladimir at gmail.com  Tue Dec 13 18:02:33 2011
From: evgrafov.vladimir at gmail.com (=?UTF-8?B?0JXQstCz0YDQsNGE0L7QsiDQktC70LDQtNC40LzQuNGA?=)
Date: Tue, 13 Dec 2011 21:02:33 +0400
Subject: [vlc-devel] VIDEO WEBSITE PARSER
Message-ID: <CAMn+YQKZJ8xpY6Ph7qu=Qfic1AhOHf3tgx0aB1LmOzuPeYv3Hg@mail.gmail.com>


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111213/eaa837a2/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: academicearth1.lua
Type: application/octet-stream
Size: 1590 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111213/eaa837a2/attachment.obj>

From dayanand at alethea.in  Tue Dec 13 09:03:27 2011
From: dayanand at alethea.in (Dayananda A)
Date: Tue, 13 Dec 2011 13:33:27 +0530
Subject: [vlc-devel] Need media descriptor access for a broadcast media
Message-ID: <CAKjtyvRWyy7rO5OjW5j8dfA5iyN39Px3bYFkrhSc2pAkEeXcmQ@mail.gmail.com>

Hi,

   I am trying to run a broadcast of a certain media using python bindings
for VLC. I have a requirement that I collect the media statistics on the
media that I currently streaming. To get media stats we need a media
descriptor and there is no native method in vlc.py to get the media
descriptor for a broadcast media. Could you please help me find an API that
could make me avail the media stats? That will be of great help. Thanks.


Regards,
-Daya.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111213/b5768bb6/attachment.html>

From var1able at var1able.ru  Tue Dec 13 18:02:57 2011
From: var1able at var1able.ru (=?utf-8?B?0JrQvtC90YHRgtCw0L3RgtC40L0g0JHQvtCz0LTQsNC90L7Qsg==?=)
Date: Tue, 13 Dec 2011 21:02:57 +0400
Subject: [vlc-devel] TuneWiki plugin
Message-ID: <641AEB2A-25BB-44BD-A763-408F2C32538D@var1able.ru>

http://www.google-melange.com/gci/task/view/google/gci2011/7131352
-------------- next part --------------
A non-text attachment was scrubbed...
Name: tunewiki.lua
Type: application/octet-stream
Size: 4520 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111213/42367753/attachment.obj>

From remi at remlab.net  Tue Dec 13 20:54:15 2011
From: remi at remlab.net (=?iso-8859-1?q?R=E9mi?= Denis-Courmont)
Date: Tue, 13 Dec 2011 21:54:15 +0200
Subject: [vlc-devel] [PATCH] faad : Reinitializing decoder when we got
	Unexpected channel configuration change.
In-Reply-To: <CAOa_K+t5pxFBzJVPc9PBPNvw+GF42Q5c1VkeN0bC7UU_GKZMcA@mail.gmail.com>
References: <CAOa_K+uWP0EoAr0Fschr-htnBjGF+BoE9a2PzgztptJeEG-Nng@mail.gmail.com>
	<CAOa_K+uv5fLVXm6Duz-9wPcfvehB7bAdCv9uDTd1QT8AVPGCQQ@mail.gmail.com>
	<CAOa_K+t5pxFBzJVPc9PBPNvw+GF42Q5c1VkeN0bC7UU_GKZMcA@mail.gmail.com>
Message-ID: <201112132154.18169.remi@remlab.net>

Le mardi 13 d?cembre 2011 17:44:29 Naohiro KORIYAMA, vous avez ?crit :
> I misunderstood what you said.
> DecodeBlock() is threaded?

No. DecodeBlock() is called repeatedly from a single thread.
This is true for all codecs.

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From remi at remlab.net  Tue Dec 13 20:54:43 2011
From: remi at remlab.net (=?iso-8859-15?q?R=E9mi?= Denis-Courmont)
Date: Tue, 13 Dec 2011 21:54:43 +0200
Subject: [vlc-devel] [PATCH] faad : Reinitializing decoder when we got
	Unexpected channel configuration change.
In-Reply-To: <CAOa_K+uWP0EoAr0Fschr-htnBjGF+BoE9a2PzgztptJeEG-Nng@mail.gmail.com>
References: <CAOa_K+uWP0EoAr0Fschr-htnBjGF+BoE9a2PzgztptJeEG-Nng@mail.gmail.com>
Message-ID: <201112132154.44146.remi@remlab.net>

Le mardi 13 d?cembre 2011 17:06:59 Naohiro KORIYAMA, vous avez ?crit :
> Duraing playing some video including aac audio, once we got an error
> "Unexpected channel configuration change",  same error occurs
> afterwards, and we got no sound.

I was asked to review this. I don't consider myself an expert on codecs. In 
any case, I see no problem, other than cosmetically overly long lines.

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From daniel.klischies at freenet.de  Tue Dec 13 19:28:00 2011
From: daniel.klischies at freenet.de (Daniel Klischies)
Date: Tue, 13 Dec 2011 19:28:00 +0100
Subject: [vlc-devel]  [PATCH] Installer improvements
Message-ID: <4EE79930.8030009@freenet.de>

Here are some improvements I made to the installer, readme.txt is
included to explain where to copy the NSIS directory.
To avoid confusion: It is not possible to use .jpegs in NSIS which is
why I had to include bitmaps. Their size should be relatively small in
the package (if LZMA compression is enabled).

I hope file-attachments are allowed here...

Best Regards
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: vlcsetup_improvements.patch
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111213/c67e8b18/attachment.ksh>

From beauze.h at gmail.com  Tue Dec 13 23:40:49 2011
From: beauze.h at gmail.com (=?UTF-8?Q?Hugo_Beauz=C3=A9e=2DLuyssen?=)
Date: Tue, 13 Dec 2011 23:40:49 +0100
Subject: [vlc-devel] [Patch] Dash: Don't try to do something without a
 valid MPD manager
In-Reply-To: <CAK-C75Rsjt92O5ho3jmrturf-ugdzHk38hEY6HMgyA4+mKTDeg@mail.gmail.com>
References: <CAK-C75TOrnUGWki6HDVk0iJ0_xE1ZO4Swd8FGAoCraupTXQW8Q@mail.gmail.com>
	<4ed3d2c2.c27f980a.2078.19a3SMTPIN_ADDED@mx.google.com>
	<CAK-C75QjQ05NLFG25mrJ7drozaSdXkv-gGzQFPsuSCerLQ5V7Q@mail.gmail.com>
	<20111128224245.GA29913@videolan.org>
	<CAK-C75Rsjt92O5ho3jmrturf-ugdzHk38hEY6HMgyA4+mKTDeg@mail.gmail.com>
Message-ID: <CAK-C75TdZ88HL0YQectcwh+FF6qO6nkRXRrM+w=iER9UR0vyoA@mail.gmail.com>

On Thu, Dec 1, 2011 at 12:43 PM, Hugo Beauz?e-Luyssen
<beauze.h at gmail.com> wrote:
> On Mon, Nov 28, 2011 at 11:42 PM, Jean-Baptiste Kempf <jb at videolan.org> wrote:
>> On Mon, Nov 28, 2011 at 10:54:07PM +0100, Hugo Beauz?e-Luyssen wrote :
>>> much, but feels more natural in my opinion.
>>
>> Natural and C++ ?
>>
>> pff
>
> Ping for review!
>
> Regards,
>

Hi,

This patch should solve this issue :
https://forum.videolan.org/viewtopic.php?f=13&t=96335&p=320156#p320156

Regards,


-- 
Hugo Beauz?e-Luyssen


From beauze.h at gmail.com  Tue Dec 13 23:46:32 2011
From: beauze.h at gmail.com (=?UTF-8?Q?Hugo_Beauz=C3=A9e=2DLuyssen?=)
Date: Tue, 13 Dec 2011 23:46:32 +0100
Subject: [vlc-devel] [Patch] Adding an helper to convert iso8601
 durations to seconds
In-Reply-To: <201112091942.47134.remi@remlab.net>
References: <CAK-C75R5BPUud6z2QxrFNNnh1wMFktu48O59oXi2bAUqZ0mYMw@mail.gmail.com>
	<CAK-C75RQNA-+iYR8wKs2LfFAE54oDOswZ_e-PXcpBJwe8W2zqg@mail.gmail.com>
	<CAK-C75T4x2x4cBjFwvz-FtSseLsV1EoHe=RvfsqUAqb70mf7Aw@mail.gmail.com>
	<201112091942.47134.remi@remlab.net>
Message-ID: <CAK-C75RtmcQd0m32+3d9ue7WZKH8v9n84izFWRp=ziwKm4yCEw@mail.gmail.com>

2011/12/9 R?mi Denis-Courmont <remi at remlab.net>:
> Function name is too long. And time_t is thecorrect type for seconds IMHO.
>

Here is a new version.

Best regards,

-- 
Hugo Beauz?e-Luyssen
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-strings-Adding-an-helper-to-convert-iso8601-duration.patch
Type: text/x-patch
Size: 3073 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111213/516aa720/attachment.bin>

From beauze.h at gmail.com  Tue Dec 13 23:48:41 2011
From: beauze.h at gmail.com (=?UTF-8?Q?Hugo_Beauz=C3=A9e=2DLuyssen?=)
Date: Tue, 13 Dec 2011 23:48:41 +0100
Subject: [vlc-devel] [Patch] Adding an helper to convert iso8601
 durations to seconds
In-Reply-To: <CAK-C75RtmcQd0m32+3d9ue7WZKH8v9n84izFWRp=ziwKm4yCEw@mail.gmail.com>
References: <CAK-C75R5BPUud6z2QxrFNNnh1wMFktu48O59oXi2bAUqZ0mYMw@mail.gmail.com>
	<CAK-C75RQNA-+iYR8wKs2LfFAE54oDOswZ_e-PXcpBJwe8W2zqg@mail.gmail.com>
	<CAK-C75T4x2x4cBjFwvz-FtSseLsV1EoHe=RvfsqUAqb70mf7Aw@mail.gmail.com>
	<201112091942.47134.remi@remlab.net>
	<CAK-C75RtmcQd0m32+3d9ue7WZKH8v9n84izFWRp=ziwKm4yCEw@mail.gmail.com>
Message-ID: <CAK-C75QD08hyZ4XNBR7nYkSFyLqR7Tm887Uxw7fWqpXRGJd=Dg@mail.gmail.com>

On Tue, Dec 13, 2011 at 11:46 PM, Hugo Beauz?e-Luyssen
<beauze.h at gmail.com> wrote:
> 2011/12/9 R?mi Denis-Courmont <remi at remlab.net>:
>> Function name is too long. And time_t is thecorrect type for seconds IMHO.
>>
>

Oops, this one's the good one.

Regards,


-- 
Hugo Beauz?e-Luyssen
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-strings-Adding-an-helper-to-convert-iso8601-duration.patch
Type: text/x-patch
Size: 3069 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111213/9975fe5a/attachment.bin>

From gotrunks at gmail.com  Wed Dec 14 01:59:38 2011
From: gotrunks at gmail.com (gotrunks at gmail.com)
Date: Wed, 14 Dec 2011 01:59:38 +0100
Subject: [vlc-devel] [PATCH] fix filename self-references
Message-ID: <CAMYtjArZ-km3cgzbOsV2BNLwYmRdH80A+JXe+Zpc=pBzxmVwQA@mail.gmail.com>

Hi,
patch attached.
Regards,--
Pere
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-fix-filename-self-references.patch
Type: text/x-patch
Size: 102372 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111214/271e5288/attachment.bin>

From RedWire at iinet.net.au  Wed Dec 14 02:20:38 2011
From: RedWire at iinet.net.au (Mark Phillips)
Date: Wed, 14 Dec 2011 12:20:38 +1100
Subject: [vlc-devel] [VLC] #5657: Mouse cursor does not disappear
	despite config settings V1.0.5 onwards
In-Reply-To: <062.b13d8d01a8a9497f92c810529070de4c@videolan.org>
References: <047.0587a377c0ab32669d466be1f67a7490@videolan.org>
	<062.b13d8d01a8a9497f92c810529070de4c@videolan.org>
Message-ID: <961C2146EE944460893356ED7CD47B53@thelook>

Thankyou for your response - although killing bittorent does not work I
think I have further isolated the problem to Matroska (mkv) encapsulated
files.

Probably fixed now and I believe problem solved by using VLC install
reversing back 5 versions and installing 1.1.11 from scratch. Weird, but
hell machines don't think, they just act.

Thanks and regards Mark


-----Original Message-----
From: VLC [mailto:trac at videolan.org] 
Sent: Tuesday, 13 December 2011 3:26 AM
To: undisclosed-recipients:
Subject: Re: [VLC] #5657: Mouse cursor does not disappear despite config
settings V1.0.5 onwards

#5657: Mouse cursor does not disappear despite config settings V1.0.5
onwards
----------------------------------------+----------------------------------
   Reporter:  Redwire                   |        Owner:  RedWire Consulting
       Type:  defect                    |       Status:  new
   Priority:  normal                    |    Milestone:  Bugs paradize
  Component:  Interface: Qt4            |      Version:  1.1.x git
   Severity:  normal                    |   Resolution:
   Keywords:  mouse cursor full screen  |   Difficulty:  hard
Platform(s):  Win32                     |  Work status:  Not started
----------------------------------------+----------------------------------
Changes (by jb):

 * component:  Unknown => Interface: Qt4


Comment:

 Well, please kill bittorrent and try again...

-- 
Ticket URL: <http://trac.videolan.org/vlc/ticket/5657#comment:1>
VLC <http://www.videolan.org/vlc/>
VLC media player



From funman at videolan.org  Wed Dec 14 07:43:34 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Wed, 14 Dec 2011 01:43:34 -0500
Subject: [vlc-devel] veda
In-Reply-To: <20111209175436.6f1dca99@tok>
References: <20111209175436.6f1dca99@tok>
Message-ID: <20111214014334.6af9ce0b@tok>

Le Fri, 9 Dec 2011 17:54:36 -0500,
Rafa?l Carr? <funman at videolan.org> a ?crit :

> Hi,
> 
> Can you add videolan to sudoers on veda?
> 
> Or give me root through other means, I need to tweak system for the buildbot.
> 
> Thanks

ping

btw Felix was speaking about moving veda at another location

-- 
Rafa?l Carr?


From nkoriyama at gmail.com  Wed Dec 14 07:48:54 2011
From: nkoriyama at gmail.com (Naohiro KORIYAMA)
Date: Wed, 14 Dec 2011 15:48:54 +0900
Subject: [vlc-devel] a pitfall when cross-compiling win32 build on 64bit
	machine
Message-ID: <CAOa_K+t1dmgxwxfZS6=FH-o2H1Y4kwNCwdBuMSNeVjXM4P0z8w@mail.gmail.com>

I compile win32 build on OS X Lion.

First, it seemed fine, but lua scripts didn't run at all.
I installed lua in /usr/local/bin, and it was built as 64bit application.
And, I found the following in the LUAC man page:
The binary files created by luac are portable only among architectures
with the same word size and byte order.

Oops, that's it.
After I made luac 32bit version, and use it by "LUAC=luac32
..../win32/configure.sh ...", lua scripts worked on win32 build.
I don't know about Linux 64bit environment, on
http://wiki.videolan.org/Win32Compile, there is no description about
it.

-- 
KORIYAMA, Naohiro
nkoriyama at gmail.com


From funman at videolan.org  Wed Dec 14 08:01:09 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Wed, 14 Dec 2011 02:01:09 -0500
Subject: [vlc-devel] a pitfall when cross-compiling win32 build on 64bit
 machine
In-Reply-To: <CAOa_K+t1dmgxwxfZS6=FH-o2H1Y4kwNCwdBuMSNeVjXM4P0z8w@mail.gmail.com>
References: <CAOa_K+t1dmgxwxfZS6=FH-o2H1Y4kwNCwdBuMSNeVjXM4P0z8w@mail.gmail.com>
Message-ID: <20111214020109.52f93878@tok>

Le Wed, 14 Dec 2011 15:48:54 +0900,
Naohiro KORIYAMA <nkoriyama at gmail.com> a ?crit :

> I compile win32 build on OS X Lion.
> 
> First, it seemed fine, but lua scripts didn't run at all.
> I installed lua in /usr/local/bin, and it was built as 64bit application.
> And, I found the following in the LUAC man page:
> The binary files created by luac are portable only among architectures
> with the same word size and byte order.
> 
> Oops, that's it.
> After I made luac 32bit version, and use it by "LUAC=luac32
> ..../win32/configure.sh ...", lua scripts worked on win32 build.
> I don't know about Linux 64bit environment, on
> http://wiki.videolan.org/Win32Compile, there is no description about
> it.

It's strange, I think the windows builds are made on a linux 64 bits
machine but they run fine on win32.

The .lua files (not compiled) should work fine, I think we used luac
only to reduce binary size of VLC package.

If that's the case, zip/7z compress the files already so there should
not be a big difference between .lua and .luac (except when unpacked on
the disk), and we could packge .lua files only.

-- 
Rafa?l Carr?


From nkoriyama at gmail.com  Wed Dec 14 08:49:13 2011
From: nkoriyama at gmail.com (Naohiro KORIYAMA)
Date: Wed, 14 Dec 2011 16:49:13 +0900
Subject: [vlc-devel] a pitfall when cross-compiling win32 build on 64bit
	machine
In-Reply-To: <20111214020109.52f93878@tok>
References: <CAOa_K+t1dmgxwxfZS6=FH-o2H1Y4kwNCwdBuMSNeVjXM4P0z8w@mail.gmail.com>
	<20111214020109.52f93878@tok>
Message-ID: <CAOa_K+v3VvKWiMy=dzaG0A8BHcxg=dVs_zFxY+XJG8rOEu+9gA@mail.gmail.com>

2011/12/14 Rafa?l Carr? <funman at videolan.org>:
> Le Wed, 14 Dec 2011 15:48:54 +0900,
> Naohiro KORIYAMA <nkoriyama at gmail.com> a ?crit :
>
>> I compile win32 build on OS X Lion.
>>
>> First, it seemed fine, but lua scripts didn't run at all.
>> I installed lua in /usr/local/bin, and it was built as 64bit application.
>> And, I found the following in the LUAC man page:
>> The binary files created by luac are portable only among architectures
>> with the same word size and byte order.
>>
>> Oops, that's it.
>> After I made luac 32bit version, and use it by "LUAC=luac32
>> ..../win32/configure.sh ...", lua scripts worked on win32 build.
>> I don't know about Linux 64bit environment, on
>> http://wiki.videolan.org/Win32Compile, there is no description about
>> it.
>
> It's strange, I think the windows builds are made on a linux 64 bits
> machine but they run fine on win32.

Header infomation in luac is different  32bit and 64bit, and liblua checks it.
liblua refuses when loading wrong luac like:

lua debug: Trying Lua playlist script D:\utils\vlc\lua\playlist\youtube.luac
lua warning: Error loading script
D:\utils\vlc\lua\playlist\youtube.luac:
D:\utils\vlc\lua\playlist\youtube.luac: bad header in precompiled
chunk


FYI, Compiled lua scripts in win32 NB are the same as one generated by
32bit luac, and in win64 NB, same as one generated by 64bit luac.
Maybe switching luac somehow...

Header Hex generated by luac 64bit that works on Mac OS X 64bit:
0000000: 1b4c 7561 5100 0104 0804 0800 5b00 0000  .LuaQ.......[...
0000010: 0000 0000 402f 5573 6572 732f 6e6b 6f72  ....@/Users/nkor

Header Hex generated by luac 32bit that works on Win32:
0000000: 1b4c 7561 5100 0104 0404 0800 5a00 0000  .LuaQ.......Z...
0000010: 402f 5573 6572 732f 6e6b 6f72 6979 616d  @/Users/nkoriyam

Header Hex in NB win 32bit:
0000000: 1b4c 7561 5100 0104 0404 0800 1a00 0000  .LuaQ...........
0000010: 406c 7561 2f70 6c61 796c 6973 742f 796f  @lua/playlist/yo

Header Hex in NB win 64bit:
0000000: 1b4c 7561 5100 0104 0804 0800 1a00 0000  .LuaQ...........
0000010: 0000 0000 406c 7561 2f70 6c61 796c 6973  .... at lua/playlis

> The .lua files (not compiled) should work fine, I think we used luac
> only to reduce binary size of VLC package.
>
> If that's the case, zip/7z compress the files already so there should
> not be a big difference between .lua and .luac (except when unpacked on
> the disk), and we could packge .lua files only.

If it's only my particular problem, please ignore it.

-- 
KORIYAMA, Naohiro
nkoriyama at gmail.com


From remi at remlab.net  Wed Dec 14 08:58:22 2011
From: remi at remlab.net (=?UTF-8?Q?R=C3=A9mi_Denis-Courmont?=)
Date: Wed, 14 Dec 2011 08:58:22 +0100
Subject: [vlc-devel] [vlc-commits] fix typo in french translation of nvlc
In-Reply-To: <20111214045035.CCC071447CA@albiero.videolan.org>
References: <20111214045035.CCC071447CA@albiero.videolan.org>
Message-ID: <223ba297b5536aae5067b5d2c3438557@chewa.net>

On Wed, 14 Dec 2011 05:50:35 +0100 (CET), git at videolan.org (Rafa?l Carr?)
wrote:
> vlc | branch: master | Rafa?l Carr? <funman at videolan.org> | Tue Dec 13
> 23:50:18 2011 -0500| [c52cfdd034558076705445510d3bbb95719f9704] |
> committer: Rafa?l Carr?
> 
> fix typo in french translation of nvlc

This will be overwritten when 1.2 translations are merged into 1.3...

-- 
R?mi Denis-Courmont
http://www.remlab.net/


From remi at remlab.net  Wed Dec 14 09:05:39 2011
From: remi at remlab.net (=?UTF-8?Q?R=C3=A9mi_Denis-Courmont?=)
Date: Wed, 14 Dec 2011 09:05:39 +0100
Subject: [vlc-devel] a pitfall when cross-compiling win32 build on 64bit
	machine
In-Reply-To: <CAOa_K+t1dmgxwxfZS6=FH-o2H1Y4kwNCwdBuMSNeVjXM4P0z8w@mail.gmail.com>
References: <CAOa_K+t1dmgxwxfZS6=FH-o2H1Y4kwNCwdBuMSNeVjXM4P0z8w@mail.gmail.com>
Message-ID: <b11d8e8953c48161d2ec0f614a324f7f@chewa.net>

   Naohiro-san hello,

On Wed, 14 Dec 2011 15:48:54 +0900, Naohiro KORIYAMA <nkoriyama at gmail.com>
wrote:
> I compile win32 build on OS X Lion.
> 
> First, it seemed fine, but lua scripts didn't run at all.
> I installed lua in /usr/local/bin, and it was built as 64bit
application.
> And, I found the following in the LUAC man page:
> The binary files created by luac are portable only among architectures
> with the same word size and byte order.

Is that so? I thought it was only byte order.

Then it depends what "word size" is? Only long has a different size on
x86-64 than on i386, while char, short, int and long long are same.

> Oops, that's it.
> After I made luac 32bit version, and use it by "LUAC=luac32
> ..../win32/configure.sh ...", lua scripts worked on win32 build.
> I don't know about Linux 64bit environment, on
> http://wiki.videolan.org/Win32Compile, there is no description about
> it.

I am not sure anybody ever tried to run the result.

Unfortunately, there is no lua cross-compiler, so there is no easy way to
fix this.

-- 
R?mi Denis-Courmont
http://www.remlab.net/


From remi at remlab.net  Wed Dec 14 09:10:28 2011
From: remi at remlab.net (=?UTF-8?Q?R=C3=A9mi_Denis-Courmont?=)
Date: Wed, 14 Dec 2011 09:10:28 +0100
Subject: [vlc-devel] a pitfall when cross-compiling win32 build on 64bit
	machine
In-Reply-To: <20111214020109.52f93878@tok>
References: <CAOa_K+t1dmgxwxfZS6=FH-o2H1Y4kwNCwdBuMSNeVjXM4P0z8w@mail.gmail.com>
	<20111214020109.52f93878@tok>
Message-ID: <da87c3e3ca0f1c3870a3278a27ed59db@chewa.net>

On Wed, 14 Dec 2011 02:01:09 -0500, Rafa?l Carr? <funman at videolan.org>
wrote:
> The .lua files (not compiled) should work fine, I think we used luac
> only to reduce binary size of VLC package.

The main motivation was speed up. Interpreting byte code is much faster
than interpreting source code.

There are a lot of lua scripts nowadays. For demuxer probing, LUA is now
certainly the main bottleneck by far now (I have not measured though).

-- 
R?mi Denis-Courmont
http://www.remlab.net/


From nkoriyama at gmail.com  Wed Dec 14 09:51:51 2011
From: nkoriyama at gmail.com (Naohiro KORIYAMA)
Date: Wed, 14 Dec 2011 17:51:51 +0900
Subject: [vlc-devel] a pitfall when cross-compiling win32 build on 64bit
	machine
In-Reply-To: <b11d8e8953c48161d2ec0f614a324f7f@chewa.net>
References: <CAOa_K+t1dmgxwxfZS6=FH-o2H1Y4kwNCwdBuMSNeVjXM4P0z8w@mail.gmail.com>
	<b11d8e8953c48161d2ec0f614a324f7f@chewa.net>
Message-ID: <CAOa_K+sKifa-fzXeWHCBXGdxnH_Lnnin4iLWqwhePV303qUe=w@mail.gmail.com>

2011/12/14 R?mi Denis-Courmont <remi at remlab.net>:
> ? Naohiro-san hello,
>
> On Wed, 14 Dec 2011 15:48:54 +0900, Naohiro KORIYAMA <nkoriyama at gmail.com>
> wrote:
>> I compile win32 build on OS X Lion.
>>
>> First, it seemed fine, but lua scripts didn't run at all.
>> I installed lua in /usr/local/bin, and it was built as 64bit
> application.
>> And, I found the following in the LUAC man page:
>> The binary files created by luac are portable only among architectures
>> with the same word size and byte order.
>
> Is that so? I thought it was only byte order.

At least on my case, byte order is the same, because my macbook air's
cpu is corei5(x86).

> Then it depends what "word size" is? Only long has a different size on
> x86-64 than on i386, while char, short, int and long long are same.

Difference between x86_64 and i386, long(4,8) and long double(12,16), maybe.
luac's header has size of size_t (long unsigned int 4,8).
If the size in header and the machine's sizeof(size_t) is different,
treated as "bad header".

luadump.c
/*
* make header
*/
void luaU_header (char* h)
{
 int x=1;
 memcpy(h,LUA_SIGNATURE,sizeof(LUA_SIGNATURE)-1);
 h+=sizeof(LUA_SIGNATURE)-1;
 *h++=(char)LUAC_VERSION;
 *h++=(char)LUAC_FORMAT;
 *h++=(char)*(char*)&x;				/* endianness */
 *h++=(char)sizeof(int);
 *h++=(char)sizeof(size_t);
 *h++=(char)sizeof(Instruction);
 *h++=(char)sizeof(lua_Number);
 *h++=(char)(((lua_Number)0.5)==0);		/* is lua_Number integral? */
}

static void LoadHeader(LoadState* S)
{
 char h[LUAC_HEADERSIZE];
 char s[LUAC_HEADERSIZE];
 luaU_header(h);
 LoadBlock(S,s,LUAC_HEADERSIZE);
 IF (memcmp(h,s,LUAC_HEADERSIZE)!=0, "bad header");
}


>
>> Oops, that's it.
>> After I made luac 32bit version, and use it by "LUAC=luac32
>> ..../win32/configure.sh ...", lua scripts worked on win32 build.
>> I don't know about Linux 64bit environment, on
>> http://wiki.videolan.org/Win32Compile, there is no description about
>> it.
>
> I am not sure anybody ever tried to run the result.
>
> Unfortunately, there is no lua cross-compiler, so there is no easy way to
> fix this.

Yes.
On my case, only difference of word size, so I can easily act, using
32bit luac.

Anyway, I wonder how to switch luac 32bit and 64bit on windows nightly
builds if the environment are the same.

-- 
KORIYAMA, Naohiro
nkoriyama at gmail.com


From hugo at videolan.org  Wed Dec 14 11:43:15 2011
From: hugo at videolan.org (=?UTF-8?Q?Hugo_Beauz=C3=A9e=2DLuyssen?=)
Date: Wed, 14 Dec 2011 11:43:15 +0100
Subject: [vlc-devel] FOSDEM 2012
Message-ID: <CAK-C75QBGcaC-GNZoWSHq_m0TO_DcDsd4xFeFvjj3G-ZNWe_Ug@mail.gmail.com>

Hi,

As some of you may know, the VideoLAN team will meet at FOSDEM! So if
you wish to join us, now is the time to speak!

FOSDEM 2012 will take place on February 4th, 5th. There's a beer event
on February 3rd, for those interested.

As usual, for active community members, VideoLAN will pay for one
night, and for the travel.  That means if you wish to attend the beer
event, you'll be paying the extra hotel night.

We will book rooms for two, so you'll have to share a room with a fellow geek.
If you wish to attend, you can either mail trez at videolan.org with some
basic info :
    - When do you arrive? Friday or Saturday?
    - Where will you be taking your train/plane from, and do you
travel back, or to another location.
    - Do you have a special geek/dev/pony you want to share the room with?
    - Will you book your ticket by yourself? Or do you want me to book
it for you? (This can ease up the refund, especially if you leave
outside the EU)

Or you can just fill this form out:
https://docs.google.com/spreadsheet/viewform?formkey=dF9VUDFCeVRIaHZGYS1OVlFYVXQyZ0E6MQ&ifq

If you book your ticket yourself, please do it quick! The sooner the cheaper.
Please answer by the end of December so we can book the hotel.

Regards,

-- 
Hugo Beauz?e-Luyssen


From edward.c.wang at compdigitec.com  Wed Dec 14 12:01:50 2011
From: edward.c.wang at compdigitec.com (Edward Wang)
Date: Wed, 14 Dec 2011 06:01:50 -0500
Subject: [vlc-devel] [PATCH] Add title time in playback menus for MP4
Message-ID: <CAHqHa+J9rOrV4Grcdvdwv=gncCBQdj3mHvSa2G_xgVijY61Jjg@mail.gmail.com>

Hi all:

This patch adds the time in the playback menu for MP4 files with chapters -
see bug #4030 (http://trac.videolan.org/vlc/ticket/4030). Attached is the
patch file.

Thanks,
Edward Wang
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111214/2091b7e0/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: mp4.patch
Type: text/x-c
Size: 552 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111214/2091b7e0/attachment.bin>

From jb at videolan.org  Wed Dec 14 16:11:54 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Wed, 14 Dec 2011 16:11:54 +0100
Subject: [vlc-devel] [PATCH] Add title time in playback menus for MP4
In-Reply-To: <CAHqHa+J9rOrV4Grcdvdwv=gncCBQdj3mHvSa2G_xgVijY61Jjg@mail.gmail.com>
References: <CAHqHa+J9rOrV4Grcdvdwv=gncCBQdj3mHvSa2G_xgVijY61Jjg@mail.gmail.com>
Message-ID: <20111214151154.GA22153@videolan.org>

On Wed, Dec 14, 2011 at 06:01:50AM -0500, Edward Wang wrote :
> This patch adds the time in the playback menu for MP4 files with chapters -
> see bug #4030 (http://trac.videolan.org/vlc/ticket/4030). Attached is the
> patch file.

Applied, thx.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From funman at videolan.org  Wed Dec 14 16:12:00 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Wed, 14 Dec 2011 10:12:00 -0500
Subject: [vlc-devel] a pitfall when cross-compiling win32 build on 64bit
 machine
In-Reply-To: <CAOa_K+v3VvKWiMy=dzaG0A8BHcxg=dVs_zFxY+XJG8rOEu+9gA@mail.gmail.com>
References: <CAOa_K+t1dmgxwxfZS6=FH-o2H1Y4kwNCwdBuMSNeVjXM4P0z8w@mail.gmail.com>
	<20111214020109.52f93878@tok>
	<CAOa_K+v3VvKWiMy=dzaG0A8BHcxg=dVs_zFxY+XJG8rOEu+9gA@mail.gmail.com>
Message-ID: <20111214101200.7aa83afc@tok>

Le Wed, 14 Dec 2011 16:49:13 +0900,
Naohiro KORIYAMA <nkoriyama at gmail.com> a ?crit :

> 2011/12/14 Rafa?l Carr? <funman at videolan.org>:
> > Le Wed, 14 Dec 2011 15:48:54 +0900,
> > Naohiro KORIYAMA <nkoriyama at gmail.com> a ?crit :
> >
> >> I compile win32 build on OS X Lion.
> >>
> >> First, it seemed fine, but lua scripts didn't run at all.
> >> I installed lua in /usr/local/bin, and it was built as 64bit application.
> >> And, I found the following in the LUAC man page:
> >> The binary files created by luac are portable only among architectures
> >> with the same word size and byte order.
> >>
> >> Oops, that's it.
> >> After I made luac 32bit version, and use it by "LUAC=luac32
> >> ..../win32/configure.sh ...", lua scripts worked on win32 build.
> >> I don't know about Linux 64bit environment, on
> >> http://wiki.videolan.org/Win32Compile, there is no description about
> >> it.
> >
> > It's strange, I think the windows builds are made on a linux 64 bits
> > machine but they run fine on win32.
> 
> Header infomation in luac is different  32bit and 64bit, and liblua checks it.
> liblua refuses when loading wrong luac like:

Yeah, now I remember that win32 buildbot is built in a 32bits chroot

> If it's only my particular problem, please ignore it.

I think it is going to bite us sooner or later

-- 
Rafa?l Carr?


From jb at videolan.org  Wed Dec 14 16:12:41 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Wed, 14 Dec 2011 16:12:41 +0100
Subject: [vlc-devel] [PATCH] faad : Reinitializing decoder when we got
 Unexpected channel configuration change.
In-Reply-To: <CAOa_K+uWP0EoAr0Fschr-htnBjGF+BoE9a2PzgztptJeEG-Nng@mail.gmail.com>
References: <CAOa_K+uWP0EoAr0Fschr-htnBjGF+BoE9a2PzgztptJeEG-Nng@mail.gmail.com>
Message-ID: <20111214151241.GB22153@videolan.org>

On Wed, Dec 14, 2011 at 12:06:59AM +0900, Naohiro KORIYAMA wrote :
> Duraing playing some video including aac audio, once we got an error
> "Unexpected channel configuration change",  same error occurs
> afterwards, and we got no sound.
> 
> This patch fixes the aac audio issue.
> There are 2 tickets about this issue. #5448, #3714

Applied, thx. Both tickets are now closed.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From funman at videolan.org  Wed Dec 14 16:12:49 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Wed, 14 Dec 2011 10:12:49 -0500
Subject: [vlc-devel] a pitfall when cross-compiling win32 build on 64bit
 machine
In-Reply-To: <b11d8e8953c48161d2ec0f614a324f7f@chewa.net>
References: <CAOa_K+t1dmgxwxfZS6=FH-o2H1Y4kwNCwdBuMSNeVjXM4P0z8w@mail.gmail.com>
	<b11d8e8953c48161d2ec0f614a324f7f@chewa.net>
Message-ID: <20111214101249.7d3d732b@tok>

Le Wed, 14 Dec 2011 09:05:39 +0100,
R?mi Denis-Courmont <remi at remlab.net> a ?crit :

>    Naohiro-san hello,
> 
> On Wed, 14 Dec 2011 15:48:54 +0900, Naohiro KORIYAMA <nkoriyama at gmail.com>
> wrote:
> > I compile win32 build on OS X Lion.
> > 
> > First, it seemed fine, but lua scripts didn't run at all.
> > I installed lua in /usr/local/bin, and it was built as 64bit
> application.
> > And, I found the following in the LUAC man page:
> > The binary files created by luac are portable only among architectures
> > with the same word size and byte order.
> 
> Is that so? I thought it was only byte order.

lua wiki says it doesn't depend on byte order (it's not in sync with
luac manpage at least)

-- 
Rafa?l Carr?


From jb at videolan.org  Wed Dec 14 16:13:14 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Wed, 14 Dec 2011 16:13:14 +0100
Subject: [vlc-devel] [PATCH] Qt4: use QWizard for sout (stream out)
 dialog (#5651)
In-Reply-To: <CAHqHa+JPacAOkvUFnD_=mUxseqJp9-kLAYmZBhj9u3LDHamFYw@mail.gmail.com>
References: <CAHqHa+JPacAOkvUFnD_=mUxseqJp9-kLAYmZBhj9u3LDHamFYw@mail.gmail.com>
Message-ID: <20111214151314.GC22153@videolan.org>

On Sun, Dec 11, 2011 at 04:34:21PM -0500, Edward Wang wrote :
> This patch will add the new feature in bug #5651 (
> http://trac.videolan.org/vlc/ticket/5651 - Qt4: use QWizard for sout
> (stream out) dialog). Attached is the patch file.

This was applied.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Wed Dec 14 16:13:28 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Wed, 14 Dec 2011 16:13:28 +0100
Subject: [vlc-devel] [PATCH] Fix EQ HTTP UI partially
In-Reply-To: <CAOgZoSd6MBJdT_DFKwWSD+AmNEBCA6EujhWyX6MH96hGur4AHg@mail.gmail.com>
References: <CAOgZoSd6MBJdT_DFKwWSD+AmNEBCA6EujhWyX6MH96hGur4AHg@mail.gmail.com>
Message-ID: <20111214151328.GD22153@videolan.org>

On Mon, Dec 12, 2011 at 12:54:25AM +0530, akash mehrotra wrote :
> Partial fix for the Lua Web Interface Equalizer

What does it fix?

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From funman at videolan.org  Wed Dec 14 16:14:00 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Wed, 14 Dec 2011 10:14:00 -0500
Subject: [vlc-devel] [vlc-commits] fix typo in french translation of nvlc
In-Reply-To: <223ba297b5536aae5067b5d2c3438557@chewa.net>
References: <20111214045035.CCC071447CA@albiero.videolan.org>
	<223ba297b5536aae5067b5d2c3438557@chewa.net>
Message-ID: <20111214101400.31cbc6a9@tok>

Le Wed, 14 Dec 2011 08:58:22 +0100,
R?mi Denis-Courmont <remi at remlab.net> a ?crit :

> On Wed, 14 Dec 2011 05:50:35 +0100 (CET), git at videolan.org (Rafa?l Carr?)
> wrote:
> > vlc | branch: master | Rafa?l Carr? <funman at videolan.org> | Tue Dec 13
> > 23:50:18 2011 -0500| [c52cfdd034558076705445510d3bbb95719f9704] |
> > committer: Rafa?l Carr?
> > 
> > fix typo in french translation of nvlc
> 
> This will be overwritten when 1.2 translations are merged into 1.3...

should I backport it then?

xtophe?

-- 
Rafa?l Carr?


From nkoriyama at gmail.com  Wed Dec 14 16:20:02 2011
From: nkoriyama at gmail.com (Naohiro KORIYAMA)
Date: Thu, 15 Dec 2011 00:20:02 +0900
Subject: [vlc-devel] a pitfall when cross-compiling win32 build on 64bit
	machine
In-Reply-To: <20111214101249.7d3d732b@tok>
References: <CAOa_K+t1dmgxwxfZS6=FH-o2H1Y4kwNCwdBuMSNeVjXM4P0z8w@mail.gmail.com>
	<b11d8e8953c48161d2ec0f614a324f7f@chewa.net>
	<20111214101249.7d3d732b@tok>
Message-ID: <CAOa_K+vptB-CS1tNDirPjSTAU4ZHHp=4N7qA_bjNX5JEO6CymA@mail.gmail.com>

2011/12/15 Rafa?l Carr? <funman at videolan.org>:
> Le Wed, 14 Dec 2011 09:05:39 +0100,
> R?mi Denis-Courmont <remi at remlab.net> a ?crit :
>
>> ? ?Naohiro-san hello,
>>
>> On Wed, 14 Dec 2011 15:48:54 +0900, Naohiro KORIYAMA <nkoriyama at gmail.com>
>> wrote:
>> > I compile win32 build on OS X Lion.
>> >
>> > First, it seemed fine, but lua scripts didn't run at all.
>> > I installed lua in /usr/local/bin, and it was built as 64bit
>> application.
>> > And, I found the following in the LUAC man page:
>> > The binary files created by luac are portable only among architectures
>> > with the same word size and byte order.
>>
>> Is that so? I thought it was only byte order.
>
> lua wiki says it doesn't depend on byte order (it's not in sync with
> luac manpage at least)

Endianness is checked at header loading.

int x=1;
 *h++=(char)*(char*)&x;                         /* endianness */


-- 
KORIYAMA, Naohiro
nkoriyama at gmail.com


From funman at videolan.org  Wed Dec 14 16:25:10 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Wed, 14 Dec 2011 10:25:10 -0500
Subject: [vlc-devel] a pitfall when cross-compiling win32 build on 64bit
 machine
In-Reply-To: <CAOa_K+vptB-CS1tNDirPjSTAU4ZHHp=4N7qA_bjNX5JEO6CymA@mail.gmail.com>
References: <CAOa_K+t1dmgxwxfZS6=FH-o2H1Y4kwNCwdBuMSNeVjXM4P0z8w@mail.gmail.com>
	<b11d8e8953c48161d2ec0f614a324f7f@chewa.net>
	<20111214101249.7d3d732b@tok>
	<CAOa_K+vptB-CS1tNDirPjSTAU4ZHHp=4N7qA_bjNX5JEO6CymA@mail.gmail.com>
Message-ID: <20111214102510.5246c1d3@tok>

Le Thu, 15 Dec 2011 00:20:02 +0900,
Naohiro KORIYAMA <nkoriyama at gmail.com> a ?crit :

> 2011/12/15 Rafa?l Carr? <funman at videolan.org>:
> > Le Wed, 14 Dec 2011 09:05:39 +0100,
> > R?mi Denis-Courmont <remi at remlab.net> a ?crit :
> >
> >> ? ?Naohiro-san hello,
> >>
> >> On Wed, 14 Dec 2011 15:48:54 +0900, Naohiro KORIYAMA <nkoriyama at gmail.com>
> >> wrote:
> >> > I compile win32 build on OS X Lion.
> >> >
> >> > First, it seemed fine, but lua scripts didn't run at all.
> >> > I installed lua in /usr/local/bin, and it was built as 64bit
> >> application.
> >> > And, I found the following in the LUAC man page:
> >> > The binary files created by luac are portable only among architectures
> >> > with the same word size and byte order.
> >>
> >> Is that so? I thought it was only byte order.
> >
> > lua wiki says it doesn't depend on byte order (it's not in sync with
> > luac manpage at least)
> 
> Endianness is checked at header loading.
> 
> int x=1;
>  *h++=(char)*(char*)&x;                         /* endianness */

We could see what luac checks exactly and add those checks to
configure.ac when cross compiling.

gcc has macros for byte order: __BYTE_ORDER__
and __SIZEOF_INT__ etc so we can check them when cross compiling

-- 
Rafa?l Carr?


From nkoriyama at gmail.com  Wed Dec 14 16:57:02 2011
From: nkoriyama at gmail.com (Naohiro KORIYAMA)
Date: Thu, 15 Dec 2011 00:57:02 +0900
Subject: [vlc-devel] a pitfall when cross-compiling win32 build on 64bit
	machine
In-Reply-To: <20111214102510.5246c1d3@tok>
References: <CAOa_K+t1dmgxwxfZS6=FH-o2H1Y4kwNCwdBuMSNeVjXM4P0z8w@mail.gmail.com>
	<b11d8e8953c48161d2ec0f614a324f7f@chewa.net>
	<20111214101249.7d3d732b@tok>
	<CAOa_K+vptB-CS1tNDirPjSTAU4ZHHp=4N7qA_bjNX5JEO6CymA@mail.gmail.com>
	<20111214102510.5246c1d3@tok>
Message-ID: <CAOa_K+vDyV+LNhdhahy1Oyxv9VfLdT8k+ZN3wXBWgYb3iL9+Tg@mail.gmail.com>

2011/12/15 Rafa?l Carr? <funman at videolan.org>:
> Le Thu, 15 Dec 2011 00:20:02 +0900,
> Naohiro KORIYAMA <nkoriyama at gmail.com> a ?crit :
>
>> 2011/12/15 Rafa?l Carr? <funman at videolan.org>:
>> > Le Wed, 14 Dec 2011 09:05:39 +0100,
>> > R?mi Denis-Courmont <remi at remlab.net> a ?crit :
>> >
>> >> ? ?Naohiro-san hello,
>> >>
>> >> On Wed, 14 Dec 2011 15:48:54 +0900, Naohiro KORIYAMA <nkoriyama at gmail.com>
>> >> wrote:
>> >> > I compile win32 build on OS X Lion.
>> >> >
>> >> > First, it seemed fine, but lua scripts didn't run at all.
>> >> > I installed lua in /usr/local/bin, and it was built as 64bit
>> >> application.
>> >> > And, I found the following in the LUAC man page:
>> >> > The binary files created by luac are portable only among architectures
>> >> > with the same word size and byte order.
>> >>
>> >> Is that so? I thought it was only byte order.
>> >
>> > lua wiki says it doesn't depend on byte order (it's not in sync with
>> > luac manpage at least)
>>
>> Endianness is checked at header loading.
>>
>> int x=1;
>> ?*h++=(char)*(char*)&x; ? ? ? ? ? ? ? ? ? ? ? ? /* endianness */
>
> We could see what luac checks exactly and add those checks to
> configure.ac when cross compiling.
>
> gcc has macros for byte order: __BYTE_ORDER__
> and __SIZEOF_INT__ etc so we can check them when cross compiling

compare byteorder and size of long between native gcc and target gcc.
if differences exist, put some warning to config.log.
"Be sure your luac ..."

Anyway, ATM  I think writing some description to the wiki page
(http://wiki.videolan.org/Win32Compile) is enough .

-- 
KORIYAMA, Naohiro
nkoriyama at gmail.com


From remi at remlab.net  Wed Dec 14 17:51:00 2011
From: remi at remlab.net (=?utf-8?q?R=C3=A9mi?= Denis-Courmont)
Date: Wed, 14 Dec 2011 18:51:00 +0200
Subject: [vlc-devel] a pitfall when cross-compiling win32 build on 64bit
	machine
In-Reply-To: <b11d8e8953c48161d2ec0f614a324f7f@chewa.net>
References: <CAOa_K+t1dmgxwxfZS6=FH-o2H1Y4kwNCwdBuMSNeVjXM4P0z8w@mail.gmail.com>
	<b11d8e8953c48161d2ec0f614a324f7f@chewa.net>
Message-ID: <201112141851.01271.remi@remlab.net>

Le mercredi 14 d?cembre 2011 10:05:39 R?mi Denis-Courmont, vous avez ?crit :
> Then it depends what "word size" is? Only long has a different size on
> x86-64 than on i386, while char, short, int and long long are same.

So the problem is size_t. It is used to store strings size for some reason. We 
could force 32-bits values instead. This should be an easy patch in contrib. 
Of course that will cause other compatibility problems with the upstream Lua.

Another option is to support all both 32-bits and 64-bits byte codes. But such 
patch would be a lot more complicated.

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From mehrotra.akash at gmail.com  Wed Dec 14 17:57:30 2011
From: mehrotra.akash at gmail.com (akash mehrotra)
Date: Wed, 14 Dec 2011 22:27:30 +0530
Subject: [vlc-devel] [PATCH] Fix EQ HTTP UI partially
In-Reply-To: <20111214151328.GD22153@videolan.org>
References: <CAOgZoSd6MBJdT_DFKwWSD+AmNEBCA6EujhWyX6MH96hGur4AHg@mail.gmail.com>
	<20111214151328.GD22153@videolan.org>
Message-ID: <CAOgZoSe2Bkrgmkr5LT7zRfT5QvosBBsSrnG5-p=baHYUfucGoA@mail.gmail.com>

On Wed, Dec 14, 2011 at 8:43 PM, Jean-Baptiste Kempf <jb at videolan.org>wrote:

> On Mon, Dec 12, 2011 at 12:54:25AM +0530, akash mehrotra wrote :
> > Partial fix for the Lua Web Interface Equalizer
>
> What does it fix?
>
> Equalizer is currently non functional through the Web UI

This makes it functional but the order of the bars to adjust the gain per
band is incorrect (labels are correct)

> Best regards,
>
> --
> Jean-Baptiste Kempf
> http://www.jbkempf.com/ - +33 672 704 734
> Sent from my Electronic Device
> _______________________________________________
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
> http://mailman.videolan.org/listinfo/vlc-devel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111214/943ab454/attachment.html>

From remi at remlab.net  Wed Dec 14 18:02:00 2011
From: remi at remlab.net (=?utf-8?q?R=C3=A9mi?= Denis-Courmont)
Date: Wed, 14 Dec 2011 19:02:00 +0200
Subject: [vlc-devel] a pitfall when cross-compiling win32 build on 64bit
	machine
In-Reply-To: <20111214101249.7d3d732b@tok>
References: <CAOa_K+t1dmgxwxfZS6=FH-o2H1Y4kwNCwdBuMSNeVjXM4P0z8w@mail.gmail.com>
	<b11d8e8953c48161d2ec0f614a324f7f@chewa.net>
	<20111214101249.7d3d732b@tok>
Message-ID: <201112141902.00629.remi@remlab.net>

Le mercredi 14 d?cembre 2011 17:12:49 Rafa?l Carr?, vous avez ?crit :
> Le Wed, 14 Dec 2011 09:05:39 +0100,
> 
> R?mi Denis-Courmont <remi at remlab.net> a ?crit :
> >    Naohiro-san hello,
> > 
> > On Wed, 14 Dec 2011 15:48:54 +0900, Naohiro KORIYAMA
> > <nkoriyama at gmail.com>
> > 
> > wrote:
> > > I compile win32 build on OS X Lion.
> > > 
> > > First, it seemed fine, but lua scripts didn't run at all.
> > > I installed lua in /usr/local/bin, and it was built as 64bit
> > 
> > application.
> > 
> > > And, I found the following in the LUAC man page:
> > > The binary files created by luac are portable only among architectures
> > > with the same word size and byte order.
> > 
> > Is that so? I thought it was only byte order.
> 
> lua wiki says it doesn't depend on byte order (it's not in sync with
> luac manpage at least)

To be more precise, it depends on:
- the endianess,
- the size of 'int',
- the size of 'size_t',
- the size of 'double'.

For the dirty little details, see lundump.c and ldump.c. Note that VLC assumes 
int has 32-bits, so 'Instruction' is always 'int' on VLC-supported platforms.

Storing double precision values portably is not completely trivial.

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From remi at remlab.net  Wed Dec 14 18:05:47 2011
From: remi at remlab.net (=?iso-8859-15?q?R=E9mi?= Denis-Courmont)
Date: Wed, 14 Dec 2011 19:05:47 +0200
Subject: [vlc-devel] [Patch] Adding an helper to convert iso8601
	durations to seconds
In-Reply-To: <CAK-C75QD08hyZ4XNBR7nYkSFyLqR7Tm887Uxw7fWqpXRGJd=Dg@mail.gmail.com>
References: <CAK-C75R5BPUud6z2QxrFNNnh1wMFktu48O59oXi2bAUqZ0mYMw@mail.gmail.com>
	<CAK-C75RtmcQd0m32+3d9ue7WZKH8v9n84izFWRp=ziwKm4yCEw@mail.gmail.com>
	<CAK-C75QD08hyZ4XNBR7nYkSFyLqR7Tm887Uxw7fWqpXRGJd=Dg@mail.gmail.com>
Message-ID: <201112141905.47669.remi@remlab.net>

Le mercredi 14 d?cembre 2011 00:48:41 Hugo Beauz?e-Luyssen, vous avez ?crit :
> On Tue, Dec 13, 2011 at 11:46 PM, Hugo Beauz?e-Luyssen
> 
> <beauze.h at gmail.com> wrote:
> > 2011/12/9 R?mi Denis-Courmont <remi at remlab.net>:
> >> Function name is too long. And time_t is thecorrect type for seconds
> >> IMHO.
> 
> Oops, this one's the good one.

I think this patch does not work due to typo. Please drop _to_s*. It's 
obviously seconds since the return type is time_t.

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From beauze.h at gmail.com  Wed Dec 14 18:33:41 2011
From: beauze.h at gmail.com (=?UTF-8?Q?Hugo_Beauz=C3=A9e=2DLuyssen?=)
Date: Wed, 14 Dec 2011 18:33:41 +0100
Subject: [vlc-devel] [Patch] Adding an helper to convert iso8601
 durations to seconds
In-Reply-To: <201112141905.47669.remi@remlab.net>
References: <CAK-C75R5BPUud6z2QxrFNNnh1wMFktu48O59oXi2bAUqZ0mYMw@mail.gmail.com>
	<CAK-C75RtmcQd0m32+3d9ue7WZKH8v9n84izFWRp=ziwKm4yCEw@mail.gmail.com>
	<CAK-C75QD08hyZ4XNBR7nYkSFyLqR7Tm887Uxw7fWqpXRGJd=Dg@mail.gmail.com>
	<201112141905.47669.remi@remlab.net>
Message-ID: <CAK-C75RayZc-hHcd9-BE-cF5gXFF_JjNB6hDGCz5wxS4OfSfAg@mail.gmail.com>

2011/12/14 R?mi Denis-Courmont <remi at remlab.net>:
> Le mercredi 14 d?cembre 2011 00:48:41 Hugo Beauz?e-Luyssen, vous avez ?crit :
>> On Tue, Dec 13, 2011 at 11:46 PM, Hugo Beauz?e-Luyssen
>>
>> <beauze.h at gmail.com> wrote:
>> > 2011/12/9 R?mi Denis-Courmont <remi at remlab.net>:
>> >> Function name is too long. And time_t is thecorrect type for seconds
>> >> IMHO.
>>
>> Oops, this one's the good one.
>
> I think this patch does not work due to typo. Please drop _to_s*. It's
> obviously seconds since the return type is time_t.
>

Hmmm yep you're right... my bad.
Here are some fixed versions of both patches.

Regards,

-- 
Hugo Beauz?e-Luyssen
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0005-tests-Running-new-strings-tests-in-the-test-suite.patch
Type: text/x-patch
Size: 3359 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111214/fbfad510/attachment.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0004-strings-Adding-an-helper-to-convert-iso8601-duration.patch
Type: text/x-patch
Size: 3037 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111214/fbfad510/attachment-0001.bin>

From jb at videolan.org  Wed Dec 14 19:06:41 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Wed, 14 Dec 2011 19:06:41 +0100
Subject: [vlc-devel] [Patch] Dash: Don't try to do something without a
 valid	MPD manager
In-Reply-To: <004601ccadfb$793402b0$6b9c0810$@mueller@itec.uni-klu.ac.at>
References: <CAK-C75TOrnUGWki6HDVk0iJ0_xE1ZO4Swd8FGAoCraupTXQW8Q@mail.gmail.com>
	<004601ccadfb$793402b0$6b9c0810$@mueller@itec.uni-klu.ac.at>
Message-ID: <20111214180641.GA23803@videolan.org>

Why do we need a NullManager, in the end, since we have the empty
interface already?
What is the use of instantiating it, if it does not do anything?
If the profile is not implemented, we just quit the module, no?

I must be missing something, but I don't see the point.

On Mon, Nov 28, 2011 at 10:28:04AM -0800, Christopher M?ller wrote :
> Why do you want to get rid of it? You have to consider that the MPD part will grow immensely if we implement the whole DASH specification. Furthermore there are a lot of profiles, therefore I don?t like that.
> 
> > -----Urspr?ngliche Nachricht-----
> > Von: vlc-devel-bounces at videolan.org [mailto:vlc-devel-
> > bounces at videolan.org] Im Auftrag von Hugo Beauz?e-Luyssen
> > Gesendet: Montag, 28. November 2011 07:04
> > An: Mailing list for VLC media player developers
> > Betreff: [vlc-devel] [Patch] Dash: Don't try to do something without a
> > valid MPD manager
> > 
> > Hi,
> > 
> > This patch removes a "no-op" MPD manager, and just stops the computing
> > if the profile is invalid.
> > Unless something was planned with the NullManager, I don't see any use
> > for it, as it returns NULL everywhere, which just postpones the failure
> > in the best case.
> > 
> > Regards,
> > 
> > --
> > Hugo Beauz?e-Luyssen
> 
> _______________________________________________
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
> http://mailman.videolan.org/listinfo/vlc-devel

-- 
Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From enrique.osuna at gmail.com  Wed Dec 14 19:21:04 2011
From: enrique.osuna at gmail.com (Faustino E. Osuna)
Date: Wed, 14 Dec 2011 19:21:04 +0100
Subject: [vlc-devel] [PATCH 0/2] extras/tools General Updates
Message-ID: <1323886866-2007-1-git-send-email-enrique.osuna@gmail.com>

Hello All,

This is my first contribution in a very (very very) long time. I'm out of touch
with today's patch submission/review protocol so let me know if I'm out of
step.

The following two patches are just some general updates to the tools subfolder:
1) Fixes a typo in one of the makefiles.

2) Updates the .gitignore folder to ignore the downloaded tools. In the
future, if you need to add a subfolder (e.g. extra/tools/pleasetrackme) you
would have to use `git add -f` to override the .gitignore file.

I look forward to your feedback.

Enrique

Faustino E. Osuna (2):
  extras/tools: fix when downloading in tools.mak
  extras/tools: update ignore files

 extras/tools/.gitignore |    2 ++
 extras/tools/tools.mak  |   16 ++++++++--------
 2 files changed, 10 insertions(+), 8 deletions(-)

-- 
1.7.7.4



From enrique.osuna at gmail.com  Wed Dec 14 19:21:05 2011
From: enrique.osuna at gmail.com (Faustino E. Osuna)
Date: Wed, 14 Dec 2011 19:21:05 +0100
Subject: [vlc-devel] [PATCH 1/2] extras/tools: fix when downloading in
	tools.mak
In-Reply-To: <1323886866-2007-1-git-send-email-enrique.osuna@gmail.com>
References: <1323886866-2007-1-git-send-email-enrique.osuna@gmail.com>
Message-ID: <1323886866-2007-2-git-send-email-enrique.osuna@gmail.com>

Fix a typo when using the $(download) function. This suppresses
malformed url errors from curl.
---
 extras/tools/tools.mak |   16 ++++++++--------
 1 files changed, 8 insertions(+), 8 deletions(-)

diff --git a/extras/tools/tools.mak b/extras/tools/tools.mak
index 81f74d4..0c85458 100644
--- a/extras/tools/tools.mak
+++ b/extras/tools/tools.mak
@@ -39,7 +39,7 @@ MOVE = mv $(UNPACK_DIR) $@ && touch $@
 # yasm
 
 yasm-$(YASM_VERSION).tar.gz:
-	$(download) $(YASM_URL)
+	$(call download,$(YASM_URL))
 
 yasm: yasm-$(YASM_VERSION).tar.gz
 	$(UNPACK)
@@ -56,7 +56,7 @@ DISTCLEAN_PKG += yasm-$(YASM_VERSION).tar.gz
 # cmake
 
 cmake-$(CMAKE_VERSION).tar.gz:
-	$(download) $(CMAKE_URL)
+	$(call download,$(CMAKE_URL))
 
 cmake: cmake-$(CMAKE_VERSION).tar.gz
 	$(UNPACK)
@@ -73,7 +73,7 @@ DISTCLEAN_PKG += cmake-$(CMAKE_VERSION).tar.gz
 # libtool
 
 libtool-$(LIBTOOL_VERSION).tar.gz:
-	$(download) $(LIBTOOL_URL)
+	$(call download,$(LIBTOOL_URL))
 
 libtool: libtool-$(LIBTOOL_VERSION).tar.gz
 	$(UNPACK)
@@ -92,7 +92,7 @@ CLEAN_FILE += .libtool
 # GNU tar (with xz support)
 
 tar-$(TAR_VERSION).tar.bz2:
-	$(download) $(TAR_URL)
+	$(call download,$(TAR_URL))
 
 tar: tar-$(TAR_VERSION).tar.bz2
 	$(UNPACK)
@@ -109,7 +109,7 @@ CLEAN_FILE += .tar
 # xz
 
 xz-$(XZ_VERSION).tar.bz2:
-	$(download) $(XZ_URL)
+	$(call download,$(XZ_URL))
 
 xz: xz-$(XZ_VERSION).tar.bz2
 	$(UNPACK)
@@ -126,7 +126,7 @@ CLEAN_FILE += .xz
 # autoconf
 
 autoconf-$(AUTOCONF_VERSION).tar.bz2:
-	$(download) $(AUTOCONF_URL)
+	$(call download,$(AUTOCONF_URL))
 
 autoconf: autoconf-$(AUTOCONF_VERSION).tar.bz2
 	$(UNPACK)
@@ -143,7 +143,7 @@ DISTCLEAN_PKG += autoconf-$(AUTOCONF_VERSION).tar.bz2
 # automake
 
 automake-$(AUTOMAKE_VERSION).tar.gz:
-	$(download) $(AUTOMAKE_URL)
+	$(call download,$(AUTOMAKE_URL))
 
 automake: automake-$(AUTOMAKE_VERSION).tar.gz
 	$(UNPACK)
@@ -160,7 +160,7 @@ DISTCLEAN_PKG += automake-$(AUTOMAKE_VERSION).tar.gz
 # pkg-config
 
 pkg-config-$(PKGCFG_VERSION).tar.gz:
-	$(download) $(PKGCFG_URL)
+	$(call download,$(PKGCFG_URL))
 
 pkgconfig: pkg-config-$(PKGCFG_VERSION).tar.gz
 	$(UNPACK)
-- 
1.7.7.4



From enrique.osuna at gmail.com  Wed Dec 14 19:21:06 2011
From: enrique.osuna at gmail.com (Faustino E. Osuna)
Date: Wed, 14 Dec 2011 19:21:06 +0100
Subject: [vlc-devel] [PATCH 2/2] extras/tools: update ignore files
In-Reply-To: <1323886866-2007-2-git-send-email-enrique.osuna@gmail.com>
References: <1323886866-2007-1-git-send-email-enrique.osuna@gmail.com>
	<1323886866-2007-2-git-send-email-enrique.osuna@gmail.com>
Message-ID: <1323886866-2007-3-git-send-email-enrique.osuna@gmail.com>

Ignore downloaded archives / subfolders and files generated by make.
---
 extras/tools/.gitignore |    2 ++
 1 files changed, 2 insertions(+), 0 deletions(-)

diff --git a/extras/tools/.gitignore b/extras/tools/.gitignore
index f23496b..6c25fda 100644
--- a/extras/tools/.gitignore
+++ b/extras/tools/.gitignore
@@ -1,4 +1,6 @@
 Makefile
 build/
+.*
+*/
 *.tar.gz
 *.tar.bz2
-- 
1.7.7.4



From funman at videolan.org  Wed Dec 14 19:45:27 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Wed, 14 Dec 2011 13:45:27 -0500
Subject: [vlc-devel] [PATCH 2/2] extras/tools: update ignore files
In-Reply-To: <1323886866-2007-3-git-send-email-enrique.osuna@gmail.com>
References: <1323886866-2007-1-git-send-email-enrique.osuna@gmail.com>
	<1323886866-2007-2-git-send-email-enrique.osuna@gmail.com>
	<1323886866-2007-3-git-send-email-enrique.osuna@gmail.com>
Message-ID: <20111214134527.7db6f013@tok>

Le Wed, 14 Dec 2011 19:21:06 +0100,
"Faustino E. Osuna" <enrique.osuna at gmail.com> a ?crit :

> Ignore downloaded archives / subfolders and files generated by make.
> ---
>  extras/tools/.gitignore |    2 ++
>  1 files changed, 2 insertions(+), 0 deletions(-)
> 
> diff --git a/extras/tools/.gitignore b/extras/tools/.gitignore
> index f23496b..6c25fda 100644
> --- a/extras/tools/.gitignore
> +++ b/extras/tools/.gitignore
> @@ -1,4 +1,6 @@
>  Makefile
>  build/
> +.*

does that also ignore .gitignore ?

the patch looks cool but i am not sure if it's right

> +*/
>  *.tar.gz
>  *.tar.bz2

-- 
Rafa?l Carr?


From enrique.osuna at gmail.com  Wed Dec 14 20:11:55 2011
From: enrique.osuna at gmail.com (Faustino Osuna)
Date: Wed, 14 Dec 2011 20:11:55 +0100
Subject: [vlc-devel] [PATCH 2/2] extras/tools: update ignore files
In-Reply-To: <20111214134527.7db6f013@tok>
References: <1323886866-2007-1-git-send-email-enrique.osuna@gmail.com>
	<1323886866-2007-2-git-send-email-enrique.osuna@gmail.com>
	<1323886866-2007-3-git-send-email-enrique.osuna@gmail.com>
	<20111214134527.7db6f013@tok>
Message-ID: <CAKhCBszkYWM0aBJtWtRwDxR7ZL-VZ3pUi5_n5xyfY=Mp5N_yJQ@mail.gmail.com>

It should only ignore subfolders.

On my local fork, I modified '.gitignore'.  If I run `git status`, it shows
up as modified.

I also created a bogus file "trackme" and it shows up as an untracked file.
 If i create a subfolder "track_me_dir" it doesn't show up as untracked.

Enrique

On Wed, Dec 14, 2011 at 7:45 PM, Rafa?l Carr? <funman at videolan.org> wrote:

> Le Wed, 14 Dec 2011 19:21:06 +0100,
> "Faustino E. Osuna" <enrique.osuna at gmail.com> a ?crit :
>
> > Ignore downloaded archives / subfolders and files generated by make.
> > ---
> >  extras/tools/.gitignore |    2 ++
> >  1 files changed, 2 insertions(+), 0 deletions(-)
> >
> > diff --git a/extras/tools/.gitignore b/extras/tools/.gitignore
> > index f23496b..6c25fda 100644
> > --- a/extras/tools/.gitignore
> > +++ b/extras/tools/.gitignore
> > @@ -1,4 +1,6 @@
> >  Makefile
> >  build/
> > +.*
>
> does that also ignore .gitignore ?
>
> the patch looks cool but i am not sure if it's right
>
> > +*/
> >  *.tar.gz
> >  *.tar.bz2
>
> --
> Rafa?l Carr?
> _______________________________________________
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
> http://mailman.videolan.org/listinfo/vlc-devel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111214/c42edb1c/attachment.html>

From enrique.osuna at gmail.com  Wed Dec 14 20:24:20 2011
From: enrique.osuna at gmail.com (Faustino E. Osuna)
Date: Wed, 14 Dec 2011 20:24:20 +0100
Subject: [vlc-devel] [PATCH 0/4] contrib General Updates
Message-ID: <1323890664-16077-1-git-send-email-enrique.osuna@gmail.com>

Here is another set of (primarily) cosmetic patches for the contribs sub
folder.
1) Update .gitignore to ignore the triplet subfolders 
(e.g. x86_64-apple-darwin10) generated by the new contribs system.

2) Update .gitignore to ignore cloned git repoisotiries.

3) Use spaces instead of tab characters for identation and character alignment.

4) Match `bootstrap` check_macosx_sdk()'s coding pattern to that of usage(). 

Enrique

Faustino E. Osuna (4):
  contrib: ignore triplets
  contrib: update ignore files for git repositories
  Cosmetic: white spacing for contrib scripts
  Cosmetic: use existing style for contrib/bootstrap

 contrib/.gitignore          |    1 +
 contrib/bootstrap           |  202 +++++++++++++++++++++---------------------
 contrib/src/get-arch.sh     |   40 ++++----
 contrib/tarballs/.gitignore |    1 +
 4 files changed, 123 insertions(+), 121 deletions(-)

-- 
1.7.7.4



From enrique.osuna at gmail.com  Wed Dec 14 20:24:21 2011
From: enrique.osuna at gmail.com (Faustino E. Osuna)
Date: Wed, 14 Dec 2011 20:24:21 +0100
Subject: [vlc-devel] [PATCH 1/4] contrib: ignore triplets
In-Reply-To: <1323890664-16077-1-git-send-email-enrique.osuna@gmail.com>
References: <1323890664-16077-1-git-send-email-enrique.osuna@gmail.com>
Message-ID: <1323890664-16077-2-git-send-email-enrique.osuna@gmail.com>

Ignore the triplet subfolders created in the contrib path.
---
 contrib/.gitignore |    1 +
 1 files changed, 1 insertions(+), 0 deletions(-)

diff --git a/contrib/.gitignore b/contrib/.gitignore
index 2b8d00e..64cbe1d 100644
--- a/contrib/.gitignore
+++ b/contrib/.gitignore
@@ -1 +1,2 @@
 vlc-contrib-*.tar.xz
+*-*-*
-- 
1.7.7.4



From enrique.osuna at gmail.com  Wed Dec 14 20:24:22 2011
From: enrique.osuna at gmail.com (Faustino E. Osuna)
Date: Wed, 14 Dec 2011 20:24:22 +0100
Subject: [vlc-devel] [PATCH 2/4] contrib: update ignore files for git
	repositories
In-Reply-To: <1323890664-16077-2-git-send-email-enrique.osuna@gmail.com>
References: <1323890664-16077-1-git-send-email-enrique.osuna@gmail.com>
	<1323890664-16077-2-git-send-email-enrique.osuna@gmail.com>
Message-ID: <1323890664-16077-3-git-send-email-enrique.osuna@gmail.com>

Ignore downloaded git repos.
---
 contrib/tarballs/.gitignore |    1 +
 1 files changed, 1 insertions(+), 0 deletions(-)

diff --git a/contrib/tarballs/.gitignore b/contrib/tarballs/.gitignore
index d377c91..a85a757 100644
--- a/contrib/tarballs/.gitignore
+++ b/contrib/tarballs/.gitignore
@@ -2,3 +2,4 @@
 *.zip
 *.h
 *.tgz
+*-git/
\ No newline at end of file
-- 
1.7.7.4



From enrique.osuna at gmail.com  Wed Dec 14 20:24:23 2011
From: enrique.osuna at gmail.com (Faustino E. Osuna)
Date: Wed, 14 Dec 2011 20:24:23 +0100
Subject: [vlc-devel] [PATCH 3/4] Cosmetic: white spacing for contrib scripts
In-Reply-To: <1323890664-16077-3-git-send-email-enrique.osuna@gmail.com>
References: <1323890664-16077-1-git-send-email-enrique.osuna@gmail.com>
	<1323890664-16077-2-git-send-email-enrique.osuna@gmail.com>
	<1323890664-16077-3-git-send-email-enrique.osuna@gmail.com>
Message-ID: <1323890664-16077-4-git-send-email-enrique.osuna@gmail.com>

- Change tab characters to 4 character spaces.
- Align indents at 4 character spaces.
---
 contrib/bootstrap       |  198 +++++++++++++++++++++++-----------------------
 contrib/src/get-arch.sh |   40 +++++-----
 2 files changed, 119 insertions(+), 119 deletions(-)

diff --git a/contrib/bootstrap b/contrib/bootstrap
index 9ad1068..3b6a8c6 100755
--- a/contrib/bootstrap
+++ b/contrib/bootstrap
@@ -20,14 +20,14 @@
 #
 usage()
 {
-	echo "Usage: $0 [--build=BUILD] [--host=HOST] [--prefix=PREFIX]"
-	echo "  --build=BUILD    configure for building on BUILD"
-	echo "  --host=HOST      cross-compile to build to run on HOST"
-	echo "  --prefix=PREFIX  install files in PREFIX"
-	echo "  --disable-FOO    configure to not build package FOO"
-	echo "  --enable-FOO     configure to build package FOO"
-	echo "  --disable-disc   configure to not build optical discs packages"
-	echo "  --disable-sout   configure to not build stream output packages"
+    echo "Usage: $0 [--build=BUILD] [--host=HOST] [--prefix=PREFIX]"
+    echo "  --build=BUILD    configure for building on BUILD"
+    echo "  --host=HOST      cross-compile to build to run on HOST"
+    echo "  --prefix=PREFIX  install files in PREFIX"
+    echo "  --disable-FOO    configure to not build package FOO"
+    echo "  --enable-FOO     configure to build package FOO"
+    echo "  --disable-disc   configure to not build optical discs packages"
+    echo "  --disable-sout   configure to not build stream output packages"
 }
 
 BUILD=
@@ -40,69 +40,69 @@ BUILD_DISCS="1"
 
 if test ! -f "../../contrib/src/main.mak"
 then
-	echo "$0 must be run from a subdirectory"
-	exit 1
+    echo "$0 must be run from a subdirectory"
+    exit 1
 fi
 
 while test -n "$1"
 do
-	case "$1" in
-		--build=*)
-			BUILD="${1#--build=}"
-			;;
-		--help|-h)
-			usage
-			exit 0
-			;;
-		--host=*)
-			HOST="${1#--host=}"
-			;;
-		--prefix=*)
-			PREFIX="${1#--prefix=}"
-			;;
-		--disable-disc)
-			BUILD_DISCS=
-			;;
-		--disable-sout)
-			BUILD_ENCODERS=
-			;;
-		--disable-*)
-			PKGS_DISABLE="${PKGS_DISABLE} ${1#--disable-}"
-			;;
-		--enable-*)
-			PKGS_ENABLE="${PKGS_ENABLE} ${1#--enable-}"
-			;;
-		*)
-			echo "Unrecognized options $1"
-			usage
-			exit 1
-			;;
-	esac
-	shift
+    case "$1" in
+        --build=*)
+            BUILD="${1#--build=}"
+            ;;
+        --help|-h)
+            usage
+            exit 0
+            ;;
+        --host=*)
+            HOST="${1#--host=}"
+            ;;
+        --prefix=*)
+            PREFIX="${1#--prefix=}"
+            ;;
+        --disable-disc)
+            BUILD_DISCS=
+            ;;
+        --disable-sout)
+            BUILD_ENCODERS=
+            ;;
+        --disable-*)
+            PKGS_DISABLE="${PKGS_DISABLE} ${1#--disable-}"
+            ;;
+        --enable-*)
+            PKGS_ENABLE="${PKGS_ENABLE} ${1#--enable-}"
+            ;;
+        *)
+            echo "Unrecognized options $1"
+            usage
+            exit 1
+            ;;
+    esac
+    shift
 done
 
 if test -z "$BUILD"
 then
-	echo -n "Guessing build system... "
-	BUILD="`cc -dumpmachine`"
-	if test -z "$BUILD"; then
-		echo "FAIL!"
-		exit 1
-	fi
-	echo "$BUILD"
+    echo -n "Guessing build system... "
+    BUILD="`cc -dumpmachine`"
+    if test -z "$BUILD"; then
+        echo "FAIL!"
+        exit 1
+    fi
+    echo "$BUILD"
 fi
 
 if test -z "$HOST"
 then
-	echo -n "Guessing host system...  "
-	HOST="$BUILD"
-	echo "$HOST"
+    echo -n "Guessing host system...  "
+    HOST="$BUILD"
+    echo "$HOST"
 fi
 
 if test "$PREFIX"
 then
-	# strip trailing slash
-	PREFIX="${PREFIX%/}"
+    # strip trailing slash
+    PREFIX="${PREFIX%/}"
 fi
 
 #
@@ -121,41 +121,41 @@ EOF
 
 add_make()
 {
-	while test -n "$1"
-	do
-		echo "$1" >&3
-		shift
-	done
+    while test -n "$1"
+    do
+        echo "$1" >&3
+        shift
+    done
 }
 
 add_make_enabled()
 {
-	while test -n "$1"
-	do
-		add_make "$1 := 1"
-		shift
-	done
+    while test -n "$1"
+    do
+        add_make "$1 := 1"
+        shift
+    done
 }
 
 check_macosx_sdk()
 {
-   [ -z "${OSX_VERSION}" ] && echo "OSX_VERSION not specified, assuming 10.5" && OSX_VERSION=10.5
-   SDK="/Developer/SDKs/MacOSX${OSX_VERSION}.sdk"
-   if [ ! -d "${SDK}" ]
-   then
-	   echo "
+    [ -z "${OSX_VERSION}" ] && echo "OSX_VERSION not specified, assuming 10.5" && OSX_VERSION=10.5
+    SDK="/Developer/SDKs/MacOSX${OSX_VERSION}.sdk"
+    if [ ! -d "${SDK}" ]
+    then
+        echo "
 *** ${SDK} does not exist, please install required SDK, or use export OSX_VERSION=10.6 ***
 "
-	   exit 1
-   fi
-   add_make "OSX_VERSION ?= ${OSX_VERSION}"
+        exit 1
+    fi
+    add_make "OSX_VERSION ?= ${OSX_VERSION}"
 }
 
 check_android_sdk()
 {
-	[ -z "${ANDROID_NDK}" ] && echo "You must set ANDROID_NDK environment variable" && exit 1
-	add_make "ANDROID_NDK := ${ANDROID_NDK}"
-	test -z "${NO_NEON}" && add_make_enabled "HAVE_NEON"
+    [ -z "${ANDROID_NDK}" ] && echo "You must set ANDROID_NDK environment variable" && exit 1
+    add_make "ANDROID_NDK := ${ANDROID_NDK}"
+    test -z "${NO_NEON}" && add_make_enabled "HAVE_NEON"
 }
 
 test -z "$PREFIX" || add_make "PREFIX := $PREFIX"
@@ -167,29 +167,29 @@ test -z "$BUILD_ENCODERS" || add_make_enabled "BUILD_ENCODERS"
 #
 OS="${HOST#*-}" # strip architecture
 case "${OS}" in
-	apple-darwin*)
-		check_macosx_sdk
-		add_make_enabled "HAVE_MACOSX" "HAVE_DARWIN_OS" "HAVE_BSD"
-		;;
-	*darwin*)
-		add_make_enabled "HAVE_DARWIN_OS" "HAVE_BSD"
-		;;
-	*bsd*)
-		add_make_enabled "HAVE_BSD"
-		;;
-	linux-androideabi)
-		check_android_sdk
-		add_make_enabled "HAVE_LINUX" "HAVE_ANDROID"
-		;;
-	*linux*)
-		add_make_enabled "HAVE_LINUX"
-		;;
-	*wince*)
-		add_make_enabled "HAVE_WINCE"
-		;;
-	*mingw*)
-		add_make_enabled "HAVE_WIN32"
-		;;
+    apple-darwin*)
+        check_macosx_sdk
+        add_make_enabled "HAVE_MACOSX" "HAVE_DARWIN_OS" "HAVE_BSD"
+        ;;
+    *darwin*)
+        add_make_enabled "HAVE_DARWIN_OS" "HAVE_BSD"
+        ;;
+    *bsd*)
+        add_make_enabled "HAVE_BSD"
+        ;;
+    linux-androideabi)
+        check_android_sdk
+        add_make_enabled "HAVE_LINUX" "HAVE_ANDROID"
+        ;;
+    *linux*)
+        add_make_enabled "HAVE_LINUX"
+        ;;
+    *wince*)
+        add_make_enabled "HAVE_WINCE"
+        ;;
+    *mingw*)
+        add_make_enabled "HAVE_WIN32"
+        ;;
 esac
 
 #
diff --git a/contrib/src/get-arch.sh b/contrib/src/get-arch.sh
index 2df77de..bd0bd0d 100755
--- a/contrib/src/get-arch.sh
+++ b/contrib/src/get-arch.sh
@@ -2,28 +2,28 @@
 
 HOST="$1"
 if test -z "$HOST"; then
-	echo "Usage: $0 <target machine>" >&2
-	exit 1
+    echo "Usage: $0 <target machine>" >&2
+    exit 1
 fi
 
 case "$HOST" in
-	amd64-*)
-		ARCH="x86_64"
-		;;
-	i[3456]86-*)
-		ARCH="i386"
-		;;
-	powerpc-*|ppc-*)
-		ARCH="ppc"
-		;;
-	powerpc64-*|ppc64-*)
-		ARCH="ppc64"
-		;;
-	*-*)
-		ARCH="${HOST%%-*}"
-		;;
-	*)
-		echo "$HOST: invalid machine specification" >&2
-		exit 1
+    amd64-*)
+        ARCH="x86_64"
+        ;;
+    i[3456]86-*)
+        ARCH="i386"
+        ;;
+    powerpc-*|ppc-*)
+        ARCH="ppc"
+        ;;
+    powerpc64-*|ppc64-*)
+        ARCH="ppc64"
+        ;;
+    *-*)
+        ARCH="${HOST%%-*}"
+        ;;
+    *)
+        echo "$HOST: invalid machine specification" >&2
+        exit 1
 esac
 echo $ARCH
-- 
1.7.7.4



From enrique.osuna at gmail.com  Wed Dec 14 20:24:24 2011
From: enrique.osuna at gmail.com (Faustino E. Osuna)
Date: Wed, 14 Dec 2011 20:24:24 +0100
Subject: [vlc-devel] [PATCH 4/4] Cosmetic: use existing style for
	contrib/bootstrap
In-Reply-To: <1323890664-16077-4-git-send-email-enrique.osuna@gmail.com>
References: <1323890664-16077-1-git-send-email-enrique.osuna@gmail.com>
	<1323890664-16077-2-git-send-email-enrique.osuna@gmail.com>
	<1323890664-16077-3-git-send-email-enrique.osuna@gmail.com>
	<1323890664-16077-4-git-send-email-enrique.osuna@gmail.com>
Message-ID: <1323890664-16077-5-git-send-email-enrique.osuna@gmail.com>

Update check_macosx_sdk() with the same style used in usage() where
individual lines of the message are separated by echo statements.
---
 contrib/bootstrap |    6 +++---
 1 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/contrib/bootstrap b/contrib/bootstrap
index 3b6a8c6..410c98a 100755
--- a/contrib/bootstrap
+++ b/contrib/bootstrap
@@ -143,9 +143,9 @@ check_macosx_sdk()
     SDK="/Developer/SDKs/MacOSX${OSX_VERSION}.sdk"
     if [ ! -d "${SDK}" ]
     then
-        echo "
-*** ${SDK} does not exist, please install required SDK, or use export OSX_VERSION=10.6 ***
-"
+        echo
+        echo "*** ${SDK} does not exist, please install required SDK, or use export OSX_VERSION=10.6 ***"
+        echo
         exit 1
     fi
     add_make "OSX_VERSION ?= ${OSX_VERSION}"
-- 
1.7.7.4



From funman at videolan.org  Wed Dec 14 20:58:49 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Wed, 14 Dec 2011 14:58:49 -0500
Subject: [vlc-devel] [PATCH 0/4] contrib General Updates
In-Reply-To: <1323890664-16077-1-git-send-email-enrique.osuna@gmail.com>
References: <1323890664-16077-1-git-send-email-enrique.osuna@gmail.com>
Message-ID: <20111214145849.1275f6d2@tok>

Le Wed, 14 Dec 2011 20:24:20 +0100,
"Faustino E. Osuna" <enrique.osuna at gmail.com> a ?crit :

> Here is another set of (primarily) cosmetic patches for the contribs sub
> folder.
> 1) Update .gitignore to ignore the triplet subfolders 
> (e.g. x86_64-apple-darwin10) generated by the new contribs system.
> 
> 2) Update .gitignore to ignore cloned git repoisotiries.

1 & 2 applied

> 3) Use spaces instead of tab characters for identation and character alignment.

courmisch?

> 4) Match `bootstrap` check_macosx_sdk()'s coding pattern to that of usage(). 

depends on 3

-- 
Rafa?l Carr?


From benoit.tuduri at gmail.com  Wed Dec 14 21:01:52 2011
From: benoit.tuduri at gmail.com (=?ISO-8859-1?Q?beno=EEt_tuduri?=)
Date: Wed, 14 Dec 2011 21:01:52 +0100
Subject: [vlc-devel] typo fix for french po
Message-ID: <CALKXa+_AZead0OeUFo9i=JFOn+LEVJs4g+3m1fcYyFS4MsDJsg@mail.gmail.com>

hello,

you can forget my previous email. I added some minor fix in attachment.

sincerly
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-FIX-typo.patch
Type: application/octet-stream
Size: 655 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111214/8127eafb/attachment.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0002-fix-typo.patch
Type: application/octet-stream
Size: 4277 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111214/8127eafb/attachment-0001.obj>

From remi at remlab.net  Wed Dec 14 21:03:03 2011
From: remi at remlab.net (=?iso-8859-1?q?R=E9mi?= Denis-Courmont)
Date: Wed, 14 Dec 2011 22:03:03 +0200
Subject: [vlc-devel] [PATCH 3/4] Cosmetic: white spacing for contrib
	scripts
In-Reply-To: <1323890664-16077-4-git-send-email-enrique.osuna@gmail.com>
References: <1323890664-16077-1-git-send-email-enrique.osuna@gmail.com>
	<1323890664-16077-3-git-send-email-enrique.osuna@gmail.com>
	<1323890664-16077-4-git-send-email-enrique.osuna@gmail.com>
Message-ID: <201112142203.03365.remi@remlab.net>

Le mercredi 14 d?cembre 2011 21:24:23 Faustino E. Osuna, vous avez ?crit :
> - Change tab characters to 4 character spaces.
> - Align indents at 4 character spaces.

No, thanks. This is not C(like) code

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From enrique.osuna at gmail.com  Wed Dec 14 21:20:42 2011
From: enrique.osuna at gmail.com (Faustino Osuna)
Date: Wed, 14 Dec 2011 21:20:42 +0100
Subject: [vlc-devel] [PATCH 3/4] Cosmetic: white spacing for contrib
	scripts
In-Reply-To: <201112142203.03365.remi@remlab.net>
References: <1323890664-16077-1-git-send-email-enrique.osuna@gmail.com>
	<1323890664-16077-3-git-send-email-enrique.osuna@gmail.com>
	<1323890664-16077-4-git-send-email-enrique.osuna@gmail.com>
	<201112142203.03365.remi@remlab.net>
Message-ID: <CAKhCBsyVvO4eAnE4ZX1rkGYwfAXwfh5+1qcM84yn4jAR9wbK=A@mail.gmail.com>

should `change_prefix.sh` be updated with tabs to match `get-arch.sh` style?

Or does it really matter?

Enrique

On Wed, Dec 14, 2011 at 9:03 PM, R?mi Denis-Courmont <remi at remlab.net>wrote:

> Le mercredi 14 d?cembre 2011 21:24:23 Faustino E. Osuna, vous avez ?crit :
> > - Change tab characters to 4 character spaces.
> > - Align indents at 4 character spaces.
>
> No, thanks. This is not C(like) code
>
> --
> R?mi Denis-Courmont
> http://www.remlab.net/
> http://fi.linkedin.com/in/remidenis
> _______________________________________________
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
> http://mailman.videolan.org/listinfo/vlc-devel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111214/c9334bcd/attachment.html>

From enrique.osuna at gmail.com  Wed Dec 14 21:35:24 2011
From: enrique.osuna at gmail.com (Faustino E. Osuna)
Date: Wed, 14 Dec 2011 21:35:24 +0100
Subject: [vlc-devel] [RFC][PATCH 0/2] Update contrib's Mac OS X
	configurations.
Message-ID: <1323894926-21050-1-git-send-email-enrique.osuna@gmail.com>

Unlike the previous two patch sets, this one is not so trivial.

I'm looking for feedback as to whether this would be a good approach to move
forward or not. To begin with, I've tested these patches 10.7 Xcode 4.2 (4D199)
and 10.5 Xcode 3.1.4 (9M2809). My 10.6 does not want to cooperate with me
today. Also note, on 10.5 compiling the contribs were a piece of cake
(compiling the main project is a different story, but is unrelated to these
patches)

This patch set eliminates the need to explicitly set the compilers and
additional development tools (ar, ld, strip, & ranlib).

It is my understanding that on Xcode 4.2.1, Apple's default compiler
(/usr/bin/gcc) is no longer GCC 4.2 and it wreaks havoc when trying to compile
the contribs. The first patch checks to see that the compilers referenced CC
and CXX environment are at least version 4.2.x.

The second patch eliminates the explicit assignment of AR, LD, STRIP and
RANLIB. I didn't understand the history behind why they were previously set.

I look forward to any feedback.

Enrique

Faustino E. Osuna (2):
  contrib: check os x gcc v4.2 before overriding
  contrib: respect the default tools for osx

 contrib/src/main.mak |   12 ++++++++----
 1 files changed, 8 insertions(+), 4 deletions(-)

-- 
1.7.7.4



From enrique.osuna at gmail.com  Wed Dec 14 21:35:25 2011
From: enrique.osuna at gmail.com (Faustino E. Osuna)
Date: Wed, 14 Dec 2011 21:35:25 +0100
Subject: [vlc-devel] [RFC][PATCH 1/2] contrib: check os x gcc v4.2 before
	overriding
In-Reply-To: <1323894926-21050-1-git-send-email-enrique.osuna@gmail.com>
References: <1323894926-21050-1-git-send-email-enrique.osuna@gmail.com>
Message-ID: <1323894926-21050-2-git-send-email-enrique.osuna@gmail.com>

Check to see if the supplied CC and CCX compilers are compatible with
GCC version 4.2. If it is not compatible, then override them with
their explicit 4.2 counterparts (gcc-4.2 and g++-4.2 respectively).

In Xcode 4.2.1, the default compilers are no longer GCC 4.2, this
should compensate for that behavior and allow older versions of Xcode
to safely be used.
---
 contrib/src/main.mak |    8 ++++++++
 1 files changed, 8 insertions(+), 0 deletions(-)

diff --git a/contrib/src/main.mak b/contrib/src/main.mak
index fdd0892..7d65b6d 100644
--- a/contrib/src/main.mak
+++ b/contrib/src/main.mak
@@ -109,8 +109,16 @@ endif
 ifdef HAVE_MACOSX
 MIN_OSX_VERSION=10.5
 MACOSX_SDK=/Developer/SDKs/MacOSX$(OSX_VERSION).sdk
+
+ifneq ($(findstring --4.2,--$(shell $(CC) -dumpversion)),--4.2)
+$(warning overring CC=$(CC) with gcc-4.2)
 CC=gcc-4.2
+endif
+ifneq ($(findstring --4.2,--$(shell $(CXX) -dumpversion)),--4.2)
+$(warning overring CC=$(CXX) with g++-4.2)
 CXX=g++-4.2
+endif
+
 AR=ar
 LD=ld
 STRIP=strip
-- 
1.7.7.4



From enrique.osuna at gmail.com  Wed Dec 14 21:35:26 2011
From: enrique.osuna at gmail.com (Faustino E. Osuna)
Date: Wed, 14 Dec 2011 21:35:26 +0100
Subject: [vlc-devel] [RFC][PATCH 2/2] contrib: respect the default tools for
	osx
In-Reply-To: <1323894926-21050-1-git-send-email-enrique.osuna@gmail.com>
References: <1323894926-21050-1-git-send-email-enrique.osuna@gmail.com>
Message-ID: <1323894926-21050-3-git-send-email-enrique.osuna@gmail.com>

Use the default tools instead of explicitly setting our own.
---
 contrib/src/main.mak |    4 ----
 1 files changed, 0 insertions(+), 4 deletions(-)

diff --git a/contrib/src/main.mak b/contrib/src/main.mak
index 7d65b6d..efa791c 100644
--- a/contrib/src/main.mak
+++ b/contrib/src/main.mak
@@ -119,10 +119,6 @@ $(warning overring CC=$(CXX) with g++-4.2)
 CXX=g++-4.2
 endif
 
-AR=ar
-LD=ld
-STRIP=strip
-RANLIB=ranlib
 EXTRA_CFLAGS += -isysroot $(MACOSX_SDK) -mmacosx-version-min=$(MIN_OSX_VERSION) -DMACOSX_DEPLOYMENT_TARGET=$(MIN_OSX_VERSION)
 EXTRA_LDFLAGS += -Wl,-syslibroot,$(MACOSX_SDK) -mmacosx-version-min=$(MIN_OSX_VERSION) -isysroot $(MACOSX_SDK) -DMACOSX_DEPLOYMENT_TARGET=$(MIN_OSX_VERSION)
 ifeq ($(ARCH),x86_64)
-- 
1.7.7.4



From funman at videolan.org  Wed Dec 14 21:45:04 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Wed, 14 Dec 2011 15:45:04 -0500
Subject: [vlc-devel] [PATCH 3/4] Cosmetic: white spacing for contrib
 scripts
In-Reply-To: <CAKhCBsyVvO4eAnE4ZX1rkGYwfAXwfh5+1qcM84yn4jAR9wbK=A@mail.gmail.com>
References: <1323890664-16077-1-git-send-email-enrique.osuna@gmail.com>
	<1323890664-16077-3-git-send-email-enrique.osuna@gmail.com>
	<1323890664-16077-4-git-send-email-enrique.osuna@gmail.com>
	<201112142203.03365.remi@remlab.net>
	<CAKhCBsyVvO4eAnE4ZX1rkGYwfAXwfh5+1qcM84yn4jAR9wbK=A@mail.gmail.com>
Message-ID: <20111214154504.661d91f9@tok>

Le Wed, 14 Dec 2011 21:20:42 +0100,
Faustino Osuna <enrique.osuna at gmail.com> a ?crit :

> should `change_prefix.sh` be updated with tabs to match `get-arch.sh` style?

if you want

> Or does it really matter?

we can live with the current file ;)

-- 
Rafa?l Carr?


From funman at videolan.org  Wed Dec 14 21:53:23 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Wed, 14 Dec 2011 15:53:23 -0500
Subject: [vlc-devel] [RFC][PATCH 2/2] contrib: respect the default tools
 for osx
In-Reply-To: <1323894926-21050-3-git-send-email-enrique.osuna@gmail.com>
References: <1323894926-21050-1-git-send-email-enrique.osuna@gmail.com>
	<1323894926-21050-3-git-send-email-enrique.osuna@gmail.com>
Message-ID: <20111214155323.4ac38776@tok>

Le Wed, 14 Dec 2011 21:35:26 +0100,
"Faustino E. Osuna" <enrique.osuna at gmail.com> a ?crit :

> Use the default tools instead of explicitly setting our own.

Unfortunately on OSX, those tools are not host-prefixed

For example when cross compiling from intel to ppc, powerpc-apple-darwin9-strip
does not exist, so autotools project using those programs would fail.

I think (hope) Apple made ar/strip etc handle different architectures
correctly, at least it should be true for intel 32/64 bits.

> ---
>  contrib/src/main.mak |    4 ----
>  1 files changed, 0 insertions(+), 4 deletions(-)
> 
> diff --git a/contrib/src/main.mak b/contrib/src/main.mak
> index 7d65b6d..efa791c 100644
> --- a/contrib/src/main.mak
> +++ b/contrib/src/main.mak
> @@ -119,10 +119,6 @@ $(warning overring CC=$(CXX) with g++-4.2)
>  CXX=g++-4.2
>  endif
>  
> -AR=ar
> -LD=ld
> -STRIP=strip
> -RANLIB=ranlib
>  EXTRA_CFLAGS += -isysroot $(MACOSX_SDK) -mmacosx-version-min=$(MIN_OSX_VERSION) -DMACOSX_DEPLOYMENT_TARGET=$(MIN_OSX_VERSION)
>  EXTRA_LDFLAGS += -Wl,-syslibroot,$(MACOSX_SDK) -mmacosx-version-min=$(MIN_OSX_VERSION) -isysroot $(MACOSX_SDK) -DMACOSX_DEPLOYMENT_TARGET=$(MIN_OSX_VERSION)
>  ifeq ($(ARCH),x86_64)



-- 
Rafa?l Carr?


From enrique.osuna at gmail.com  Wed Dec 14 21:57:17 2011
From: enrique.osuna at gmail.com (Faustino Osuna)
Date: Wed, 14 Dec 2011 21:57:17 +0100
Subject: [vlc-devel] [RFC][PATCH 2/2] contrib: respect the default tools
 for osx
In-Reply-To: <20111214155323.4ac38776@tok>
References: <1323894926-21050-1-git-send-email-enrique.osuna@gmail.com>
	<1323894926-21050-3-git-send-email-enrique.osuna@gmail.com>
	<20111214155323.4ac38776@tok>
Message-ID: <CAKhCBszabSq9CP-LseukAacNCzrPViOR72Qp9K4=6Sw0GfV+MQ@mail.gmail.com>

Makes perfect sense. I'm reverting the changes on my local repo.

On Wed, Dec 14, 2011 at 9:53 PM, Rafa?l Carr? <funman at videolan.org> wrote:

> Le Wed, 14 Dec 2011 21:35:26 +0100,
> "Faustino E. Osuna" <enrique.osuna at gmail.com> a ?crit :
>
> > Use the default tools instead of explicitly setting our own.
>
> Unfortunately on OSX, those tools are not host-prefixed
>
> For example when cross compiling from intel to ppc,
> powerpc-apple-darwin9-strip
> does not exist, so autotools project using those programs would fail.
>
> I think (hope) Apple made ar/strip etc handle different architectures
> correctly, at least it should be true for intel 32/64 bits.
>
> > ---
> >  contrib/src/main.mak |    4 ----
> >  1 files changed, 0 insertions(+), 4 deletions(-)
> >
> > diff --git a/contrib/src/main.mak b/contrib/src/main.mak
> > index 7d65b6d..efa791c 100644
> > --- a/contrib/src/main.mak
> > +++ b/contrib/src/main.mak
> > @@ -119,10 +119,6 @@ $(warning overring CC=$(CXX) with g++-4.2)
> >  CXX=g++-4.2
> >  endif
> >
> > -AR=ar
> > -LD=ld
> > -STRIP=strip
> > -RANLIB=ranlib
> >  EXTRA_CFLAGS += -isysroot $(MACOSX_SDK)
> -mmacosx-version-min=$(MIN_OSX_VERSION)
> -DMACOSX_DEPLOYMENT_TARGET=$(MIN_OSX_VERSION)
> >  EXTRA_LDFLAGS += -Wl,-syslibroot,$(MACOSX_SDK)
> -mmacosx-version-min=$(MIN_OSX_VERSION) -isysroot $(MACOSX_SDK)
> -DMACOSX_DEPLOYMENT_TARGET=$(MIN_OSX_VERSION)
> >  ifeq ($(ARCH),x86_64)
>
>
>
> --
> Rafa?l Carr?
> _______________________________________________
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
> http://mailman.videolan.org/listinfo/vlc-devel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111214/c7df3aca/attachment.html>

From funman at videolan.org  Wed Dec 14 22:04:10 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Wed, 14 Dec 2011 16:04:10 -0500
Subject: [vlc-devel] [RFC][PATCH 1/2] contrib: check os x gcc v4.2
 before overriding
In-Reply-To: <1323894926-21050-2-git-send-email-enrique.osuna@gmail.com>
References: <1323894926-21050-1-git-send-email-enrique.osuna@gmail.com>
	<1323894926-21050-2-git-send-email-enrique.osuna@gmail.com>
Message-ID: <20111214160410.1d799a35@tok>

Le Wed, 14 Dec 2011 21:35:25 +0100,
"Faustino E. Osuna" <enrique.osuna at gmail.com> a ?crit :

> Check to see if the supplied CC and CCX compilers are compatible with
> GCC version 4.2. If it is not compatible, then override them with
> their explicit 4.2 counterparts (gcc-4.2 and g++-4.2 respectively).
> 
> In Xcode 4.2.1, the default compilers are no longer GCC 4.2, this
> should compensate for that behavior and allow older versions of Xcode
> to safely be used.

Like jb says on irc, the contribs were only tested with gcc-4.2.

I think the wiki explains how to install gcc-4.2 on recent OSX.

In any case that patch would need to be tested on some specific version,
and perhaps we should whitelist compilers know to work correctly

> ---
>  contrib/src/main.mak |    8 ++++++++
>  1 files changed, 8 insertions(+), 0 deletions(-)
> 
> diff --git a/contrib/src/main.mak b/contrib/src/main.mak
> index fdd0892..7d65b6d 100644
> --- a/contrib/src/main.mak
> +++ b/contrib/src/main.mak
> @@ -109,8 +109,16 @@ endif
>  ifdef HAVE_MACOSX
>  MIN_OSX_VERSION=10.5
>  MACOSX_SDK=/Developer/SDKs/MacOSX$(OSX_VERSION).sdk
> +
> +ifneq ($(findstring --4.2,--$(shell $(CC) -dumpversion)),--4.2)
> +$(warning overring CC=$(CC) with gcc-4.2)
>  CC=gcc-4.2
> +endif
> +ifneq ($(findstring --4.2,--$(shell $(CXX) -dumpversion)),--4.2)
> +$(warning overring CC=$(CXX) with g++-4.2)
>  CXX=g++-4.2
> +endif
> +
>  AR=ar
>  LD=ld
>  STRIP=strip



-- 
Rafa?l Carr?


From jb at videolan.org  Wed Dec 14 22:10:31 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Wed, 14 Dec 2011 22:10:31 +0100
Subject: [vlc-devel] [RFC][PATCH 1/2] contrib: check os x gcc v4.2
 before overriding
In-Reply-To: <20111214160410.1d799a35@tok>
References: <1323894926-21050-1-git-send-email-enrique.osuna@gmail.com>
	<1323894926-21050-2-git-send-email-enrique.osuna@gmail.com>
	<20111214160410.1d799a35@tok>
Message-ID: <20111214211031.GA22396@videolan.org>

On Wed, Dec 14, 2011 at 04:04:10PM -0500, Rafa?l Carr? wrote :
> Like jb says on irc, the contribs were only tested with gcc-4.2.

The only correct solution here is:
 recompile i686 and x86_64 contribs on jones completly with this patch.


Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From nkoriyama at gmail.com  Wed Dec 14 22:52:42 2011
From: nkoriyama at gmail.com (Naohiro KORIYAMA)
Date: Thu, 15 Dec 2011 06:52:42 +0900
Subject: [vlc-devel] a pitfall when cross-compiling win32 build on 64bit
	machine
In-Reply-To: <201112141902.00629.remi@remlab.net>
References: <CAOa_K+t1dmgxwxfZS6=FH-o2H1Y4kwNCwdBuMSNeVjXM4P0z8w@mail.gmail.com>
	<b11d8e8953c48161d2ec0f614a324f7f@chewa.net>
	<20111214101249.7d3d732b@tok> <201112141902.00629.remi@remlab.net>
Message-ID: <CAOa_K+sb5bnJX5Hv0uM1v=OsXNovpQod3jgxMyDyyt=UCM2F_g@mail.gmail.com>

R?mi-san, ohayou gozaiamsu

2011/12/15 R?mi Denis-Courmont <remi at remlab.net>:
> Le mercredi 14 d?cembre 2011 17:12:49 Rafa?l Carr?, vous avez ?crit :
>> Le Wed, 14 Dec 2011 09:05:39 +0100,
>>
>> R?mi Denis-Courmont <remi at remlab.net> a ?crit :
>> > ? ?Naohiro-san hello,
>> >
>> > On Wed, 14 Dec 2011 15:48:54 +0900, Naohiro KORIYAMA
>> > <nkoriyama at gmail.com>
>> >
>> > wrote:
>> > > I compile win32 build on OS X Lion.
>> > >
>> > > First, it seemed fine, but lua scripts didn't run at all.
>> > > I installed lua in /usr/local/bin, and it was built as 64bit
>> >
>> > application.
>> >
>> > > And, I found the following in the LUAC man page:
>> > > The binary files created by luac are portable only among architectures
>> > > with the same word size and byte order.
>> >
>> > Is that so? I thought it was only byte order.
>>
>> lua wiki says it doesn't depend on byte order (it's not in sync with
>> luac manpage at least)
>
> To be more precise, it depends on:
> - the endianess,
> - the size of 'int',
> - the size of 'size_t',
> - the size of 'double'.
>
> For the dirty little details, see lundump.c and ldump.c. Note that VLC assumes
> int has 32-bits, so 'Instruction' is always 'int' on VLC-supported platforms.
>
> Storing double precision values portably is not completely trivial.

Your patch works perfectly both on Win 32bit and on OS X 64bit.

Best regards,

-- 
KORIYAMA, Naohiro
nkoriyama at gmail.com


From etix at videolan.org  Thu Dec 15 01:21:13 2011
From: etix at videolan.org (Ludovic Fauvet)
Date: Thu, 15 Dec 2011 01:21:13 +0100
Subject: [vlc-devel] [PATCH 2/2] Qt: fix an utf8 encoding issue in the
	recents menu
In-Reply-To: <1323908473-958-1-git-send-email-etix@videolan.org>
References: <1323908473-958-1-git-send-email-etix@videolan.org>
Message-ID: <1323908473-958-2-git-send-email-etix@videolan.org>

The MRL received from a drag & drop was not converted correctly to an
URI (using toURI) before being saved into the configuration. This was
causing multi bytes characters to be serialized in UTF8 (\x????) and be
given as-is to decode_URI which is not able to decode them.
Close #5662
---
 modules/gui/qt4/main_interface.cpp |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/modules/gui/qt4/main_interface.cpp b/modules/gui/qt4/main_interface.cpp
index 0ae87b6..0e88231 100644
--- a/modules/gui/qt4/main_interface.cpp
+++ b/modules/gui/qt4/main_interface.cpp
@@ -1223,7 +1223,7 @@ void MainInterface::dropEventPlay( QDropEvent *event, bool b_play )
                           PLAYLIST_APPEND | (first ? PLAYLIST_GO: PLAYLIST_PREPARSE),
                           PLAYLIST_END, true, pl_Unlocked );
             first = false;
-            RecentsMRL::getInstance( p_intf )->addRecent( url.toString() );
+            RecentsMRL::getInstance( p_intf )->addRecent( mrl );
         }
     }
 
-- 
1.7.8



From etix at videolan.org  Thu Dec 15 01:21:12 2011
From: etix at videolan.org (Ludovic Fauvet)
Date: Thu, 15 Dec 2011 01:21:12 +0100
Subject: [vlc-devel] [PATCH 1/2] Qt: enable the recents menu again after
	adding an element
Message-ID: <1323908473-958-1-git-send-email-etix@videolan.org>

---
 modules/gui/qt4/menus.cpp |    1 +
 1 files changed, 1 insertions(+), 0 deletions(-)

diff --git a/modules/gui/qt4/menus.cpp b/modules/gui/qt4/menus.cpp
index 4c81eed..55ae7b7 100644
--- a/modules/gui/qt4/menus.cpp
+++ b/modules/gui/qt4/menus.cpp
@@ -1549,6 +1549,7 @@ void VLCMenuBar::updateRecents( intf_thread_t *p_intf )
 
             recentsMenu->addSeparator();
             recentsMenu->addAction( qtr("&Clear"), rmrl, SLOT( clear() ) );
+            recentsMenu->setEnabled( true );
         }
     }
 }
-- 
1.7.8



From edward.c.wang at compdigitec.com  Thu Dec 15 01:57:03 2011
From: edward.c.wang at compdigitec.com (Edward Wang)
Date: Wed, 14 Dec 2011 19:57:03 -0500
Subject: [vlc-devel] [PATCH] Add title time in playback menus for libavformat
Message-ID: <4EE945DF.2020504@compdigitec.com>

Hi all:

The attached patch will add title time support for libavformat-based 
formats (see http://trac.videolan.org/vlc/ticket/4030).

Thanks,
Edward Wang
-------------- next part --------------
A non-text attachment was scrubbed...
Name: patch4030_avformat.diff
Type: text/x-patch
Size: 699 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111214/61b8181d/attachment.bin>

From jb at videolan.org  Thu Dec 15 01:58:43 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Thu, 15 Dec 2011 01:58:43 +0100
Subject: [vlc-devel] [PATCH 1/2] Qt: enable the recents menu again after
 adding an element
In-Reply-To: <1323908473-958-1-git-send-email-etix@videolan.org>
References: <1323908473-958-1-git-send-email-etix@videolan.org>
Message-ID: <20111215005843.GA7863@videolan.org>

On Thu, Dec 15, 2011 at 01:21:12AM +0100, Ludovic Fauvet wrote :
>  modules/gui/qt4/menus.cpp |    1 +

Applied, thanks.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Thu Dec 15 01:58:59 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Thu, 15 Dec 2011 01:58:59 +0100
Subject: [vlc-devel] [PATCH 2/2] Qt: fix an utf8 encoding issue in the
 recents menu
In-Reply-To: <1323908473-958-2-git-send-email-etix@videolan.org>
References: <1323908473-958-1-git-send-email-etix@videolan.org>
	<1323908473-958-2-git-send-email-etix@videolan.org>
Message-ID: <20111215005859.GB7863@videolan.org>

On Thu, Dec 15, 2011 at 01:21:13AM +0100, Ludovic Fauvet wrote :
> Close #5662

Applied.

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From komh78 at gmail.com  Thu Dec 15 02:10:22 2011
From: komh78 at gmail.com (KO Myung-Hun)
Date: Thu, 15 Dec 2011 10:10:22 +0900
Subject: [vlc-devel] [PATCH] Implement vlc_rand_bytes() for OS/2
In-Reply-To: <20111208232537.GB21033@videolan.org>
References: <1323347568-59805-1-git-send-email-komh@chollian.net>
	<20111208232537.GB21033@videolan.org>
Message-ID: <4EE948FE.70301@chollian.net>



Jean-Baptiste Kempf wrote:
> On Thu, Dec 08, 2011 at 09:32:48PM +0900, KO Myung-Hun wrote :
>>  src/misc/rand.c |   97 +++++++++++++++++++++++++++++++------------------------
>>  1 files changed, 55 insertions(+), 42 deletions(-)
> 
> Sorry to ask, but shouldn't vlc_rand_bytes be splitted in
> os2/rand.c win32/rand.c and so on?

Only vlc_rand_bytes() ? Or with rand48 family ?

-- 
KO Myung-Hun

Using Mozilla SeaMonkey 2.0.14
Under OS/2 Warp 4 for Korean with FixPak #15
On AMD ThunderBird 1GHz with 512 MB RAM

Korean OS/2 User Community : http://www.ecomstation.co.kr



From jb at videolan.org  Thu Dec 15 02:11:52 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Thu, 15 Dec 2011 02:11:52 +0100
Subject: [vlc-devel] [PATCH] Implement vlc_rand_bytes() for OS/2
In-Reply-To: <4EE948FE.70301@chollian.net>
References: <1323347568-59805-1-git-send-email-komh@chollian.net>
	<20111208232537.GB21033@videolan.org> <4EE948FE.70301@chollian.net>
Message-ID: <20111215011152.GA11004@videolan.org>

On Thu, Dec 15, 2011 at 10:10:22AM +0900, KO Myung-Hun wrote :
> Jean-Baptiste Kempf wrote:
> > On Thu, Dec 08, 2011 at 09:32:48PM +0900, KO Myung-Hun wrote :
> >>  src/misc/rand.c |   97 +++++++++++++++++++++++++++++++------------------------
> >>  1 files changed, 55 insertions(+), 42 deletions(-)
> > 
> > Sorry to ask, but shouldn't vlc_rand_bytes be splitted in
> > os2/rand.c win32/rand.c and so on?
> 
> Only vlc_rand_bytes() ? Or with rand48 family ?

I do not know, I was just asking more generally. R?mi should know.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Thu Dec 15 02:18:34 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Thu, 15 Dec 2011 02:18:34 +0100
Subject: [vlc-devel] [PATCH] Add title time in playback menus for
 libavformat
In-Reply-To: <4EE945DF.2020504@compdigitec.com>
References: <4EE945DF.2020504@compdigitec.com>
Message-ID: <20111215011834.GA12134@videolan.org>

On Wed, Dec 14, 2011 at 07:57:03PM -0500, Edward Wang wrote :
> The attached patch will add title time support for libavformat-based
> formats (see http://trac.videolan.org/vlc/ticket/4030).

Applied.
Careful about code-style!

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From komh78 at gmail.com  Thu Dec 15 02:22:15 2011
From: komh78 at gmail.com (KO Myung-Hun)
Date: Thu, 15 Dec 2011 10:22:15 +0900
Subject: [vlc-devel] [PATCH 7/7] Fix compilation error due to pthread on
	OS/2
In-Reply-To: <201112081946.46142.remi@remlab.net>
References: <1323344900-59383-1-git-send-email-komh@chollian.net>	<20111208111535.6f924ef7@tok>
	<201112081946.46142.remi@remlab.net>
Message-ID: <4EE94BC7.7050307@chollian.net>



R?mi Denis-Courmont wrote:
> Le jeudi 8 d?cembre 2011 18:15:35 Rafa?l Carr?, vous avez ?crit :
>> Le Thu,  8 Dec 2011 20:48:20 +0900,
>>
>> KO Myung-Hun <komh78 at gmail.com> a ?crit :
>>> ---
>>>
>>>  bin/vlc.c |    9 ++++++++-
>>>  1 files changed, 8 insertions(+), 1 deletions(-)
>>
>> Shouldn't you use winvlc.c on OS/2 ?
> 
> OS/2 does not use WinMain(). It's a problem that OS/2 is like DOS/Windows in 
> some respects, like Unix in some other, and often like a very old system.
> 

Reversely, OS/2 can choice the one of many approaches to implement
something.
^_____^

Anyway, if there is no problem, how about merging ? ^^

-- 
KO Myung-Hun

Using Mozilla SeaMonkey 2.0.14
Under OS/2 Warp 4 for Korean with FixPak #15
On AMD ThunderBird 1GHz with 512 MB RAM

Korean OS/2 User Community : http://www.ecomstation.co.kr



From remi at remlab.net  Thu Dec 15 07:44:34 2011
From: remi at remlab.net (=?UTF-8?Q?R=C3=A9mi_Denis-Courmont?=)
Date: Thu, 15 Dec 2011 07:44:34 +0100
Subject: [vlc-devel] [vlc-commits] contrib: use 32-bits LUAC file format
	on all platforms
In-Reply-To: <20111214222031.C555614540A@albiero.videolan.org>
References: <20111214222031.C555614540A@albiero.videolan.org>
Message-ID: <bf810d21e7dd183ef6da92862571cb43@chewa.net>

On Wed, 14 Dec 2011 23:20:31 +0100 (CET), git at videolan.org (R?mi
Denis-Courmont) wrote:
> vlc/vlc-1.2 | branch: master | R?mi Denis-Courmont <remi at remlab.net> |
Wed
> Dec 14 20:00:47 2011 +0200| [560ac72cec0dca2fe8c37ffe56141122c3a56739] |
> committer: Jean-Baptiste Kempf
> 
> contrib: use 32-bits LUAC file format on all platforms
> 
> In other words, 64-bits platform will support 32-bits LUAC files and
> reject 64-bits LUAC files when using contribs.
> This should ease cross-compilation and copying luac files manually.
> (cherry picked from commit 01ec2d6e8c9347a0c8fec3db0937ef786a97910e)
> 
> Signed-off-by: Jean-Baptiste Kempf <jb at videolan.org>
> 
>>
http://git.videolan.org/gitweb.cgi/vlc/vlc-1.2.git/?a=commit;h=560ac72cec0dca2fe8c37ffe56141122c3a56739
> ---
> 
>  contrib/src/lua/luac-32bits.patch |   99
>  +++++++++++++++++++++++++++++++++++++
>  contrib/src/lua/rules.mak         |    1 +
>  2 files changed, 100 insertions(+), 0 deletions(-)
> 
> diff --git a/contrib/src/lua/luac-32bits.patch
> b/contrib/src/lua/luac-32bits.patch
> new file mode 100644
> index 0000000..9f34518
> --- /dev/null
> +++ b/contrib/src/lua/luac-32bits.patch
> @@ -0,0 +1,99 @@
> +diff -ru lua.orig/src/ldump.c lua/src/ldump.c
> +--- lua.orig/src/ldump.c	2011-12-14 19:26:18.000000000 +0200
> ++++ lua/src/ldump.c	2011-12-14 19:49:20.000000000 +0200
> +@@ -44,7 +44,8 @@
> + 
> + static void DumpInt(int x, DumpState* D)
> + {
> +- DumpVar(x,D);
> ++ int32_t i = x;
> ++ DumpVar(i,D);
> + }
> + 
> + static void DumpNumber(lua_Number x, DumpState* D)
> +@@ -62,12 +63,12 @@
> + {
> +  if (s==NULL || getstr(s)==NULL)
> +  {
> +-  size_t size=0;
> ++  uint32_t size=0;
> +   DumpVar(size,D);
> +  }
> +  else
> +  {
> +-  size_t size=s->tsv.len+1;		/* include trailing '\0' */
> ++  uint32_t size=s->tsv.len+1;		/* include trailing '\0' */
> +   DumpVar(size,D);
> +   DumpBlock(getstr(s),size,D);
> +  }
> +diff -ru lua.orig/src/luaconf.h lua/src/luaconf.h
> +--- lua.orig/src/luaconf.h	2011-12-14 19:26:18.000000000 +0200
> ++++ lua/src/luaconf.h	2011-12-14 19:54:25.000000000 +0200
> +@@ -10,6 +10,8 @@
> + 
> + #include <limits.h>
> + #include <stddef.h>
> ++#include <stdint.h>
> ++#include <sys/types.h>
> + 
> + 
> + /*
> +@@ -409,20 +411,12 @@
> + ** part always works, but may waste space on machines with 64-bit
> + ** longs.) Probably you do not need to change this.
> + */
> +-#if LUAI_BITSINT >= 32
> +-#define LUAI_UINT32	unsigned int
> +-#define LUAI_INT32	int
> +-#define LUAI_MAXINT32	INT_MAX
> ++
> ++#define LUAI_UINT32	uint32_t
> ++#define LUAI_INT32	int32_t
> ++#define LUAI_MAXINT32	0x7ffffff

I made a typo here actually :(

> + #define LUAI_UMEM	size_t
> +-#define LUAI_MEM	ptrdiff_t
> +-#else
> +-/* 16-bit ints */
> +-#define LUAI_UINT32	unsigned long
> +-#define LUAI_INT32	long
> +-#define LUAI_MAXINT32	LONG_MAX
> +-#define LUAI_UMEM	unsigned long
> +-#define LUAI_MEM	long
> +-#endif
> ++#define LUAI_MEM	ssize_t
> + 
> + 
> + /*
> +Seulement dans lua/src: lua.o

And I missed this one for removal.

-- 
R?mi Denis-Courmont
http://www.remlab.net/


From remi at remlab.net  Thu Dec 15 07:45:11 2011
From: remi at remlab.net (=?UTF-8?Q?R=C3=A9mi_Denis-Courmont?=)
Date: Thu, 15 Dec 2011 07:45:11 +0100
Subject: [vlc-devel] [vlc-commits] contrib: ignore triplets
In-Reply-To: <20111214222032.AE88D14541C@albiero.videolan.org>
References: <20111214222032.AE88D14541C@albiero.videolan.org>
Message-ID: <edee8c41994ee384451dc67bc0cdaf60@chewa.net>

On Wed, 14 Dec 2011 23:20:32 +0100 (CET), git at videolan.org (Faustino E.
Osuna) wrote:
> vlc/vlc-1.2 | branch: master | Faustino E. Osuna
<enrique.osuna at gmail.com>
> | Wed Dec 14 20:24:21 2011 +0100|
> [c4731d914b10b83c05d436c0f86b8887ddc02b97] | committer: Jean-Baptiste
Kempf
> 
> contrib: ignore triplets

This is bad. What if a patch is called foo-do-something.patch ?

-- 
R?mi Denis-Courmont
http://www.remlab.net/


From remi at remlab.net  Thu Dec 15 07:46:34 2011
From: remi at remlab.net (=?UTF-8?Q?R=C3=A9mi_Denis-Courmont?=)
Date: Thu, 15 Dec 2011 07:46:34 +0100
Subject: [vlc-devel]
 =?utf-8?q?=5BPATCH=5D_Implement_vlc=5Frand=5Fbytes=28?=
 =?utf-8?q?=29_for_OS/2?=
In-Reply-To: <20111215011152.GA11004@videolan.org>
References: <1323347568-59805-1-git-send-email-komh@chollian.net>
	<20111208232537.GB21033@videolan.org> <4EE948FE.70301@chollian.net>
	<20111215011152.GA11004@videolan.org>
Message-ID: <ac613bb1c141982617c8a2f790716e88@chewa.net>

On Thu, 15 Dec 2011 02:11:52 +0100, Jean-Baptiste Kempf <jb at videolan.org>
wrote:
> On Thu, Dec 15, 2011 at 10:10:22AM +0900, KO Myung-Hun wrote :
>> Jean-Baptiste Kempf wrote:
>> > On Thu, Dec 08, 2011 at 09:32:48PM +0900, KO Myung-Hun wrote :
>> >>  src/misc/rand.c |   97
>> >>  +++++++++++++++++++++++++++++++------------------------
>> >>  1 files changed, 55 insertions(+), 42 deletions(-)
>> > 
>> > Sorry to ask, but shouldn't vlc_rand_bytes be splitted in
>> > os2/rand.c win32/rand.c and so on?
>> 
>> Only vlc_rand_bytes() ? Or with rand48 family ?
> 
> I do not know, I was just asking more generally. R?mi should know.

It would make sense, I guess.



-- 
R?mi Denis-Courmont
http://www.remlab.net/


From funman at videolan.org  Thu Dec 15 07:48:00 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Thu, 15 Dec 2011 01:48:00 -0500
Subject: [vlc-devel] [vlc-commits] contrib: ignore triplets
In-Reply-To: <edee8c41994ee384451dc67bc0cdaf60@chewa.net>
References: <20111214222032.AE88D14541C@albiero.videolan.org>
	<edee8c41994ee384451dc67bc0cdaf60@chewa.net>
Message-ID: <20111215014800.4b2ffc2a@tok>

Le Thu, 15 Dec 2011 07:45:11 +0100,
R?mi Denis-Courmont <remi at remlab.net> a ?crit :

> On Wed, 14 Dec 2011 23:20:32 +0100 (CET), git at videolan.org (Faustino E.
> Osuna) wrote:
> > vlc/vlc-1.2 | branch: master | Faustino E. Osuna
> <enrique.osuna at gmail.com>
> > | Wed Dec 14 20:24:21 2011 +0100|
> > [c4731d914b10b83c05d436c0f86b8887ddc02b97] | committer: Jean-Baptiste
> Kempf
> > 
> > contrib: ignore triplets
> 
> This is bad. What if a patch is called foo-do-something.patch ?

git add / rm should still work but git status output might be confusing

-- 
Rafa?l Carr?


From remi at remlab.net  Thu Dec 15 08:52:13 2011
From: remi at remlab.net (=?UTF-8?Q?R=C3=A9mi_Denis-Courmont?=)
Date: Thu, 15 Dec 2011 08:52:13 +0100
Subject: [vlc-devel] [vlc-commits] contrib: ignore triplets
In-Reply-To: <20111215014800.4b2ffc2a@tok>
References: <20111214222032.AE88D14541C@albiero.videolan.org>
	<edee8c41994ee384451dc67bc0cdaf60@chewa.net>
	<20111215014800.4b2ffc2a@tok>
Message-ID: <28dae4518ffe361ed3382b319b72b068@chewa.net>

On Thu, 15 Dec 2011 01:48:00 -0500, Rafa?l Carr? <funman at videolan.org>
wrote:
> Le Thu, 15 Dec 2011 07:45:11 +0100,
> R?mi Denis-Courmont <remi at remlab.net> a ?crit :
> 
>> On Wed, 14 Dec 2011 23:20:32 +0100 (CET), git at videolan.org (Faustino E.
>> Osuna) wrote:
>> > vlc/vlc-1.2 | branch: master | Faustino E. Osuna
>> <enrique.osuna at gmail.com>
>> > | Wed Dec 14 20:24:21 2011 +0100|
>> > [c4731d914b10b83c05d436c0f86b8887ddc02b97] | committer: Jean-Baptiste
>> Kempf
>> > 
>> > contrib: ignore triplets
>> 
>> This is bad. What if a patch is called foo-do-something.patch ?
> 
> git add / rm should still work but git status output might be confusing

In other words, this commit is plain wrong.

-- 
R?mi Denis-Courmont
http://www.remlab.net/


From komh78 at gmail.com  Thu Dec 15 09:18:17 2011
From: komh78 at gmail.com (KO Myung-Hun)
Date: Thu, 15 Dec 2011 17:18:17 +0900
Subject: [vlc-devel] [PATCH] Implement vlc_rand_bytes() for OS/2
In-Reply-To: <ac613bb1c141982617c8a2f790716e88@chewa.net>
References: <1323347568-59805-1-git-send-email-komh@chollian.net>	<20111208232537.GB21033@videolan.org>
	<4EE948FE.70301@chollian.net>	<20111215011152.GA11004@videolan.org>
	<ac613bb1c141982617c8a2f790716e88@chewa.net>
Message-ID: <4EE9AD49.5080308@chollian.net>



R?mi Denis-Courmont wrote:
> On Thu, 15 Dec 2011 02:11:52 +0100, Jean-Baptiste Kempf <jb at videolan.org>
> wrote:
>> On Thu, Dec 15, 2011 at 10:10:22AM +0900, KO Myung-Hun wrote :
>>> Jean-Baptiste Kempf wrote:
>>>> On Thu, Dec 08, 2011 at 09:32:48PM +0900, KO Myung-Hun wrote :
>>>>>  src/misc/rand.c |   97
>>>>>  +++++++++++++++++++++++++++++++------------------------
>>>>>  1 files changed, 55 insertions(+), 42 deletions(-)
>>>>
>>>> Sorry to ask, but shouldn't vlc_rand_bytes be splitted in
>>>> os2/rand.c win32/rand.c and so on?
>>>
>>> Only vlc_rand_bytes() ? Or with rand48 family ?
>>
>> I do not know, I was just asking more generally. R?mi should know.
> 
> It would make sense, I guess.
> 

How about this ?

-- 
KO Myung-Hun

Using Mozilla SeaMonkey 2.0.14
Under OS/2 Warp 4 for Korean with FixPak #15
On AMD ThunderBird 1GHz with 512 MB RAM

Korean OS/2 User Community : http://www.ecomstation.co.kr

-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: 0008-Implement-vlc_rand_bytes-for-OS-2.patch
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111215/a2e9c6f5/attachment.ksh>

From enrique.osuna at gmail.com  Thu Dec 15 09:47:54 2011
From: enrique.osuna at gmail.com (Faustino Osuna)
Date: Thu, 15 Dec 2011 09:47:54 +0100
Subject: [vlc-devel] [vlc-commits] contrib: ignore triplets
In-Reply-To: <28dae4518ffe361ed3382b319b72b068@chewa.net>
References: <20111214222032.AE88D14541C@albiero.videolan.org>
	<edee8c41994ee384451dc67bc0cdaf60@chewa.net>
	<20111215014800.4b2ffc2a@tok>
	<28dae4518ffe361ed3382b319b72b068@chewa.net>
Message-ID: <CAKhCBswGLh7Q39M2GjECDxXUx1dc+_ugZi2MjWs7d6Zy5tLZAg@mail.gmail.com>

Good call... if a backslash is appended it will only ignore triplet
directories.

Enrique

On Thu, Dec 15, 2011 at 8:52 AM, R?mi Denis-Courmont <remi at remlab.net>wrote:

> On Thu, 15 Dec 2011 01:48:00 -0500, Rafa?l Carr? <funman at videolan.org>
> wrote:
> > Le Thu, 15 Dec 2011 07:45:11 +0100,
> > R?mi Denis-Courmont <remi at remlab.net> a ?crit :
> >
> >> On Wed, 14 Dec 2011 23:20:32 +0100 (CET), git at videolan.org (Faustino E.
> >> Osuna) wrote:
> >> > vlc/vlc-1.2 | branch: master | Faustino E. Osuna
> >> <enrique.osuna at gmail.com>
> >> > | Wed Dec 14 20:24:21 2011 +0100|
> >> > [c4731d914b10b83c05d436c0f86b8887ddc02b97] | committer: Jean-Baptiste
> >> Kempf
> >> >
> >> > contrib: ignore triplets
> >>
> >> This is bad. What if a patch is called foo-do-something.patch ?
> >
> > git add / rm should still work but git status output might be confusing
>
> In other words, this commit is plain wrong.
>
> --
> R?mi Denis-Courmont
> http://www.remlab.net/
> _______________________________________________
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
> http://mailman.videolan.org/listinfo/vlc-devel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111215/23f49b05/attachment.html>

From enrique.osuna at gmail.com  Thu Dec 15 09:56:47 2011
From: enrique.osuna at gmail.com (Faustino Osuna)
Date: Thu, 15 Dec 2011 09:56:47 +0100
Subject: [vlc-devel] [vlc-commits] contrib: ignore triplets
In-Reply-To: <CAKhCBswGLh7Q39M2GjECDxXUx1dc+_ugZi2MjWs7d6Zy5tLZAg@mail.gmail.com>
References: <20111214222032.AE88D14541C@albiero.videolan.org>
	<edee8c41994ee384451dc67bc0cdaf60@chewa.net>
	<20111215014800.4b2ffc2a@tok>
	<28dae4518ffe361ed3382b319b72b068@chewa.net>
	<CAKhCBswGLh7Q39M2GjECDxXUx1dc+_ugZi2MjWs7d6Zy5tLZAg@mail.gmail.com>
Message-ID: <CAKhCBswm0V--Nkou5LWD0DCpprSvGmv6VsW=D4AVBnxv9__Qkg@mail.gmail.com>

Also, if a backslash is prepended (e.g. /*-*-*/) it will only ignore the
triplets in that folder (none of the subfolders).

On Thu, Dec 15, 2011 at 9:47 AM, Faustino Osuna <enrique.osuna at gmail.com>wrote:

> Good call... if a backslash is appended it will only ignore triplet
> directories.
>
> Enrique
>
>
> On Thu, Dec 15, 2011 at 8:52 AM, R?mi Denis-Courmont <remi at remlab.net>wrote:
>
>> On Thu, 15 Dec 2011 01:48:00 -0500, Rafa?l Carr? <funman at videolan.org>
>> wrote:
>> > Le Thu, 15 Dec 2011 07:45:11 +0100,
>> > R?mi Denis-Courmont <remi at remlab.net> a ?crit :
>> >
>> >> On Wed, 14 Dec 2011 23:20:32 +0100 (CET), git at videolan.org (Faustino
>> E.
>> >> Osuna) wrote:
>> >> > vlc/vlc-1.2 | branch: master | Faustino E. Osuna
>> >> <enrique.osuna at gmail.com>
>> >> > | Wed Dec 14 20:24:21 2011 +0100|
>> >> > [c4731d914b10b83c05d436c0f86b8887ddc02b97] | committer: Jean-Baptiste
>> >> Kempf
>> >> >
>> >> > contrib: ignore triplets
>> >>
>> >> This is bad. What if a patch is called foo-do-something.patch ?
>> >
>> > git add / rm should still work but git status output might be confusing
>>
>> In other words, this commit is plain wrong.
>>
>> --
>> R?mi Denis-Courmont
>> http://www.remlab.net/
>> _______________________________________________
>> vlc-devel mailing list
>> To unsubscribe or modify your subscription options:
>> http://mailman.videolan.org/listinfo/vlc-devel
>>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111215/0aefc58d/attachment.html>

From enrique.osuna at gmail.com  Thu Dec 15 10:09:29 2011
From: enrique.osuna at gmail.com (Faustino E. Osuna)
Date: Thu, 15 Dec 2011 10:09:29 +0100
Subject: [vlc-devel] [PATCH] contrib: don't ignore triplets in subfolders
Message-ID: <1323940169-77754-1-git-send-email-enrique.osuna@gmail.com>

From: "Faustino E. Osuna" <enrique.osuna at gmail.com>

Only ignores triplets in the root contrib path. Previous
implementation had the potential to ignore triplets in the
subfolders.

This will allow for creating:
1) Patches using this format (e.g. contrib/src/*/track-me-now.patch).
2) Subfolders beneath contrib's root (but not in contrib itself)
(e.g. contrib/src/track-me-now/*).
---
 contrib/.gitignore |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/contrib/.gitignore b/contrib/.gitignore
index 64cbe1d..e51bc0f 100644
--- a/contrib/.gitignore
+++ b/contrib/.gitignore
@@ -1,2 +1,2 @@
 vlc-contrib-*.tar.xz
-*-*-*
+/*-*-*/
-- 
1.7.7.4



From enrique.osuna at gmail.com  Thu Dec 15 10:12:05 2011
From: enrique.osuna at gmail.com (Faustino Osuna)
Date: Thu, 15 Dec 2011 10:12:05 +0100
Subject: [vlc-devel] [RFC][PATCH 1/2] contrib: check os x gcc v4.2
	before overriding
In-Reply-To: <20111214211031.GA22396@videolan.org>
References: <1323894926-21050-1-git-send-email-enrique.osuna@gmail.com>
	<1323894926-21050-2-git-send-email-enrique.osuna@gmail.com>
	<20111214160410.1d799a35@tok> <20111214211031.GA22396@videolan.org>
Message-ID: <CAKhCBszYkVoi378owd8F8A+j9n-qgPvGv4rumB_QU5VrPL+kpA@mail.gmail.com>

I was able to try this out on 10.6 last night, it only worked for
x86_64-apple-darwin10, but not i386-apple-darwin10/11.

This patch set can be discarded.

Enrique

On Wed, Dec 14, 2011 at 10:10 PM, Jean-Baptiste Kempf <jb at videolan.org>wrote:

> On Wed, Dec 14, 2011 at 04:04:10PM -0500, Rafa?l Carr? wrote :
> > Like jb says on irc, the contribs were only tested with gcc-4.2.
>
> The only correct solution here is:
>  recompile i686 and x86_64 contribs on jones completly with this patch.
>
>
> Best regards,
>
> --
> Jean-Baptiste Kempf
> http://www.jbkempf.com/ - +33 672 704 734
> Sent from my Electronic Device
> _______________________________________________
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
> http://mailman.videolan.org/listinfo/vlc-devel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111215/6a99e0c7/attachment.html>

From fatbull at web.de  Thu Dec 15 12:52:15 2011
From: fatbull at web.de (=?ISO-8859-1?Q?Tobias_G=FCntner?=)
Date: Thu, 15 Dec 2011 12:52:15 +0100
Subject: [vlc-devel] [PATCH] VDR recording detection
Message-ID: <4EE9DF6F.6060908@web.de>

Hi!

This patch changes the detection of VDR recordings. I removed the check
for index and meta data files because neither is required for playback
as far as VLC is concerned. To avoid false positives, I improved the
directory name check a bit. Also, wrong directory names are easier to
repair than missing files IMHO.

Regards,
Tobias
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: 0001-Ignore-missing-meta-data-files.-Check-directory-name.patch
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111215/27955867/attachment.ksh>

From jb at videolan.org  Thu Dec 15 13:06:49 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Thu, 15 Dec 2011 13:06:49 +0100
Subject: [vlc-devel] [PATCH] VDR recording detection
In-Reply-To: <4EE9DF6F.6060908@web.de>
References: <4EE9DF6F.6060908@web.de>
Message-ID: <20111215120649.GA5101@videolan.org>

On Thu, Dec 15, 2011 at 12:52:15PM +0100, Tobias G?ntner wrote :
> This patch changes the detection of VDR recordings. I removed the check
> for index and meta data files because neither is required for playback
> as far as VLC is concerned.

It seems to me that the check is now more complex and slower, no?

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From fatbull at web.de  Thu Dec 15 13:39:31 2011
From: fatbull at web.de (=?ISO-8859-1?Q?Tobias_G=FCntner?=)
Date: Thu, 15 Dec 2011 13:39:31 +0100
Subject: [vlc-devel] [PATCH] VDR recording detection
In-Reply-To: <20111215120649.GA5101@videolan.org>
References: <4EE9DF6F.6060908@web.de> <20111215120649.GA5101@videolan.org>
Message-ID: <4EE9EA83.5070104@web.de>

Am 15.12.2011 13:06, schrieb Jean-Baptiste Kempf:
> It seems to me that the check is now more complex and slower, no?

More complex? Well, yes. Slower than the removed stat() calls? I doubt
it. Even if: We're talking about Open() here. Would it really matter?

Regards,
Tobias


From remi at remlab.net  Thu Dec 15 13:41:05 2011
From: remi at remlab.net (=?UTF-8?Q?R=C3=A9mi_Denis-Courmont?=)
Date: Thu, 15 Dec 2011 13:41:05 +0100
Subject: [vlc-devel] [PATCH] VDR recording detection
In-Reply-To: <20111215120649.GA5101@videolan.org>
References: <4EE9DF6F.6060908@web.de> <20111215120649.GA5101@videolan.org>
Message-ID: <295808a31e6ac83c3ab503a2b42360cc@chewa.net>

On Thu, 15 Dec 2011 13:06:49 +0100, Jean-Baptiste Kempf <jb at videolan.org>
wrote:
> On Thu, Dec 15, 2011 at 12:52:15PM +0100, Tobias G?ntner wrote :
>> This patch changes the detection of VDR recordings. I removed the check
>> for index and meta data files because neither is required for playback
>> as far as VLC is concerned.
> 
> It seems to me that the check is now more complex and slower, no?

To me, it looks like it should be a lot faster. At least sscanf() should
be a lot faster than stat().

-- 
R?mi Denis-Courmont
http://www.remlab.net/


From jb at videolan.org  Thu Dec 15 13:48:54 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Thu, 15 Dec 2011 13:48:54 +0100
Subject: [vlc-devel] [PATCH] VDR recording detection
In-Reply-To: <295808a31e6ac83c3ab503a2b42360cc@chewa.net>
References: <4EE9DF6F.6060908@web.de> <20111215120649.GA5101@videolan.org>
	<295808a31e6ac83c3ab503a2b42360cc@chewa.net>
Message-ID: <20111215124854.GA13866@videolan.org>

On Thu, Dec 15, 2011 at 01:41:05PM +0100, R?mi Denis-Courmont wrote :
> On Thu, 15 Dec 2011 13:06:49 +0100, Jean-Baptiste Kempf <jb at videolan.org>
> wrote:
> > On Thu, Dec 15, 2011 at 12:52:15PM +0100, Tobias G?ntner wrote :
> >> This patch changes the detection of VDR recordings. I removed the check
> >> for index and meta data files because neither is required for playback
> >> as far as VLC is concerned.
> > 
> > It seems to me that the check is now more complex and slower, no?
> 
> To me, it looks like it should be a lot faster. At least sscanf() should
> be a lot faster than stat().

OK, but I thought that the check with strcasecmp on ".rec" made it
faster.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From remi at remlab.net  Thu Dec 15 16:08:22 2011
From: remi at remlab.net (=?iso-8859-15?q?R=E9mi?= Denis-Courmont)
Date: Thu, 15 Dec 2011 17:08:22 +0200
Subject: [vlc-devel] [PATCH] Implement vlc_rand_bytes() for OS/2
In-Reply-To: <4EE9AD49.5080308@chollian.net>
References: <1323347568-59805-1-git-send-email-komh@chollian.net>
	<ac613bb1c141982617c8a2f790716e88@chewa.net>
	<4EE9AD49.5080308@chollian.net>
Message-ID: <201112151708.28357.remi@remlab.net>

Le jeudi 15 d?cembre 2011 10:18:17 KO Myung-Hun, vous avez ?crit :
> > It would make sense, I guess.
> 
> How about this ?

Merged, thanks.

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From remi at remlab.net  Thu Dec 15 16:08:46 2011
From: remi at remlab.net (=?iso-8859-1?q?R=E9mi?= Denis-Courmont)
Date: Thu, 15 Dec 2011 17:08:46 +0200
Subject: [vlc-devel] [PATCH 7/7] Fix compilation error due to pthread on
	OS/2
In-Reply-To: <1323344900-59383-1-git-send-email-komh@chollian.net>
References: <1323344900-59383-1-git-send-email-komh@chollian.net>
Message-ID: <201112151708.47345.remi@remlab.net>

Merged, thanks!

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From remi at remlab.net  Thu Dec 15 16:10:04 2011
From: remi at remlab.net (=?iso-8859-15?q?R=E9mi?= Denis-Courmont)
Date: Thu, 15 Dec 2011 17:10:04 +0200
Subject: [vlc-devel] [PATCH] fix filename self-references
In-Reply-To: <CAMYtjArZ-km3cgzbOsV2BNLwYmRdH80A+JXe+Zpc=pBzxmVwQA@mail.gmail.com>
References: <CAMYtjArZ-km3cgzbOsV2BNLwYmRdH80A+JXe+Zpc=pBzxmVwQA@mail.gmail.com>
Message-ID: <201112151710.04906.remi@remlab.net>

Merged, thanks.

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From remi at remlab.net  Thu Dec 15 16:12:10 2011
From: remi at remlab.net (=?iso-8859-15?q?R=E9mi?= Denis-Courmont)
Date: Thu, 15 Dec 2011 17:12:10 +0200
Subject: [vlc-devel] [PATCH] VDR recording detection
In-Reply-To: <4EE9DF6F.6060908@web.de>
References: <4EE9DF6F.6060908@web.de>
Message-ID: <201112151712.11215.remi@remlab.net>

Le jeudi 15 d?cembre 2011 13:52:15 Tobias G?ntner, vous avez ?crit :
> Hi!
> 
> This patch changes the detection of VDR recordings. I removed the check
> for index and meta data files because neither is required for playback
> as far as VLC is concerned. To avoid false positives, I improved the
> directory name check a bit. Also, wrong directory names are easier to
> repair than missing files IMHO.

Maybe it's only me. But I dislike mixing and and or without parenthesis.

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From funman at videolan.org  Thu Dec 15 18:05:36 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Thu, 15 Dec 2011 12:05:36 -0500
Subject: [vlc-devel] [PATCH] contrib: don't ignore triplets in subfolders
In-Reply-To: <1323940169-77754-1-git-send-email-enrique.osuna@gmail.com>
References: <1323940169-77754-1-git-send-email-enrique.osuna@gmail.com>
Message-ID: <20111215120536.6b10d028@tok>

Le Thu, 15 Dec 2011 10:09:29 +0100,
"Faustino E. Osuna" <enrique.osuna at gmail.com> a ?crit :

> From: "Faustino E. Osuna" <enrique.osuna at gmail.com>
> 
> Only ignores triplets in the root contrib path. Previous
> implementation had the potential to ignore triplets in the
> subfolders.
> 
> This will allow for creating:
> 1) Patches using this format (e.g. contrib/src/*/track-me-now.patch).
> 2) Subfolders beneath contrib's root (but not in contrib itself)
> (e.g. contrib/src/track-me-now/*).
> ---
>  contrib/.gitignore |    2 +-
>  1 files changed, 1 insertions(+), 1 deletions(-)
> 
> diff --git a/contrib/.gitignore b/contrib/.gitignore
> index 64cbe1d..e51bc0f 100644
> --- a/contrib/.gitignore
> +++ b/contrib/.gitignore
> @@ -1,2 +1,2 @@
>  vlc-contrib-*.tar.xz
> -*-*-*
> +/*-*-*/

lgtm

-- 
Rafa?l Carr?


From funman at videolan.org  Thu Dec 15 18:06:14 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Thu, 15 Dec 2011 12:06:14 -0500
Subject: [vlc-devel] [RFC][PATCH 1/2] contrib: check os x gcc v4.2
 before overriding
In-Reply-To: <CAKhCBszYkVoi378owd8F8A+j9n-qgPvGv4rumB_QU5VrPL+kpA@mail.gmail.com>
References: <1323894926-21050-1-git-send-email-enrique.osuna@gmail.com>
	<1323894926-21050-2-git-send-email-enrique.osuna@gmail.com>
	<20111214160410.1d799a35@tok> <20111214211031.GA22396@videolan.org>
	<CAKhCBszYkVoi378owd8F8A+j9n-qgPvGv4rumB_QU5VrPL+kpA@mail.gmail.com>
Message-ID: <20111215120614.71b2d93b@tok>

Le Thu, 15 Dec 2011 10:12:05 +0100,
Faustino Osuna <enrique.osuna at gmail.com> a ?crit :

> I was able to try this out on 10.6 last night, it only worked for
> x86_64-apple-darwin10, but not i386-apple-darwin10/11.
> 
> This patch set can be discarded.

For curiosity, what was the problem with i386 ?

-- 
Rafa?l Carr?


From chengsun9 at gmail.com  Thu Dec 15 18:14:42 2011
From: chengsun9 at gmail.com (Cheng Sun)
Date: Thu, 15 Dec 2011 17:14:42 +0000
Subject: [vlc-devel] [PATCH] Fix stack overflow in ExecuteCommand
Message-ID: <4EEA2B02.1030201@gmail.com>

Fix ticket #5675

Because ExecuteCommand allocates temporary string space on the stack,
proportional to the length of the command to execute, a stack overflow can
occur when the the command is too long.

This can be triggered remotely e.g. from the VLC Web interface, by
running this
JavaScript:

sendVLMCmd(Array.prototype.join.call({length:300000},'a'));

which sends a string of length 300000 to ExecuteCommand, crashing VLC.

Patch attached.


- Cheng Sun
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-Fix-stack-overflow-in-ExecuteCommand.patch
Type: text/x-patch
Size: 2268 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111215/4955b86f/attachment.bin>

From remi at remlab.net  Thu Dec 15 18:38:26 2011
From: remi at remlab.net (=?iso-8859-15?q?R=E9mi?= Denis-Courmont)
Date: Thu, 15 Dec 2011 19:38:26 +0200
Subject: [vlc-devel] [PATCH] Fix stack overflow in ExecuteCommand
In-Reply-To: <4EEA2B02.1030201@gmail.com>
References: <4EEA2B02.1030201@gmail.com>
Message-ID: <201112151938.26981.remi@remlab.net>

Le jeudi 15 d?cembre 2011 19:14:42 Cheng Sun, vous avez ?crit :
> Fix ticket #5675
> 
> Because ExecuteCommand allocates temporary string space on the stack,
> proportional to the length of the command to execute, a stack overflow can
> occur when the the command is too long.

With this patch, VLC will do undefined behaviour as in dereferencing NULL 
instead of undefined behaviour as in overflowing the stack.

It could be better...

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From cheng.sun at ymail.com  Thu Dec 15 18:49:13 2011
From: cheng.sun at ymail.com (Cheng Sun)
Date: Thu, 15 Dec 2011 17:49:13 +0000
Subject: [vlc-devel] [PATCH] Fix stack overflow in ExecuteCommand
In-Reply-To: <4EEA2B02.1030201@gmail.com>
References: <4EEA2B02.1030201@gmail.com>
Message-ID: <4EEA3319.70307@ymail.com>

Previous patch forgot to modify an assert.

- Cheng

On 15/12/11 17:14, Cheng Sun wrote:
> Fix ticket #5675
>
> Because ExecuteCommand allocates temporary string space on the stack,
> proportional to the length of the command to execute, a stack overflow can
> occur when the the command is too long.
>
> This can be triggered remotely e.g. from the VLC Web interface, by
> running this
> JavaScript:
>
> sendVLMCmd(Array.prototype.join.call({length:300000},'a'));
>
> which sends a string of length 300000 to ExecuteCommand, crashing VLC.
>
> Patch attached.
>
>
> - Cheng Sun
> _______________________________________________
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
> http://mailman.videolan.org/listinfo/vlc-devel

-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-Fix-stack-overflow-in-ExecuteCommand.patch
Type: text/x-patch
Size: 2252 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111215/de898cb1/attachment.bin>

From cheng.sun at ymail.com  Thu Dec 15 18:57:16 2011
From: cheng.sun at ymail.com (Cheng Sun)
Date: Thu, 15 Dec 2011 17:57:16 +0000
Subject: [vlc-devel] [PATCH] Fix stack overflow in ExecuteCommand
In-Reply-To: <201112151938.26981.remi@remlab.net>
References: <4EEA2B02.1030201@gmail.com> <201112151938.26981.remi@remlab.net>
Message-ID: <4EEA34FC.2080103@ymail.com>

On 15/12/11 17:38, R?mi Denis-Courmont wrote:
> Le jeudi 15 d?cembre 2011 19:14:42 Cheng Sun, vous avez ?crit :
>> Fix ticket #5675
>>
>> Because ExecuteCommand allocates temporary string space on the stack,
>> proportional to the length of the command to execute, a stack overflow can
>> occur when the the command is too long.
> With this patch, VLC will do undefined behaviour as in dereferencing NULL 
> instead of undefined behaviour as in overflowing the stack.
>
> It could be better...
>
Revised patch which checks for malloc returning NULL

- Cheng
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-Fix-stack-overflow-in-ExecuteCommand.patch
Type: text/x-patch
Size: 2449 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111215/dbb7c300/attachment.bin>

From cheng.sun at ymail.com  Thu Dec 15 19:05:39 2011
From: cheng.sun at ymail.com (Cheng Sun)
Date: Thu, 15 Dec 2011 18:05:39 +0000
Subject: [vlc-devel] [PATCH] Fix stack overflow in ExecuteCommand
In-Reply-To: <4EEA34FC.2080103@ymail.com>
References: <4EEA2B02.1030201@gmail.com> <201112151938.26981.remi@remlab.net>
	<4EEA34FC.2080103@ymail.com>
Message-ID: <4EEA36F3.2020100@ymail.com>

On 15/12/11 17:57, Cheng Sun wrote:
> On 15/12/11 17:38, R?mi Denis-Courmont wrote:
>> With this patch, VLC will do undefined behaviour as in dereferencing NULL 
>> instead of undefined behaviour as in overflowing the stack.
>>
>> It could be better...
>>
> Revised patch which checks for malloc returning NULL
>
> - Cheng
> _______________________________________________
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
> http://mailman.videolan.org/listinfo/vlc-devel
Revised patch, fixing coding style

- Cheng
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-Fix-stack-overflow-in-ExecuteCommand.patch
Type: text/x-patch
Size: 2472 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111215/61d3f1aa/attachment.bin>

From cheng.sun at ymail.com  Thu Dec 15 19:17:52 2011
From: cheng.sun at ymail.com (Cheng Sun)
Date: Thu, 15 Dec 2011 18:17:52 +0000
Subject: [vlc-devel] [PATCH] Fix stack overflow in ExecuteCommand
In-Reply-To: <4EEA36F3.2020100@ymail.com>
References: <4EEA2B02.1030201@gmail.com> <201112151938.26981.remi@remlab.net>
	<4EEA34FC.2080103@ymail.com> <4EEA36F3.2020100@ymail.com>
Message-ID: <4EEA39D0.7090106@ymail.com>

On 15/12/11 18:05, Cheng Sun wrote:
> On 15/12/11 17:57, Cheng Sun wrote:
>> On 15/12/11 17:38, R?mi Denis-Courmont wrote:
>>> With this patch, VLC will do undefined behaviour as in dereferencing NULL 
>>> instead of undefined behaviour as in overflowing the stack.
>>>
>>> It could be better...
>>>
>> Revised patch which checks for malloc returning NULL
>>
>> - Cheng
>> _______________________________________________
>> vlc-devel mailing list
>> To unsubscribe or modify your subscription options:
>> http://mailman.videolan.org/listinfo/vlc-devel
> Revised patch, fixing coding style
>
> - Cheng
> _______________________________________________
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
> http://mailman.videolan.org/listinfo/vlc-devel
Revised patch, which corrects a size calculation

- Cheng
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-Fix-stack-overflow-in-ExecuteCommand.patch
Type: text/x-patch
Size: 2472 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111215/c9896a13/attachment.bin>

From enrique.osuna at gmail.com  Thu Dec 15 21:45:04 2011
From: enrique.osuna at gmail.com (Faustino Osuna)
Date: Thu, 15 Dec 2011 21:45:04 +0100
Subject: [vlc-devel] [RFC][PATCH 1/2] contrib: check os x gcc v4.2
	before overriding
In-Reply-To: <20111215120614.71b2d93b@tok>
References: <1323894926-21050-1-git-send-email-enrique.osuna@gmail.com>
	<1323894926-21050-2-git-send-email-enrique.osuna@gmail.com>
	<20111214160410.1d799a35@tok> <20111214211031.GA22396@videolan.org>
	<CAKhCBszYkVoi378owd8F8A+j9n-qgPvGv4rumB_QU5VrPL+kpA@mail.gmail.com>
	<20111215120614.71b2d93b@tok>
Message-ID: <CAKhCBsxPQssH2Ti569TkqFxB_D2nNr1vpcDXUpJkA6mwFC8CHg@mail.gmail.com>

libcaca refused to compile...can't remember exactly what the error was but
it had something to do with inline assembly.

Enrique

On Thu, Dec 15, 2011 at 6:06 PM, Rafa?l Carr? <funman at videolan.org> wrote:

> Le Thu, 15 Dec 2011 10:12:05 +0100,
> Faustino Osuna <enrique.osuna at gmail.com> a ?crit :
>
> > I was able to try this out on 10.6 last night, it only worked for
> > x86_64-apple-darwin10, but not i386-apple-darwin10/11.
> >
> > This patch set can be discarded.
>
> For curiosity, what was the problem with i386 ?
>
> --
> Rafa?l Carr?
> _______________________________________________
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
> http://mailman.videolan.org/listinfo/vlc-devel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111215/7e7e4cc2/attachment.html>

From fatbull at web.de  Thu Dec 15 23:28:25 2011
From: fatbull at web.de (=?ISO-8859-15?Q?Tobias_G=FCntner?=)
Date: Thu, 15 Dec 2011 23:28:25 +0100
Subject: [vlc-devel] [PATCH] VDR recording detection
In-Reply-To: <201112151712.11215.remi@remlab.net>
References: <4EE9DF6F.6060908@web.de> <201112151712.11215.remi@remlab.net>
Message-ID: <4EEA7489.8060706@web.de>

Am 15.12.2011 16:12, schrieb R?mi Denis-Courmont:
> Maybe it's only me. But I dislike mixing and and or without parenthesis.

Well, I dislike redundant parentheses.

Whatever.

See attachment.

Regards,
Tobias
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-Ignore-missing-meta-data-files.-Check-directory-name.patch
Type: text/x-patch
Size: 6608 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111215/708937ae/attachment.bin>

From funman at videolan.org  Thu Dec 15 23:46:07 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Thu, 15 Dec 2011 17:46:07 -0500
Subject: [vlc-devel] [RFC][PATCH 1/2] contrib: check os x gcc v4.2
 before overriding
In-Reply-To: <CAKhCBsxPQssH2Ti569TkqFxB_D2nNr1vpcDXUpJkA6mwFC8CHg@mail.gmail.com>
References: <1323894926-21050-1-git-send-email-enrique.osuna@gmail.com>
	<1323894926-21050-2-git-send-email-enrique.osuna@gmail.com>
	<20111214160410.1d799a35@tok> <20111214211031.GA22396@videolan.org>
	<CAKhCBszYkVoi378owd8F8A+j9n-qgPvGv4rumB_QU5VrPL+kpA@mail.gmail.com>
	<20111215120614.71b2d93b@tok>
	<CAKhCBsxPQssH2Ti569TkqFxB_D2nNr1vpcDXUpJkA6mwFC8CHg@mail.gmail.com>
Message-ID: <20111215174607.1c1555be@tok>

Le Thu, 15 Dec 2011 21:45:04 +0100,
Faustino Osuna <enrique.osuna at gmail.com> a ?crit :

> libcaca refused to compile...can't remember exactly what the error was but
> it had something to do with inline assembly.

right, i remember about this.

it might be possible to rewrite the assembly but if no one is willing
to do that i think it's better to stick to gcc 4.2

-- 
Rafa?l Carr?


From enrique.osuna at gmail.com  Thu Dec 15 23:50:01 2011
From: enrique.osuna at gmail.com (Faustino Osuna)
Date: Thu, 15 Dec 2011 23:50:01 +0100
Subject: [vlc-devel] [RFC][PATCH 1/2] contrib: check os x gcc v4.2
	before overriding
In-Reply-To: <20111215174607.1c1555be@tok>
References: <1323894926-21050-1-git-send-email-enrique.osuna@gmail.com>
	<1323894926-21050-2-git-send-email-enrique.osuna@gmail.com>
	<20111214160410.1d799a35@tok> <20111214211031.GA22396@videolan.org>
	<CAKhCBszYkVoi378owd8F8A+j9n-qgPvGv4rumB_QU5VrPL+kpA@mail.gmail.com>
	<20111215120614.71b2d93b@tok>
	<CAKhCBsxPQssH2Ti569TkqFxB_D2nNr1vpcDXUpJkA6mwFC8CHg@mail.gmail.com>
	<20111215174607.1c1555be@tok>
Message-ID: <CAKhCBsz0yVJ8xfu3SGDtk9ggBf3H_XOmScvhnJ-Oe588m_YuSg@mail.gmail.com>

In the mean time, should I update the Wiki to reflect that LLVM GCC 4.2 is
not a viable option to compile the contribs?

On Thu, Dec 15, 2011 at 11:46 PM, Rafa?l Carr? <funman at videolan.org> wrote:

> Le Thu, 15 Dec 2011 21:45:04 +0100,
> Faustino Osuna <enrique.osuna at gmail.com> a ?crit :
>
> > libcaca refused to compile...can't remember exactly what the error was
> but
> > it had something to do with inline assembly.
>
> right, i remember about this.
>
> it might be possible to rewrite the assembly but if no one is willing
> to do that i think it's better to stick to gcc 4.2
>
> --
> Rafa?l Carr?
> _______________________________________________
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
> http://mailman.videolan.org/listinfo/vlc-devel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111215/e5b337f0/attachment.html>

From jb at videolan.org  Fri Dec 16 00:03:38 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Fri, 16 Dec 2011 00:03:38 +0100
Subject: [vlc-devel] [PATCH] contrib: don't ignore triplets in subfolders
In-Reply-To: <20111215120536.6b10d028@tok>
References: <1323940169-77754-1-git-send-email-enrique.osuna@gmail.com>
	<20111215120536.6b10d028@tok>
Message-ID: <20111215230338.GA4331@videolan.org>

On Thu, Dec 15, 2011 at 12:05:36PM -0500, Rafa?l Carr? wrote :
> lgtm

Applied.

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Fri Dec 16 00:45:59 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Fri, 16 Dec 2011 00:45:59 +0100
Subject: [vlc-devel] [PATCHES] update live555 requirements
Message-ID: <1323992761-22393-1-git-send-email-jb@videolan.org>

Small set of patches to update live555 in contribs and update
configure.ac requirements, since upstream seems to have fixed the
issues that were raised.

[PATCH 1/2] Contribs: update live555 to latest snapshot
[PATCH 2/2] Configure: Update live555 requirements


From jb at videolan.org  Fri Dec 16 00:46:00 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Fri, 16 Dec 2011 00:46:00 +0100
Subject: [vlc-devel] [PATCH 1/2] Contribs: update live555 to latest snapshot
In-Reply-To: <1323992761-22393-1-git-send-email-jb@videolan.org>
References: <1323992761-22393-1-git-send-email-jb@videolan.org>
Message-ID: <1323992761-22393-2-git-send-email-jb@videolan.org>

---
 contrib/src/live555/SHA512SUMS |    2 +-
 contrib/src/live555/rules.mak  |   11 +----------
 2 files changed, 2 insertions(+), 11 deletions(-)

diff --git a/contrib/src/live555/SHA512SUMS b/contrib/src/live555/SHA512SUMS
index da74cf1..9e416a4 100644
--- a/contrib/src/live555/SHA512SUMS
+++ b/contrib/src/live555/SHA512SUMS
@@ -1 +1 @@
-42e424e8743cdbce2d6c87b6ceeb182a8868e870a0b233df987004a99e97e80c216e6c996774e936be40b1c3397a7c95ac8aa74eb6dcd8fac42a81dfcb4fd767  live.2011.11.08.tar.gz
+00578686c1e154cea7638d19a6ab7675fa79d7da8f1848e9765954e3b02b82c54fd05e4480e7ffcdc399c7ad3a0ad26aff0a8ce97222e26e66e345fe82906e02  live.2011.12.02.tar.gz
diff --git a/contrib/src/live555/rules.mak b/contrib/src/live555/rules.mak
index a267981..f5bb7c1 100644
--- a/contrib/src/live555/rules.mak
+++ b/contrib/src/live555/rules.mak
@@ -1,7 +1,7 @@
 # live555
 
 #LIVEDOTCOM_URL := http://live555.com/liveMedia/public/live555-latest.tar.gz
-LIVE555_FILE := live.2011.11.08.tar.gz
+LIVE555_FILE := live.2011.12.02.tar.gz
 LIVEDOTCOM_URL := http://live555sourcecontrol.googlecode.com/files/$(LIVE555_FILE)
 
 PKGS += live555
@@ -15,15 +15,6 @@ live555: $(LIVE555_FILE) .sum-live555
 	rm -Rf live
 	$(UNPACK)
 	chmod -R u+w live
-	patch -p0 < $(SRC)/live555/live-uselocale.patch
-	patch -p0 < $(SRC)/live555/live-inet_ntop.patch
-	patch -p0 < $(SRC)/live555/live-intptr.patch
-ifndef HAVE_WIN32
-ifndef HAVE_WINCE
-	patch -p0 < $(SRC)/live555/live-getaddrinfo.patch
-endif
-endif
-	patch -p0 < $(SRC)/live555/live-cloexec.patch
 	mv live $@
 	touch $@
 
-- 
1.7.8.rc3



From jb at videolan.org  Fri Dec 16 00:46:01 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Fri, 16 Dec 2011 00:46:01 +0100
Subject: [vlc-devel] [PATCH 2/2] Configure: Update live555 requirements
In-Reply-To: <1323992761-22393-1-git-send-email-jb@videolan.org>
References: <1323992761-22393-1-git-send-email-jb@videolan.org>
Message-ID: <1323992761-22393-3-git-send-email-jb@videolan.org>

---
 configure.ac |   32 +++-----------------------------
 1 files changed, 3 insertions(+), 29 deletions(-)

diff --git a/configure.ac b/configure.ac
index 4f37ad2..40a3d73 100644
--- a/configure.ac
+++ b/configure.ac
@@ -1655,10 +1655,10 @@ AS_IF([test "${enable_live555}" != "no" -a -n "${CXX}"], [
   ])
   CPPFLAGS="${CPPFLAGS} ${CPPFLAGS_live555}"
 
-  AC_CACHE_CHECK([for live555 version 1275091200 or later], [ac_cv_live555], [
+  AC_CACHE_CHECK([for live555 version 1322784000 or later], [ac_cv_live555], [
     AC_PREPROC_IFELSE([AC_LANG_PROGRAM([
 [#include <liveMedia_version.hh>
-#if !defined (LIVEMEDIA_LIBRARY_VERSION_INT) || (LIVEMEDIA_LIBRARY_VERSION_INT < 1275091200)
+#if !defined (LIVEMEDIA_LIBRARY_VERSION_INT) || (LIVEMEDIA_LIBRARY_VERSION_INT < 1322784000)
 # error BOOM
 #endif]])
     ], [
@@ -1669,39 +1669,13 @@ AS_IF([test "${enable_live555}" != "no" -a -n "${CXX}"], [
   ])
   AS_IF([test "$ac_cv_live555" = "no"], [
     AC_MSG_WARN([liveMedia is missing or its installed version is too old:
-Version 2010.05.29 or later is required to proceed.
+Version 2011.12.02 or later is required to proceed.
 You can get an updated one from http://www.live555.com/liveMedia .])
     AS_IF([test -n "${enable_live555}"], [
       AC_MSG_ERROR([Update live555 or pass --disable-live555 to disable RTSP input support.])
     ])
     ac_cv_live555_usable=""
   ], [
-    dnl Check that live555 is patched correctly.
-    AC_CACHE_CHECK([for live555 usability], [ac_cv_live555_usable], [
-      AC_COMPILE_IFELSE([AC_LANG_PROGRAM([
-[#include <locale.h>
-#ifdef __APPLE__
-# include <xlocale.h>
-#endif
-#undef LC_ALL
-#define LC_ALL BOOM
-#ifdef LC_ALL_MASK
-#include <Locale.hh>
-#endif
-#include <GroupsockHelper.hh>]], [
-[struct in_addr addr;
-our_inet_ntoa(addr, NULL);]])
-      ], [
-        ac_cv_live555_usable="yes"
-      ], [
-        ac_cv_live555_usable="no"
-      ])
-    ])
-    AS_IF([test "$ac_cv_live555_usable" = "no"], [
-      AC_MSG_ERROR([liveMedia lacks patches and is not usable.
-Please apply our patches from the VLC contrib (contrib/src/live555/).
-You can also disable RTSP input with --disable-live555. You were warned.])
-    ])
     other_libs="-lgroupsock -lBasicUsageEnvironment -lUsageEnvironment"
     other_libs_pic="-lgroupsock_pic -lBasicUsageEnvironment_pic -lUsageEnvironment_pic"
     AS_IF([test "${SYS}" = "mingw32"], [
-- 
1.7.8.rc3



From funman at videolan.org  Fri Dec 16 01:17:50 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Thu, 15 Dec 2011 19:17:50 -0500
Subject: [vlc-devel] [PATCH 1/2] Contribs: update live555 to latest
 snapshot
In-Reply-To: <1323992761-22393-2-git-send-email-jb@videolan.org>
References: <1323992761-22393-1-git-send-email-jb@videolan.org>
	<1323992761-22393-2-git-send-email-jb@videolan.org>
Message-ID: <20111215191750.7449df23@tok>

Le Fri, 16 Dec 2011 00:46:00 +0100,
Jean-Baptiste Kempf <jb at videolan.org> a ?crit :

> ---
>  contrib/src/live555/SHA512SUMS |    2 +-
>  contrib/src/live555/rules.mak  |   11 +----------
>  2 files changed, 2 insertions(+), 11 deletions(-)
> 
> diff --git a/contrib/src/live555/SHA512SUMS b/contrib/src/live555/SHA512SUMS
> index da74cf1..9e416a4 100644
> --- a/contrib/src/live555/SHA512SUMS
> +++ b/contrib/src/live555/SHA512SUMS
> @@ -1 +1 @@
> -42e424e8743cdbce2d6c87b6ceeb182a8868e870a0b233df987004a99e97e80c216e6c996774e936be40b1c3397a7c95ac8aa74eb6dcd8fac42a81dfcb4fd767  live.2011.11.08.tar.gz
> +00578686c1e154cea7638d19a6ab7675fa79d7da8f1848e9765954e3b02b82c54fd05e4480e7ffcdc399c7ad3a0ad26aff0a8ce97222e26e66e345fe82906e02  live.2011.12.02.tar.gz
> diff --git a/contrib/src/live555/rules.mak b/contrib/src/live555/rules.mak
> index a267981..f5bb7c1 100644
> --- a/contrib/src/live555/rules.mak
> +++ b/contrib/src/live555/rules.mak
> @@ -1,7 +1,7 @@
>  # live555
>  
>  #LIVEDOTCOM_URL := http://live555.com/liveMedia/public/live555-latest.tar.gz
> -LIVE555_FILE := live.2011.11.08.tar.gz
> +LIVE555_FILE := live.2011.12.02.tar.gz
>  LIVEDOTCOM_URL := http://live555sourcecontrol.googlecode.com/files/$(LIVE555_FILE)
>  
>  PKGS += live555
> @@ -15,15 +15,6 @@ live555: $(LIVE555_FILE) .sum-live555
>  	rm -Rf live
>  	$(UNPACK)
>  	chmod -R u+w live
> -	patch -p0 < $(SRC)/live555/live-uselocale.patch
> -	patch -p0 < $(SRC)/live555/live-inet_ntop.patch
> -	patch -p0 < $(SRC)/live555/live-intptr.patch
> -ifndef HAVE_WIN32
> -ifndef HAVE_WINCE
> -	patch -p0 < $(SRC)/live555/live-getaddrinfo.patch
> -endif
> -endif
> -	patch -p0 < $(SRC)/live555/live-cloexec.patch

don't forget to git rm those patches when you'll commit

>  	mv live $@
>  	touch $@
>  



-- 
Rafa?l Carr?


From jb at videolan.org  Fri Dec 16 01:30:15 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Fri, 16 Dec 2011 01:30:15 +0100
Subject: [vlc-devel] [PATCH 1/2] Contribs: update live555 to latest
 snapshot
In-Reply-To: <20111215191750.7449df23@tok>
References: <1323992761-22393-1-git-send-email-jb@videolan.org>
	<1323992761-22393-2-git-send-email-jb@videolan.org>
	<20111215191750.7449df23@tok>
Message-ID: <20111216003014.GA20509@videolan.org>

On Thu, Dec 15, 2011 at 07:17:50PM -0500, Rafa?l Carr? wrote :
> don't forget to git rm those patches when you'll commit

Good point.


-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From var1able at var1able.ru  Fri Dec 16 15:00:45 2011
From: var1able at var1able.ru (=?utf-8?B?0JrQvtC90YHRgtCw0L3RgtC40L0g0JHQvtCz0LTQsNC90L7Qsg==?=)
Date: Fri, 16 Dec 2011 18:00:45 +0400
Subject: [vlc-devel] Mobile web interface patch
Message-ID: <3366F855-5D14-460F-A81E-B2094FEA5C9F@var1able.ru>

http://www.google-melange.com/gci/task/view/google/gci2011/7180436
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111216/0ecac400/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: mobile_interface.patch
Type: application/octet-stream
Size: 12296 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111216/0ecac400/attachment.obj>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111216/0ecac400/attachment-0001.html>

From beauze.h at gmail.com  Fri Dec 16 16:40:55 2011
From: beauze.h at gmail.com (=?UTF-8?Q?Hugo_Beauz=C3=A9e=2DLuyssen?=)
Date: Fri, 16 Dec 2011 16:40:55 +0100
Subject: [vlc-devel] [Patch] Adding an helper to convert iso8601
 durations to seconds
In-Reply-To: <CAK-C75RayZc-hHcd9-BE-cF5gXFF_JjNB6hDGCz5wxS4OfSfAg@mail.gmail.com>
References: <CAK-C75R5BPUud6z2QxrFNNnh1wMFktu48O59oXi2bAUqZ0mYMw@mail.gmail.com>
	<CAK-C75RtmcQd0m32+3d9ue7WZKH8v9n84izFWRp=ziwKm4yCEw@mail.gmail.com>
	<CAK-C75QD08hyZ4XNBR7nYkSFyLqR7Tm887Uxw7fWqpXRGJd=Dg@mail.gmail.com>
	<201112141905.47669.remi@remlab.net>
	<CAK-C75RayZc-hHcd9-BE-cF5gXFF_JjNB6hDGCz5wxS4OfSfAg@mail.gmail.com>
Message-ID: <CAK-C75R-sJ7AOJ9AGOh680VM2YAHdbnUTU2nFbJPSKhVpiPnFg@mail.gmail.com>

On Wed, Dec 14, 2011 at 6:33 PM, Hugo Beauz?e-Luyssen
<beauze.h at gmail.com> wrote:
> 2011/12/14 R?mi Denis-Courmont <remi at remlab.net>:
>> Le mercredi 14 d?cembre 2011 00:48:41 Hugo Beauz?e-Luyssen, vous avez ?crit :
>>> On Tue, Dec 13, 2011 at 11:46 PM, Hugo Beauz?e-Luyssen
>>>
>>> <beauze.h at gmail.com> wrote:
>>> > 2011/12/9 R?mi Denis-Courmont <remi at remlab.net>:
>>> >> Function name is too long. And time_t is thecorrect type for seconds
>>> >> IMHO.
>>>
>>> Oops, this one's the good one.
>>
>> I think this patch does not work due to typo. Please drop _to_s*. It's
>> obviously seconds since the return type is time_t.
>>
>
> Hmmm yep you're right... my bad.
> Here are some fixed versions of both patches.
>
> Regards,
>

Hello,

Are these patches ok?

Regards,

-- 
Hugo Beauz?e-Luyssen


From beauze.h at gmail.com  Fri Dec 16 16:42:46 2011
From: beauze.h at gmail.com (=?UTF-8?Q?Hugo_Beauz=C3=A9e=2DLuyssen?=)
Date: Fri, 16 Dec 2011 16:42:46 +0100
Subject: [vlc-devel] [Patch] Dash: Don't try to do something without a
 valid MPD manager
In-Reply-To: <20111214180641.GA23803@videolan.org>
References: <CAK-C75TOrnUGWki6HDVk0iJ0_xE1ZO4Swd8FGAoCraupTXQW8Q@mail.gmail.com>
	<20111214180641.GA23803@videolan.org>
Message-ID: <CAK-C75TtOYuwx2d=WQX4XUvqY6bjON-P=X2zbPFnig331usRXA@mail.gmail.com>

On Wed, Dec 14, 2011 at 7:06 PM, Jean-Baptiste Kempf <jb at videolan.org> wrote:
> Why do we need a NullManager, in the end, since we have the empty
> interface already?
> What is the use of instantiating it, if it does not do anything?
> If the profile is not implemented, we just quit the module, no?
>
> I must be missing something, but I don't see the point.
>

As far as I'm concerned, there's no much use, but it forces us to
maintain the NullManager implementation, in case we edit the
interface, which is a waste of time.

Regards,

-- 
Hugo Beauz?e-Luyssen


From martin at martin.st  Fri Dec 16 22:47:01 2011
From: martin at martin.st (=?UTF-8?q?Martin=20Storsj=C3=B6?=)
Date: Fri, 16 Dec 2011 23:47:01 +0200
Subject: [vlc-devel] [PATCH 0/5] android/iomx as a separate shared library
Message-ID: <1324072028-6485-1-git-send-email-martin@martin.st>

Hi,

This patchset fixes the iomx code for android to be built
as a small separate shared library, that can be loaded by
the android application. This approach has four benefits:

- Avoids having to link libvlcjni.so directly against private
  functions, which could lead to it failing to load on some
  devices
- Avoids having to use private headers during the build of the
  main VLC code (only needed when building the IOMX wrapper)
- Eases supporting different ABI versions. Since the IOMX API
  is a private API, the ABI has changed between gingerbread
  and honeycomb (even though the gingerbread version works
  on most froyo devices, too). With this variant, one apk
  can include a separate version of the IOMX wrapper code
  shared library for each targeted ABI version - this is
  only about 20 KB per library.
- The separate IOMX wrapper library has exactly the same external
  API/ABI as any OMX Core, and one could load it just like the
  other ones if one wanted. (dlopen on android doesn't look in
  the app's private library directory though, so it would require
  a bit of extra work to find the path. Therefore it's loaded
  by the android java code instead, which also can figure out
  which version to load.)

Also, I kinda like the amount of removed code in this
patchset. :-)

Martin Storsj? (5):
  omxil: Make the iomx wrapper not depend on the omxil utils.c file
  omxil: Build the iomx wrapper code as a separate shared library
  omxil: Remove the header check for the iomx module
  android: Don't add private header include directories when building
    the full VLC tree
  omxil: Use a fake library name when loading IOMX

 configure.ac                        |   13 +-----
 extras/package/android/configure.sh |    5 --
 modules/codec/omxil/Modules.am      |    2 +-
 modules/codec/omxil/iomx.cpp        |   83 +++++++++++++++--------------------
 modules/codec/omxil/iomx.h          |   33 --------------
 modules/codec/omxil/omxil.c         |   12 +++--
 6 files changed, 45 insertions(+), 103 deletions(-)
 delete mode 100644 modules/codec/omxil/iomx.h

-- 
1.7.2.5



From martin at martin.st  Fri Dec 16 22:47:02 2011
From: martin at martin.st (=?UTF-8?q?Martin=20Storsj=C3=B6?=)
Date: Fri, 16 Dec 2011 23:47:02 +0200
Subject: [vlc-devel] [PATCH 1/5] omxil: Make the iomx wrapper not depend on
	the omxil utils.c file
In-Reply-To: <1324072028-6485-1-git-send-email-martin@martin.st>
References: <1324072028-6485-1-git-send-email-martin@martin.st>
Message-ID: <1324072028-6485-2-git-send-email-martin@martin.st>

---
 modules/codec/omxil/iomx.cpp |   34 +++++++++++++++++++++++++---------
 1 files changed, 25 insertions(+), 9 deletions(-)

diff --git a/modules/codec/omxil/iomx.cpp b/modules/codec/omxil/iomx.cpp
index d0cfe59..a207d9c 100644
--- a/modules/codec/omxil/iomx.cpp
+++ b/modules/codec/omxil/iomx.cpp
@@ -23,11 +23,6 @@
 /*****************************************************************************
  * Preamble
  *****************************************************************************/
-#ifdef HAVE_CONFIG_H
-# include "config.h"
-#endif
-
-#include <vlc_common.h>
 
 #include <media/stagefright/OMXClient.h>
 #include <media/IOMX.h>
@@ -35,9 +30,6 @@
 #include <OMX_Component.h>
 
 #include "iomx.h"
-extern "C" {
-#include "omxil_utils.h"
-}
 
 using namespace android;
 
@@ -147,8 +139,32 @@ static int get_param_size(OMX_INDEXTYPE param_index)
         return sizeof(OMX_PORT_PARAM_TYPE);
     case OMX_IndexParamNumAvailableStreams:
         return sizeof(OMX_PARAM_U32TYPE);
+    case OMX_IndexParamAudioPcm:
+        return sizeof(OMX_AUDIO_PARAM_PCMMODETYPE);
+    case OMX_IndexParamAudioAdpcm:
+        return sizeof(OMX_AUDIO_PARAM_AMRTYPE);
+    case OMX_IndexParamAudioAmr:
+        return sizeof(OMX_AUDIO_PARAM_AMRTYPE);
+    case OMX_IndexParamAudioG723:
+        return sizeof(OMX_AUDIO_PARAM_G723TYPE);
+    case OMX_IndexParamAudioG726:
+        return sizeof(OMX_AUDIO_PARAM_G726TYPE);
+    case OMX_IndexParamAudioG729:
+        return sizeof(OMX_AUDIO_PARAM_G729TYPE);
+    case OMX_IndexParamAudioAac:
+        return sizeof(OMX_AUDIO_PARAM_AACPROFILETYPE);
+    case OMX_IndexParamAudioMp3:
+        return sizeof(OMX_AUDIO_PARAM_MP3TYPE);
+    case OMX_IndexParamAudioSbc:
+        return sizeof(OMX_AUDIO_PARAM_SBCTYPE);
+    case OMX_IndexParamAudioVorbis:
+        return sizeof(OMX_AUDIO_PARAM_VORBISTYPE);
+    case OMX_IndexParamAudioWma:
+        return sizeof(OMX_AUDIO_PARAM_WMATYPE);
+    case OMX_IndexParamAudioRa:
+        return sizeof(OMX_AUDIO_PARAM_RATYPE);
     default:
-        return GetAudioParamSize(param_index);
+        return 0;
     }
 }
 
-- 
1.7.2.5



From martin at martin.st  Fri Dec 16 22:47:03 2011
From: martin at martin.st (=?UTF-8?q?Martin=20Storsj=C3=B6?=)
Date: Fri, 16 Dec 2011 23:47:03 +0200
Subject: [vlc-devel] [PATCH 2/5] omxil: Build the iomx wrapper code as a
	separate shared library
In-Reply-To: <1324072028-6485-1-git-send-email-martin@martin.st>
References: <1324072028-6485-1-git-send-email-martin@martin.st>
Message-ID: <1324072028-6485-3-git-send-email-martin@martin.st>

The separate shared library needs to be loaded into the process
before using the iomx module, preferrably by the java part
of the android application.

This allows building the iomx glue code for a number of different
ABI versions, and additionally avoids linking the main VLC
library to any private symbols that might make it unloadable.
---
 modules/codec/omxil/Modules.am |    2 +-
 modules/codec/omxil/iomx.cpp   |   49 ++++++++-------------------------------
 modules/codec/omxil/iomx.h     |   33 ---------------------------
 modules/codec/omxil/omxil.c    |    8 +++---
 4 files changed, 15 insertions(+), 77 deletions(-)
 delete mode 100644 modules/codec/omxil/iomx.h

diff --git a/modules/codec/omxil/Modules.am b/modules/codec/omxil/Modules.am
index 8e1695c..a33b2e7 100644
--- a/modules/codec/omxil/Modules.am
+++ b/modules/codec/omxil/Modules.am
@@ -3,4 +3,4 @@ SOURCES_omxil = omxil.c utils.c omxil.h omxil_utils.h \
         OMX_Audio.h OMX_Index.h OMX_Other.h OMX_Video.h ../h264_nal.h
 
 CPPFLAGS_iomx = -DUSE_IOMX
-SOURCES_iomx = $(SOURCES_omxil) iomx.cpp iomx.h
+SOURCES_iomx = $(SOURCES_omxil)
diff --git a/modules/codec/omxil/iomx.cpp b/modules/codec/omxil/iomx.cpp
index a207d9c..2491c6c 100644
--- a/modules/codec/omxil/iomx.cpp
+++ b/modules/codec/omxil/iomx.cpp
@@ -29,8 +29,6 @@
 #include <binder/MemoryDealer.h>
 #include <OMX_Component.h>
 
-#include "iomx.h"
-
 using namespace android;
 
 class IOMXContext {
@@ -272,7 +270,7 @@ static OMX_ERRORTYPE iomx_set_config(OMX_HANDLETYPE component, OMX_INDEXTYPE ind
     return get_error(ctx->iomx->setConfig(node->node, index, param, sizeof(OMX_BOOL)));
 }
 
-static OMX_ERRORTYPE iomx_get_handle(OMX_HANDLETYPE *handle_ptr, const char *component_name, OMX_PTR app_data, const OMX_CALLBACKTYPE *callbacks)
+OMX_ERRORTYPE OMX_GetHandle(OMX_HANDLETYPE *handle_ptr, OMX_STRING component_name, OMX_PTR app_data, OMX_CALLBACKTYPE *callbacks)
 {
     OMXNode* node = new OMXNode();
     node->app_data = app_data;
@@ -310,7 +308,7 @@ static OMX_ERRORTYPE iomx_get_handle(OMX_HANDLETYPE *handle_ptr, const char *com
     return OMX_ErrorNone;
 }
 
-static OMX_ERRORTYPE iomx_free_handle(OMX_HANDLETYPE handle)
+OMX_ERRORTYPE OMX_FreeHandle(OMX_HANDLETYPE handle)
 {
     OMXNode* node = (OMXNode*) ((OMX_COMPONENTTYPE*)handle)->pComponentPrivate;
     ctx->iomx->freeNode( node->node );
@@ -320,24 +318,28 @@ static OMX_ERRORTYPE iomx_free_handle(OMX_HANDLETYPE handle)
     return OMX_ErrorNone;
 }
 
-static OMX_ERRORTYPE iomx_init()
+OMX_ERRORTYPE OMX_Init(void)
 {
     OMXClient client;
     if (client.connect() != OK)
         return OMX_ErrorUndefined;
 
+    if (!ctx)
+        ctx = new IOMXContext();
     ctx->iomx = client.interface();
     ctx->iomx->listNodes(&ctx->components);
     return OMX_ErrorNone;
 }
 
-static OMX_ERRORTYPE iomx_deinit()
+OMX_ERRORTYPE OMX_Deinit(void)
 {
     ctx->iomx = NULL;
+    delete ctx;
+    ctx = NULL;
     return OMX_ErrorNone;
 }
 
-static OMX_ERRORTYPE iomx_component_name_enum(OMX_STRING component_name, OMX_U32 name_length, OMX_U32 index)
+OMX_ERRORTYPE OMX_ComponentNameEnum(OMX_STRING component_name, OMX_U32 name_length, OMX_U32 index)
 {
     if (index >= ctx->components.size())
         return OMX_ErrorNoMore;
@@ -349,7 +351,7 @@ static OMX_ERRORTYPE iomx_component_name_enum(OMX_STRING component_name, OMX_U32
     return OMX_ErrorNone;
 }
 
-static OMX_ERRORTYPE iomx_get_roles_of_component(OMX_STRING component_name, OMX_U32 *num_roles, OMX_U8 **roles)
+OMX_ERRORTYPE OMX_GetRolesOfComponent(OMX_STRING component_name, OMX_U32 *num_roles, OMX_U8 **roles)
 {
     for( List<IOMX::ComponentInfo>::iterator it = ctx->components.begin(); it != ctx->components.end(); it++ ) {
         if (!strcmp(component_name, it->mName.string())) {
@@ -371,34 +373,3 @@ static OMX_ERRORTYPE iomx_get_roles_of_component(OMX_STRING component_name, OMX_
     return OMX_ErrorInvalidComponentName;
 }
 
-void* iomx_dlopen(const char *)
-{
-    if (!ctx)
-        ctx = new IOMXContext();
-    return ctx;
-}
-
-void iomx_dlclose(void *handle)
-{
-    IOMXContext *ctx = (IOMXContext*) handle;
-    delete ctx;
-    ::ctx = NULL;
-}
-
-void *iomx_dlsym(void *, const char *name)
-{
-    if (!strcmp(name, "OMX_Init"))
-        return (void*) iomx_init;
-    if (!strcmp(name, "OMX_Deinit"))
-        return (void*) iomx_deinit;
-    if (!strcmp(name, "OMX_GetHandle"))
-        return (void*) iomx_get_handle;
-    if (!strcmp(name, "OMX_FreeHandle"))
-        return (void*) iomx_free_handle;
-    if (!strcmp(name, "OMX_ComponentNameEnum"))
-        return (void*) iomx_component_name_enum;
-    if (!strcmp(name, "OMX_GetRolesOfComponent"))
-        return (void*) iomx_get_roles_of_component;
-    return NULL;
-}
-
diff --git a/modules/codec/omxil/iomx.h b/modules/codec/omxil/iomx.h
deleted file mode 100644
index 453f817..0000000
--- a/modules/codec/omxil/iomx.h
+++ /dev/null
@@ -1,33 +0,0 @@
-/*****************************************************************************
- * iomx.h: Interface for the IOMX OpenMAX wrapper
- *****************************************************************************
- * Copyright (C) 2011 VLC authors and VideoLAN
- *
- * Authors: Martin Storsjo <martin at martin.st>
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU Lesser General Public License as published by
- * the Free Software Foundation; either version 2.1 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU Lesser General Public License for more details.
- *
- * You should have received a copy of the GNU Lesser General Public License
- * along with this program; if not, write to the Free Software Foundation,
- * Inc., 51 Franklin Street, Fifth Floor, Boston MA 02110-1301, USA.
- *****************************************************************************/
-
-#ifdef __cplusplus
-extern "C" {
-#endif
-
-void* iomx_dlopen(const char *name);
-void iomx_dlclose(void *handle);
-void* iomx_dlsym(void *handle, const char *name);
-
-#ifdef __cplusplus
-}
-#endif
diff --git a/modules/codec/omxil/omxil.c b/modules/codec/omxil/omxil.c
index aad2b02..a46fc7d 100644
--- a/modules/codec/omxil/omxil.c
+++ b/modules/codec/omxil/omxil.c
@@ -30,10 +30,10 @@
 
 #include <dlfcn.h>
 #if defined(USE_IOMX)
-# include "iomx.h"
-# define dll_open(name) iomx_dlopen(name)
-# define dll_close(handle) iomx_dlclose(handle)
-# define dlsym(handle, name) iomx_dlsym(handle, name)
+/* On dll_open, just check that the OMX_Init symbol already is loaded */
+# define dll_open(name) dlsym(RTLD_DEFAULT, "OMX_Init")
+# define dll_close(handle) do { } while (0)
+# define dlsym(handle, name) dlsym(RTLD_DEFAULT, name)
 #else
 # define dll_open(name) dlopen( name, RTLD_NOW )
 # define dll_close(handle) dlclose(handle)
-- 
1.7.2.5



From martin at martin.st  Fri Dec 16 22:47:05 2011
From: martin at martin.st (=?UTF-8?q?Martin=20Storsj=C3=B6?=)
Date: Fri, 16 Dec 2011 23:47:05 +0200
Subject: [vlc-devel] [PATCH 4/5] android: Don't add private header include
	directories when building the full VLC tree
In-Reply-To: <1324072028-6485-1-git-send-email-martin@martin.st>
References: <1324072028-6485-1-git-send-email-martin@martin.st>
Message-ID: <1324072028-6485-5-git-send-email-martin@martin.st>

These aren't necessary any longer.

The approach of adding private headers to the build of the VLC
core doesn't work well if one wants to target multiple private
ABI versions in one build.

Instead, small shared libraries for each private ABI version
can be built separately.
---
 extras/package/android/configure.sh |    5 -----
 1 files changed, 0 insertions(+), 5 deletions(-)

diff --git a/extras/package/android/configure.sh b/extras/package/android/configure.sh
index 5461daf..7ed5e4c 100755
--- a/extras/package/android/configure.sh
+++ b/extras/package/android/configure.sh
@@ -29,11 +29,6 @@ fi
 CPPFLAGS="-I${ANDROID_NDK}/sources/cxx-stl/gnu-libstdc++/include -I${ANDROID_NDK}/sources/cxx-stl/gnu-libstdc++/libs/${CXX_TARGET}/include"
 LDFLAGS="$LDFLAGS -L${ANDROID_NDK}/sources/cxx-stl/gnu-libstdc++/libs/${CXX_TARGET}"
 
-if [ -n "$ANDROID_SYS_HEADERS" ]; then
-    CPPFLAGS="$CPPFLAGS -I${ANDROID_SYS_HEADERS}/frameworks/base/include -I${ANDROID_SYS_HEADERS}/system/core/include"
-    LDFLAGS="$LDFLAGS -L${ANDROID_LIBS}"
-fi
-
 SYSROOT=$ANDROID_NDK/platforms/$ANDROID_API/arch-arm
 ANDROID_BIN=$ANDROID_NDK/toolchains/arm-linux-androideabi-4.4.3/prebuilt/*-x86/bin/
 CROSS_COMPILE=${ANDROID_BIN}/arm-linux-androideabi-
-- 
1.7.2.5



From martin at martin.st  Fri Dec 16 22:47:04 2011
From: martin at martin.st (=?UTF-8?q?Martin=20Storsj=C3=B6?=)
Date: Fri, 16 Dec 2011 23:47:04 +0200
Subject: [vlc-devel] [PATCH 3/5] omxil: Remove the header check for the iomx
	module
In-Reply-To: <1324072028-6485-1-git-send-email-martin@martin.st>
References: <1324072028-6485-1-git-send-email-martin@martin.st>
Message-ID: <1324072028-6485-4-git-send-email-martin@martin.st>

Building the iomx version of the omxil module itself doesn't
require the private headers any longer, they're only necessary
when building the iomx wrapper code (which should be built by the
NDK makefile instead).
---
 configure.ac |   13 +------------
 1 files changed, 1 insertions(+), 12 deletions(-)

diff --git a/configure.ac b/configure.ac
index ae42157..53325ef 100644
--- a/configure.ac
+++ b/configure.ac
@@ -2253,18 +2253,7 @@ AC_ARG_ENABLE(iomx,
   [  --enable-iomx           iomx codec module (default disabled)])
 if test "${enable_iomx}" = "yes"
 then
-  VLC_SAVE_FLAGS
-  CPPFLAGS="${CPPFLAGS} -I${srcdir}/modules/codec/omxil" # For the OMX headers
-  AC_LANG_PUSH([C++])
-  AC_CHECK_HEADER(media/stagefright/OMXClient.h, [
-    VLC_ADD_PLUGIN([iomx])
-    VLC_ADD_CXXFLAGS([iomx], [-fno-exceptions -fno-rtti])
-    VLC_ADD_LIBS([iomx], [-lstagefright -lmedia -lutils -lbinder])
-  ], [
-    AC_MSG_ERROR("Could not find IOMX headers")
-  ])
-  AC_LANG_POP([C++])
-  VLC_RESTORE_FLAGS
+  VLC_ADD_PLUGIN([iomx])
 fi
 
 dnl
-- 
1.7.2.5



From martin at martin.st  Fri Dec 16 22:47:06 2011
From: martin at martin.st (=?UTF-8?q?Martin=20Storsj=C3=B6?=)
Date: Fri, 16 Dec 2011 23:47:06 +0200
Subject: [vlc-devel] [PATCH 5/5] omxil: Use a fake library name when loading
	IOMX
In-Reply-To: <1324072028-6485-1-git-send-email-martin@martin.st>
References: <1324072028-6485-1-git-send-email-martin@martin.st>
Message-ID: <1324072028-6485-6-git-send-email-martin@martin.st>

The iomx loading stub doesn't use the library name.
If there's errors in loading the function pointers, this
gives slightly more correct error messages.
---
 modules/codec/omxil/omxil.c |    4 ++++
 1 files changed, 4 insertions(+), 0 deletions(-)

diff --git a/modules/codec/omxil/omxil.c b/modules/codec/omxil/omxil.c
index a46fc7d..ef23a0d 100644
--- a/modules/codec/omxil/omxil.c
+++ b/modules/codec/omxil/omxil.c
@@ -58,9 +58,13 @@
  *****************************************************************************/
 static const char *ppsz_dll_list[] =
 {
+#if defined(USE_IOMX)
+    "libiomx.so", /* Not used when using IOMX, the lib should already be loaded */
+#else
     "libOMX_Core.so", /* TI OMAP IL core */
     "libOmxCore.so", /* Qualcomm IL core */
     "libomxil-bellagio.so",  /* Bellagio IL core */
+#endif
     0
 };
 
-- 
1.7.2.5



From martin at martin.st  Fri Dec 16 22:47:07 2011
From: martin at martin.st (=?UTF-8?q?Martin=20Storsj=C3=B6?=)
Date: Fri, 16 Dec 2011 23:47:07 +0200
Subject: [vlc-devel] [PATCH 1/2] Build the iomx glue code as a separate
	shared library
In-Reply-To: <1324072028-6485-1-git-send-email-martin@martin.st>
References: <1324072028-6485-1-git-send-email-martin@martin.st>
Message-ID: <1324072028-6485-7-git-send-email-martin@martin.st>

---
This is for the vlc-android repo, and is needed to use the
new form of the iomx wrapper.

 vlc-android/jni/Android.mk                         |   11 ++++++++++-
 .../src/org/videolan/vlc/android/LibVLC.java       |    5 +++++
 2 files changed, 15 insertions(+), 1 deletions(-)

diff --git a/vlc-android/jni/Android.mk b/vlc-android/jni/Android.mk
index f8a1242..06b547c 100644
--- a/vlc-android/jni/Android.mk
+++ b/vlc-android/jni/Android.mk
@@ -22,7 +22,16 @@ LOCAL_LDLIBS := -L$(VLC_CONTRIB)/lib \
 	-lavformat -lavcodec -lswscale -lavutil -lpostproc \
 	-lmpeg2 -lpng -ldca -ldvbpsi -ltwolame -lkate -llog -la52 \
 	-lebml -lmatroska -ltag \
-	-L$(ANDROID_LIBS) -lgcc -lstagefright -lmedia -lutils -lbinder \
 	$(ANDROID_NDK)/sources/cxx-stl/gnu-libstdc++/libs/$(ARCH)/libstdc++.a
 
 include $(BUILD_SHARED_LIBRARY)
+
+
+include $(CLEAR_VARS)
+
+LOCAL_MODULE     := libiomx
+LOCAL_SRC_FILES  := ../$(VLC_SRC_DIR)/modules/codec/omxil/iomx.cpp
+LOCAL_C_INCLUDES := $(VLC_SRC_DIR)/modules/codec/omxil $(ANDROID_SYS_HEADERS)/frameworks/base/include $(ANDROID_SYS_HEADERS)/system/core/include
+LOCAL_LDLIBS     := -L$(ANDROID_LIBS) -lgcc -lstagefright -lmedia -lutils -lbinder
+
+include $(BUILD_SHARED_LIBRARY)
diff --git a/vlc-android/src/org/videolan/vlc/android/LibVLC.java b/vlc-android/src/org/videolan/vlc/android/LibVLC.java
index 8cd436e..8f9b67a 100644
--- a/vlc-android/src/org/videolan/vlc/android/LibVLC.java
+++ b/vlc-android/src/org/videolan/vlc/android/LibVLC.java
@@ -30,6 +30,11 @@ public class LibVLC {
     /* Load library before object instantiation */
     static {
         try {
+            System.loadLibrary("iomx");
+        } catch (Throwable t) {
+            t.printStackTrace();
+        }
+        try {
             System.loadLibrary("vlcjni");
         } catch (UnsatisfiedLinkError ule) {
             Log.e(TAG, "Can't load vlcjni library: " + ule);
-- 
1.7.2.5



From martin at martin.st  Fri Dec 16 22:47:08 2011
From: martin at martin.st (=?UTF-8?q?Martin=20Storsj=C3=B6?=)
Date: Fri, 16 Dec 2011 23:47:08 +0200
Subject: [vlc-devel] [PATCH 2/2] Build the iomx shared library for ice cream
	sandwich, too
In-Reply-To: <1324072028-6485-1-git-send-email-martin@martin.st>
References: <1324072028-6485-1-git-send-email-martin@martin.st>
Message-ID: <1324072028-6485-8-git-send-email-martin@martin.st>

---
I haven't got any proper working ICS device with working
hw decoding yet, but this shows the principle of how to build
different versions of the IOMX plugin into one single apk.

 install.sh                                         |   10 +++++++---
 vlc-android/jni/Android.mk                         |   13 +++++++++++--
 .../src/org/videolan/vlc/android/LibVLC.java       |    6 +++++-
 3 files changed, 23 insertions(+), 6 deletions(-)

diff --git a/install.sh b/install.sh
index d2bec3a..116405c 100644
--- a/install.sh
+++ b/install.sh
@@ -13,9 +13,13 @@ echo "Fetching Android system headers"
 # "Stagefright: Memcpy optimization on output port." (available
 # upstream at https://www.codeaurora.org/gitweb/quic/la/?p=platform/frameworks/base.git;a=commit;h=052368f194c9fc180b9b0335b60114a2f1fb88d8),
 # which adds some vtable entries needed on newer qualcomm devices.
-$GIT clone -b gingerbread --depth=1 git://github.com/CyanogenMod/android_frameworks_base.git android-headers/frameworks/base
-$GIT clone -b gingerbread --depth=1 git://github.com/CyanogenMod/android_system_core.git android-headers/system/core
-export ANDROID_SYS_HEADERS=${PWD}/android-headers
+$GIT clone -b gingerbread --depth=1 git://github.com/CyanogenMod/android_frameworks_base.git android-headers-gingerbread/frameworks/base
+$GIT clone -b gingerbread --depth=1 git://github.com/CyanogenMod/android_system_core.git android-headers-gingerbread/system/core
+$GIT clone -b ics --depth=1 git://github.com/CyanogenMod/android_frameworks_base.git android-headers-gingerbread/frameworks/base
+$GIT clone -b ics --depth=1 git://github.com/CyanogenMod/android_system_core.git android-headers-ics/system/core
+$GIT clone -b ics --depth=1  git://github.com/CyanogenMod/android_hardware_libhardware.git android-headers-ics/hardware/libhardware
+export ANDROID_SYS_HEADERS_GINGERBREAD=${PWD}/android-headers-gingerbread
+export ANDROID_SYS_HEADERS_ICS=${PWD}/android-headers-ics
 
 echo "Fetching Android libraries for linking"
 # Libraries from any froyo/gingerbread device/emulator should work
diff --git a/vlc-android/jni/Android.mk b/vlc-android/jni/Android.mk
index 06b547c..2d0b6a6 100644
--- a/vlc-android/jni/Android.mk
+++ b/vlc-android/jni/Android.mk
@@ -29,9 +29,18 @@ include $(BUILD_SHARED_LIBRARY)
 
 include $(CLEAR_VARS)
 
-LOCAL_MODULE     := libiomx
+LOCAL_MODULE     := libiomx-gingerbread
 LOCAL_SRC_FILES  := ../$(VLC_SRC_DIR)/modules/codec/omxil/iomx.cpp
-LOCAL_C_INCLUDES := $(VLC_SRC_DIR)/modules/codec/omxil $(ANDROID_SYS_HEADERS)/frameworks/base/include $(ANDROID_SYS_HEADERS)/system/core/include
+LOCAL_C_INCLUDES := $(VLC_SRC_DIR)/modules/codec/omxil $(ANDROID_SYS_HEADERS_GINGERBREAD)/frameworks/base/include $(ANDROID_SYS_HEADERS_GINGERBREAD)/system/core/include
+LOCAL_LDLIBS     := -L$(ANDROID_LIBS) -lgcc -lstagefright -lmedia -lutils -lbinder
+
+include $(BUILD_SHARED_LIBRARY)
+
+include $(CLEAR_VARS)
+
+LOCAL_MODULE     := libiomx-ics
+LOCAL_SRC_FILES  := ../$(VLC_SRC_DIR)/modules/codec/omxil/iomx.cpp
+LOCAL_C_INCLUDES := $(VLC_SRC_DIR)/modules/codec/omxil $(ANDROID_SYS_HEADERS_ICS)/frameworks/base/include $(ANDROID_SYS_HEADERS_ICS)/frameworks/base/native/include $(ANDROID_SYS_HEADERS_ICS)/system/core/include $(ANDROID_SYS_HEADERS_ICS)/hardware/libhardware/include
 LOCAL_LDLIBS     := -L$(ANDROID_LIBS) -lgcc -lstagefright -lmedia -lutils -lbinder
 
 include $(BUILD_SHARED_LIBRARY)
diff --git a/vlc-android/src/org/videolan/vlc/android/LibVLC.java b/vlc-android/src/org/videolan/vlc/android/LibVLC.java
index 8f9b67a..c07c384 100644
--- a/vlc-android/src/org/videolan/vlc/android/LibVLC.java
+++ b/vlc-android/src/org/videolan/vlc/android/LibVLC.java
@@ -4,6 +4,7 @@ import android.util.Log;
 import android.view.Surface;
 import android.preference.PreferenceManager;
 import android.content.SharedPreferences;
+import android.os.Build;
 
 public class LibVLC {
     private static final String TAG = "VLC/LibVLC";
@@ -30,7 +31,10 @@ public class LibVLC {
     /* Load library before object instantiation */
     static {
         try {
-            System.loadLibrary("iomx");
+            if (Build.VERSION.SDK_INT <= 10)
+                System.loadLibrary("iomx-gingerbread");
+            else /* No honeycomb build for now */
+                System.loadLibrary("iomx-ics");
         } catch (Throwable t) {
             t.printStackTrace();
         }
-- 
1.7.2.5



From jb at videolan.org  Sat Dec 17 04:21:40 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Sat, 17 Dec 2011 04:21:40 +0100
Subject: [vlc-devel] [PATCH 0/5] android/iomx as a separate shared
 library
In-Reply-To: <1324072028-6485-1-git-send-email-martin@martin.st>
References: <1324072028-6485-1-git-send-email-martin@martin.st>
Message-ID: <20111217032140.GA1066@videolan.org>

On Fri, Dec 16, 2011 at 11:47:01PM +0200, Martin Storsj? wrote :
> Also, I kinda like the amount of removed code in this
> patchset. :-)

I do too...

It's a bit too late to understand it correctly, but it seems cool.


-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Sat Dec 17 04:24:08 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Sat, 17 Dec 2011 04:24:08 +0100
Subject: [vlc-devel] [PATCH] Fix stack overflow in ExecuteCommand
In-Reply-To: <4EEA39D0.7090106@ymail.com>
References: <4EEA2B02.1030201@gmail.com> <201112151938.26981.remi@remlab.net>
	<4EEA34FC.2080103@ymail.com> <4EEA36F3.2020100@ymail.com>
	<4EEA39D0.7090106@ymail.com>
Message-ID: <20111217032408.GA1423@videolan.org>

On Thu, Dec 15, 2011 at 06:17:52PM +0000, Cheng Sun wrote :
> Revised patch, which corrects a size calculation

Ping for review.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Sat Dec 17 04:32:08 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Sat, 17 Dec 2011 04:32:08 +0100
Subject: [vlc-devel] Mobile web interface patch
In-Reply-To: <3366F855-5D14-460F-A81E-B2094FEA5C9F@var1able.ru>
References: <3366F855-5D14-460F-A81E-B2094FEA5C9F@var1able.ru>
Message-ID: <20111217033208.GA3087@videolan.org>

On Fri, Dec 16, 2011 at 06:00:45PM +0400, ?????????? ???????? wrote :
> http://www.google-melange.com/gci/task/view/google/gci2011/7180436

Patch does not apply...

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Sat Dec 17 04:34:32 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Sat, 17 Dec 2011 04:34:32 +0100
Subject: [vlc-devel] [PATCH] VDR recording detection
In-Reply-To: <4EEA7489.8060706@web.de>
References: <4EE9DF6F.6060908@web.de> <201112151712.11215.remi@remlab.net>
	<4EEA7489.8060706@web.de>
Message-ID: <20111217033432.GA3468@videolan.org>

On Thu, Dec 15, 2011 at 11:28:25PM +0100, Tobias G?ntner wrote :
> See attachment.

Applied.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Sat Dec 17 04:35:53 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Sat, 17 Dec 2011 04:35:53 +0100
Subject: [vlc-devel] [PATCH] VDR recording detection
In-Reply-To: <4EE9DF6F.6060908@web.de>
References: <4EE9DF6F.6060908@web.de>
Message-ID: <20111217033553.GB3468@videolan.org>

On Thu, Dec 15, 2011 at 12:52:15PM +0100, Tobias G?ntner wrote :
> This patch changes the detection of VDR recordings. I removed the check
> for index and meta data files because neither is required for playback
> as far as VLC is concerned. To avoid false positives, I improved the
> directory name check a bit. Also, wrong directory names are easier to
> repair than missing files IMHO.

Speaking of VDR, do you set the duration for titles?

Cf #4030

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From funman at videolan.org  Sat Dec 17 04:36:27 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Fri, 16 Dec 2011 22:36:27 -0500
Subject: [vlc-devel] [PATCH] Fix stack overflow in ExecuteCommand
In-Reply-To: <4EEA39D0.7090106@ymail.com>
References: <4EEA2B02.1030201@gmail.com> <201112151938.26981.remi@remlab.net>
	<4EEA34FC.2080103@ymail.com> <4EEA36F3.2020100@ymail.com>
	<4EEA39D0.7090106@ymail.com>
Message-ID: <20111216223627.1c1d0b43@tok>

Le Thu, 15 Dec 2011 18:17:52 +0000,
Cheng Sun <cheng.sun at ymail.com> a ?crit :

> From 1075e863da941207cdab90713baff16c1aad76bd Mon Sep 17 00:00:00 2001
> From: Cheng Sun <chengsun9 at gmail.com>
> Date: Thu, 15 Dec 2011 17:10:52 +0000
> Subject: [PATCH] Fix stack overflow in ExecuteCommand
> 
> ---
>  src/input/vlmshell.c |   27 ++++++++++++++++++++++-----
>  1 files changed, 22 insertions(+), 5 deletions(-)
> 
> diff --git a/src/input/vlmshell.c b/src/input/vlmshell.c
> index ab00d67..71fd952 100644
> --- a/src/input/vlmshell.c
> +++ b/src/input/vlmshell.c
> @@ -847,9 +847,19 @@ int ExecuteCommand( vlm_t *p_vlm, const char *psz_command,
>                             vlm_message_t **pp_message )
>  {
>      size_t i_command = 0;
> -    char buf[strlen (psz_command) + 1], *psz_buf = buf;
> -    char *ppsz_command[3+sizeof (buf) / 2];
> +    size_t i_command_len = strlen( psz_command );
> +    char *buf = malloc( i_command_len + 1 ), *psz_buf = buf;
> +    size_t i_ppsz_command_len = (3+i_command_len / 2);

Your patch reduce the size of ppsz_command, is that wanted?

> +    char **ppsz_command = malloc( i_ppsz_command_len * sizeof(char *) );
>      vlm_message_t *p_message = NULL;
> +    int i_ret = 0;
> +
> +    if( !psz_buf || !ppsz_command )
> +    {
> +        p_message = vlm_MessageNew( ppsz_command[0],
> +                "Memory allocation failed - command too long?" );

Is it possible to have the command size in the message? (with %zu or something)

> +        goto error;
> +    }
>  
>      /* First, parse the line and cut it */
>      while( *psz_command != '\0' )
> @@ -877,7 +887,7 @@ int ExecuteCommand( vlm_t *p_vlm, const char *psz_command,
>              goto error;
>          }
>  
> -        assert (i_command < (sizeof (ppsz_command) / sizeof (ppsz_command[0])));
> +        assert (i_command < i_ppsz_command_len);
>  
>          ppsz_command[i_command] = psz_buf;
>          memcpy (psz_buf, psz_command, psz_temp - psz_command);
> @@ -889,7 +899,7 @@ int ExecuteCommand( vlm_t *p_vlm, const char *psz_command,
>          psz_buf += psz_temp - psz_command + 1;
>          psz_command = psz_temp;
>  
> -        assert (buf + sizeof (buf) >= psz_buf);
> +        assert (buf + i_command_len + 1 >= psz_buf);
>      }
>  
>      /*
> @@ -920,13 +930,20 @@ int ExecuteCommand( vlm_t *p_vlm, const char *psz_command,
>  
>  success:
>      *pp_message = p_message;
> +    free( buf );
> +    free( ppsz_command );
>      return VLC_SUCCESS;
>  
>  syntax_error:
> -    return ExecuteSyntaxError( ppsz_command[0], pp_message );
> +    i_ret = ExecuteSyntaxError( ppsz_command[0], pp_message );
> +    free( buf );
> +    free( ppsz_command );
> +    return i_ret;
>  
>  error:
>      *pp_message = p_message;
> +    free( buf );
> +    free( ppsz_command );
>      return VLC_EGENERIC;
>  }
>  



-- 
Rafa?l Carr?


From jb at videolan.org  Sat Dec 17 04:45:51 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Sat, 17 Dec 2011 04:45:51 +0100
Subject: [vlc-devel] [RFC][PATCH 1/2] contrib: check os x gcc v4.2
 before overriding
In-Reply-To: <CAKhCBsz0yVJ8xfu3SGDtk9ggBf3H_XOmScvhnJ-Oe588m_YuSg@mail.gmail.com>
References: <1323894926-21050-1-git-send-email-enrique.osuna@gmail.com>
	<1323894926-21050-2-git-send-email-enrique.osuna@gmail.com>
	<20111214160410.1d799a35@tok> <20111214211031.GA22396@videolan.org>
	<CAKhCBszYkVoi378owd8F8A+j9n-qgPvGv4rumB_QU5VrPL+kpA@mail.gmail.com>
	<20111215120614.71b2d93b@tok>
	<CAKhCBsxPQssH2Ti569TkqFxB_D2nNr1vpcDXUpJkA6mwFC8CHg@mail.gmail.com>
	<20111215174607.1c1555be@tok>
	<CAKhCBsz0yVJ8xfu3SGDtk9ggBf3H_XOmScvhnJ-Oe588m_YuSg@mail.gmail.com>
Message-ID: <20111217034551.GA5710@videolan.org>

On Thu, Dec 15, 2011 at 11:50:01PM +0100, Faustino Osuna wrote :
> In the mean time, should I update the Wiki to reflect that LLVM GCC 4.2 is
> not a viable option to compile the contribs?

Yes.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Sat Dec 17 04:52:23 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Sat, 17 Dec 2011 04:52:23 +0100
Subject: [vlc-devel] typo fix for french po
In-Reply-To: <CALKXa+_AZead0OeUFo9i=JFOn+LEVJs4g+3m1fcYyFS4MsDJsg@mail.gmail.com>
References: <CALKXa+_AZead0OeUFo9i=JFOn+LEVJs4g+3m1fcYyFS4MsDJsg@mail.gmail.com>
Message-ID: <20111217035223.GA7113@videolan.org>

On Wed, Dec 14, 2011 at 09:01:52PM +0100, beno?t tuduri wrote :
> you can forget my previous email. I added some minor fix in attachment.

Applied.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Sat Dec 17 04:53:00 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Sat, 17 Dec 2011 04:53:00 +0100
Subject: [vlc-devel] [PATCH] yadif : Add SSSE3 and SSE2 support. porting
 from FFmpeg.
In-Reply-To: <CAOa_K+shj1Wy4pm3=sFZxw4_nM+k-VCZnRauiQujeV0+LS4CGw@mail.gmail.com>
References: <CAOa_K+vdRUzBj4cE9qoLyU19+NYnVuPx03UqdOEXgkT6fw60Gg@mail.gmail.com>
	<20111205214242.GA27741@videolan.org>
	<CAOa_K+vcob=0Af2vCLHH7jeQeBXhO5TnQtieo_LWNmC9_msgSQ@mail.gmail.com>
	<201112091941.15730.remi@remlab.net>
	<CAOa_K+shj1Wy4pm3=sFZxw4_nM+k-VCZnRauiQujeV0+LS4CGw@mail.gmail.com>
Message-ID: <20111217035300.GB7113@videolan.org>

On Sat, Dec 10, 2011 at 02:55:16PM +0900, Naohiro KORIYAMA wrote :
> OK. I reverted that part. and used VLC_MMX and VLC_SSE.
> I confirmed the compilation issue is fixed!

R?mi, is it ok now?

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Sat Dec 17 04:54:18 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Sat, 17 Dec 2011 04:54:18 +0100
Subject: [vlc-devel] patch for french translation on macosx
In-Reply-To: <CALKXa+-31pVS-aOUvk-OhXpOyG-dfME9w+M=2B-TtS-Z-uinRQ@mail.gmail.com>
References: <CALKXa+-31pVS-aOUvk-OhXpOyG-dfME9w+M=2B-TtS-Z-uinRQ@mail.gmail.com>
Message-ID: <20111217035418.GC7113@videolan.org>

On Sun, Dec 11, 2011 at 04:19:57PM +0100, beno?t tuduri wrote :
> You can find a patch to fix a typo error.

Applied, later on.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From mamontor at upvnet.upv.es  Fri Dec 16 19:41:19 2011
From: mamontor at upvnet.upv.es (Mario Montagud Climent)
Date: Fri, 16 Dec 2011 19:41:19 +0100
Subject: [vlc-devel] VLC Support for RTCP extensions anb playback buffer
	management
Message-ID: <20111216194119.21193386eadn5qkv@wm.upv.es>


Dear VLC Developers,

This is my first message to this mailing list.

My name is Mario Montagud. I am a PhD student at Politechnic  
University of Valencia (UPV) in Spain. I am working under Prof.  
Fernando Boronat supervision.  
(http://personales.gan.upv.es/~fboronat/Research/index_investig_en.html).

I would like to ask you some information about the ?netsync? module in  
VLC and the VLC capabilities.

My PhD work is about Multimedia Communications and Synchronization  
Techniques. Concretely, our research group is designing an  
Inter-Destination Multimedia Synchronization (IDMS) proposal that aims  
to synchronize the playout of media streams across separated clients  
(which can be placed in different networks, with different network  
latency). For that purpose, we take advantage of the feedback and  
extension capabilities of RTP/RTCP protocols.

You can find more information about the preliminary version of our  
IDMS approach in the reference [Bor08], in which it was implemented by  
using VIC (for video) and RAT (for audio) mbone tools, and tested in a  
real scenario (our University WAN).

In the initial stages of my PhD, I have implemented this IDMS approach  
in a network simulator (NS-2), and I?ve added some enhancements to  
that solution (see [Bor11]). The simulation results have been  
satisfactory.

Currently, an evolved version of our IDMS proposal is under  
standardization status within the IETF (see ref [Bra11]). We are  
collaborating in its specification/design in conjunction with some  
Dutch researchers from TNO (Delft).

On the one hand, a new RTCP report block (RTCP XR for IDMS block) has  
been specified to inform the source (or a sync manager) about receipt  
and playout times for specific RTP media packets (for that purpose all  
the participants must share a common available clock source, e.g.  
NTP). On the other hand a new RTCP packet type (RTCP IDMS packet) has  
been defined to send playout setting instructions to the sync clients  
(if an out of sync situation is detected by the sync manager).

My next objective is to implement and test that IDMS proposal in a  
real media framework/player. So, I?m searching for the most  
appropriated solution for it.

I am a newbie in VLC media player (as a developer). For that reason, I  
send this mail asking for your recommendation.

I know that the ?netsync? module in VLC can be used to keep several  
clients synchronised on a single VLC stream (using UDP datagrams, not  
using RTP/RTCP). It can also be used to synchronise several computers  
playing parts of a video wall.

Our goal is to synchronize a RTP network stream across multiple  
separated clients. We use for that purpose the global time instant  
(64-bit wall clock time, eg. NTP) at which a specific RTP packet has  
been received (or played out) by each sync client.

For that purpose, first, I need to implement the new RTCP packet types  
and block reports defined in [Bra11]. I think it should be done in  
live555 library.

I think the most difficult task will be the management of the playout  
processes in each media client to get in sync (e.g. to report on  
presentation time for a specific RTP packet ?e.g. the last one  
encapsulating a video frame-, and to manage the necessary playback  
adjustments (playout speed variation? Skips/pauses?).

Do you think VLC is a good choice for implementing our RTP/RTCP-based  
IDMS solution? Can I access to the playback buffer to know about  
presentation times of a specific video frame and the current playback  
buffer occupancy?

Could the "netsync" module be adapted for our IDMS purposes?

Your comments will be much appreciated.

Thank you very much in advance.

Regards from Spain,

Mario

PD: I apologize for the extension of the mail.

REFERENCES: (If you are interested, and don?t have access to them,  
please, tell me)

[Bor08] F. Boronat, J.C. Guerri, and J. Lloret, ?An RTP/RTCP based  
approach for multimedia group and inter-stream synchronization?,  
Multimedia Tools and Applications Journal, Vol. 40 (2), 285-319, June  
2008.

[Bor11] F. Boronat, M. Montagud, and V. Vidal, ?Smooth Control of  
Adaptive Media Playout to Acquire IDMS in Cluster-based Applications?,  
IEEE LCN 2011, Bonn, October 2011.

[Bra11] R. V. Brandenburg, H. Stokking, M.O. V. Deventer, Boronat, F.  
M. Montagud, K. Gross, ?RTCP for Inter-Destination Media  
Synchronization?, IETF Internet Draft, October 2011.




From fatbull at web.de  Sat Dec 17 05:20:47 2011
From: fatbull at web.de (=?ISO-8859-1?Q?Tobias_G=FCntner?=)
Date: Sat, 17 Dec 2011 05:20:47 +0100
Subject: [vlc-devel] [PATCH] VDR recording detection
In-Reply-To: <20111217033553.GB3468@videolan.org>
References: <4EE9DF6F.6060908@web.de> <20111217033553.GB3468@videolan.org>
Message-ID: <4EEC189F.6030101@web.de>

Am 17.12.2011 04:35, schrieb Jean-Baptiste Kempf:
> Speaking of VDR, do you set the duration for titles?
> 
> Cf #4030

See attachment.

Regards,
Tobias
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: 0001-Set-title-length-and-size.patch
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111217/52e3f8c0/attachment.ksh>

From jb at videolan.org  Sat Dec 17 05:28:10 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Sat, 17 Dec 2011 05:28:10 +0100
Subject: [vlc-devel] [PATCH] VDR recording detection
In-Reply-To: <4EEC189F.6030101@web.de>
References: <4EE9DF6F.6060908@web.de> <20111217033553.GB3468@videolan.org>
	<4EEC189F.6030101@web.de>
Message-ID: <20111217042810.GA13921@videolan.org>

On Sat, Dec 17, 2011 at 05:20:47AM +0100, Tobias G?ntner wrote :
> See attachment.

Cool.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Sat Dec 17 05:30:53 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Sat, 17 Dec 2011 05:30:53 +0100
Subject: [vlc-devel] [PATCH] Fix EQ HTTP UI partially
In-Reply-To: <CAOgZoSe2Bkrgmkr5LT7zRfT5QvosBBsSrnG5-p=baHYUfucGoA@mail.gmail.com>
References: <CAOgZoSd6MBJdT_DFKwWSD+AmNEBCA6EujhWyX6MH96hGur4AHg@mail.gmail.com>
	<20111214151328.GD22153@videolan.org>
	<CAOgZoSe2Bkrgmkr5LT7zRfT5QvosBBsSrnG5-p=baHYUfucGoA@mail.gmail.com>
Message-ID: <20111217043053.GA14734@videolan.org>

On Wed, Dec 14, 2011 at 10:27:30PM +0530, akash mehrotra wrote :
> This makes it functional but the order of the bars to adjust the gain per
> band is incorrect (labels are correct)
 
Applied, then.


-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From edward.c.wang at compdigitec.com  Sat Dec 17 05:50:09 2011
From: edward.c.wang at compdigitec.com (Edward Wang)
Date: Fri, 16 Dec 2011 23:50:09 -0500
Subject: [vlc-devel] [PATCH] Add filename sort option to playlist dialogue
In-Reply-To: <4EE945DF.2020504@compdigitec.com>
References: <4EE945DF.2020504@compdigitec.com>
Message-ID: <4EEC1F81.6090705@compdigitec.com>

Hi all:

The attached patch will add a requested feature - the option to sort by 
filename (in addition to the URI) in the playlist dialogue (see 
http://trac.videolan.org/vlc/ticket/5455).

Thanks,
     Edward Wang
-------------- next part --------------
A non-text attachment was scrubbed...
Name: patch5455.diff
Type: text/x-patch
Size: 5818 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111216/735f24e2/attachment.bin>

From remi at remlab.net  Sat Dec 17 09:42:00 2011
From: remi at remlab.net (=?iso-8859-15?q?R=E9mi?= Denis-Courmont)
Date: Sat, 17 Dec 2011 10:42:00 +0200
Subject: [vlc-devel] [PATCH] yadif : Add SSSE3 and SSE2 support. porting
	from FFmpeg.
In-Reply-To: <CAOa_K+shj1Wy4pm3=sFZxw4_nM+k-VCZnRauiQujeV0+LS4CGw@mail.gmail.com>
References: <CAOa_K+vdRUzBj4cE9qoLyU19+NYnVuPx03UqdOEXgkT6fw60Gg@mail.gmail.com>
	<201112091941.15730.remi@remlab.net>
	<CAOa_K+shj1Wy4pm3=sFZxw4_nM+k-VCZnRauiQujeV0+LS4CGw@mail.gmail.com>
Message-ID: <201112171042.02575.remi@remlab.net>

Le samedi 10 d?cembre 2011 07:55:16 Naohiro KORIYAMA, vous avez ?crit :
> OK. I reverted that part. and used VLC_MMX and VLC_SSE.
> I confirmed the compilation issue is fixed!

I cannot link it here...

.libs/libdeinterlace_plugin_la-algo_yadif.o: In function 
`yadif_filter_line_ssse3':
/home/remi/videolan/vlc/build/modules/video_filter/../../../modules/video_filter/deinterlace/yadif_template.h:249: 
undefined reference to `_pw_1'
/home/remi/videolan/vlc/build/modules/video_filter/../../../modules/video_filter/deinterlace/yadif_template.h:249: 
undefined reference to `_pb_1'

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From cheng.sun at ymail.com  Sat Dec 17 11:49:25 2011
From: cheng.sun at ymail.com (Cheng Sun)
Date: Sat, 17 Dec 2011 10:49:25 +0000
Subject: [vlc-devel] [PATCH] Fix stack overflow in ExecuteCommand
In-Reply-To: <20111216223627.1c1d0b43@tok>
References: <4EEA2B02.1030201@gmail.com> <201112151938.26981.remi@remlab.net>
	<4EEA34FC.2080103@ymail.com> <4EEA36F3.2020100@ymail.com>
	<4EEA39D0.7090106@ymail.com> <20111216223627.1c1d0b43@tok>
Message-ID: <4EEC73B5.2070806@ymail.com>

On 17/12/11 03:36, Rafa?l Carr? wrote:
> Your patch reduce the size of ppsz_command, is that wanted?
Nope.

> Is it possible to have the command size in the message? (with %zu or something)
Good point.

Revised patch attached.

- Cheng
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-Fix-stack-overflow-in-ExecuteCommand.patch
Type: text/x-patch
Size: 2965 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111217/d9b5f572/attachment.bin>

From nkoriyama at gmail.com  Sat Dec 17 15:38:52 2011
From: nkoriyama at gmail.com (Naohiro KORIYAMA)
Date: Sat, 17 Dec 2011 23:38:52 +0900
Subject: [vlc-devel] [PATCH] yadif : Add SSSE3 and SSE2 support. porting
 from FFmpeg.
In-Reply-To: <201112171042.02575.remi@remlab.net>
References: <CAOa_K+vdRUzBj4cE9qoLyU19+NYnVuPx03UqdOEXgkT6fw60Gg@mail.gmail.com>
	<201112091941.15730.remi@remlab.net>
	<CAOa_K+shj1Wy4pm3=sFZxw4_nM+k-VCZnRauiQujeV0+LS4CGw@mail.gmail.com>
	<201112171042.02575.remi@remlab.net>
Message-ID: <CAOa_K+tBjZCS5sSve_vwCt4DbzX2qpJGa37tTiZkX-tKOaUQgA@mail.gmail.com>

2011/12/17 R?mi Denis-Courmont <remi at remlab.net>:
> Le samedi 10 d?cembre 2011 07:55:16 Naohiro KORIYAMA, vous avez ?crit :
>> OK. I reverted that part. and used VLC_MMX and VLC_SSE.
>> I confirmed the compilation issue is fixed!
>
> I cannot link it here...
>
> .libs/libdeinterlace_plugin_la-algo_yadif.o: In function
> `yadif_filter_line_ssse3':
> /home/remi/videolan/vlc/build/modules/video_filter/../../../modules/video_filter/deinterlace/yadif_template.h:249:
> undefined reference to `_pw_1'
> /home/remi/videolan/vlc/build/modules/video_filter/../../../modules/video_filter/deinterlace/yadif_template.h:249:
> undefined reference to `_pb_1'
>

I didn't have Linux environment. So I just installed Ubuntu 64 bit,
and made a building environment.
And I just got same error.

I attached a new patch, it fixed the compilation issue.
I built and testedya the module on Linux 64bit, Mac OS X 64bit and
WIndows 32bit.

I'm sorry for inconvenience.

Best regards,

-- 
KORIYAMA, Naohiro
nkoriyama at gmail.com
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-yadif-Add-SSSE3-and-SSE2-support.-porting-from-FFmpe.patch
Type: application/octet-stream
Size: 25966 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111217/a3bf7570/attachment.obj>

From funman at videolan.org  Sat Dec 17 17:09:38 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Sat, 17 Dec 2011 11:09:38 -0500
Subject: [vlc-devel] [PATCH] Fix stack overflow in ExecuteCommand
In-Reply-To: <4EEC73B5.2070806@ymail.com>
References: <4EEA2B02.1030201@gmail.com> <201112151938.26981.remi@remlab.net>
	<4EEA34FC.2080103@ymail.com> <4EEA36F3.2020100@ymail.com>
	<4EEA39D0.7090106@ymail.com> <20111216223627.1c1d0b43@tok>
	<4EEC73B5.2070806@ymail.com>
Message-ID: <20111217110938.28a79c2c@tok>

Le Sat, 17 Dec 2011 10:49:25 +0000,
Cheng Sun <cheng.sun at ymail.com> a ?crit :

> On 17/12/11 03:36, Rafa?l Carr? wrote:
> > Your patch reduce the size of ppsz_command, is that wanted?
> Nope.
> 
> > Is it possible to have the command size in the message? (with %zu or something)
> Good point.
> 
> Revised patch attached.

looks ok

-- 
Rafa?l Carr?


From cheng.sun at ymail.com  Sat Dec 17 22:37:33 2011
From: cheng.sun at ymail.com (Cheng Sun)
Date: Sat, 17 Dec 2011 21:37:33 +0000
Subject: [vlc-devel] [PATCH] Fix QToolButtonExt edge cases
Message-ID: <4EED0B9D.6030703@ymail.com>

Fixes bug #5691

Previously QToolButtonExt had some undesirable edge case behaviour.

Firstly, once the user presses down on a QToolButtonExt there is no
way to cancel the action; either a short or a long click will be
generated. Compare this to a normal button, which can be cancelled by
releasing the mouse outside of the button area.

Secondly, with the mouse button held down, moving the mouse in and out
of the button area will generate multiple short click events, when no
event at all is desired.

This patch corrects this: by releasing the mouse outside the button area
no event is generated; additionally no short click events are generated
simply by moving the depressed mouse in and out of the button.

-- Cheng
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-Fix-QToolButtonExt-edge-cases.patch
Type: text/x-patch
Size: 3692 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111217/ed5c4550/attachment.bin>

From jb at videolan.org  Sun Dec 18 01:51:39 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Sun, 18 Dec 2011 01:51:39 +0100
Subject: [vlc-devel] [PATCH] Fix QToolButtonExt edge cases
In-Reply-To: <4EED0B9D.6030703@ymail.com>
References: <4EED0B9D.6030703@ymail.com>
Message-ID: <20111218005139.GA13403@videolan.org>

On Sat, Dec 17, 2011 at 09:37:33PM +0000, Cheng Sun wrote :
> Firstly, once the user presses down on a QToolButtonExt there is no
> way to cancel the action; either a short or a long click will be
> generated. Compare this to a normal button, which can be cancelled by
> releasing the mouse outside of the button area.

Agreed.

> Secondly, with the mouse button held down, moving the mouse in and out
> of the button area will generate multiple short click events, when no
> event at all is desired.

Agreed too.

> -    bool longClick;
> +    bool releasedIsDown;
> +    bool ignoreNextClick;

Very badly chosen variables, if I could say.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From chris2553 at googlemail.com  Sun Dec 18 00:00:47 2011
From: chris2553 at googlemail.com (Chris Clayton)
Date: Sat, 17 Dec 2011 23:00:47 +0000
Subject: [vlc-devel] Patch: fix  a tag name in MediaServer xml parsing
Message-ID: <201112172300.47189.chris2553@googlemail.com>

Hi,

According to http://upnp.org/specs/av/UPnP-av-MediaServer-v1-Device.pdf, the tag name of the element 
that provides the base URL for relative URLs is "URLBase".  The same is true for a v2 MediaServer. 
Make it so in MediaServer:: parseDeviceDescription(). The patch is against the 20111216 snapshot.

Chris

--- vlc-1.3.0-git/modules/services_discovery/upnp.cpp~  2011-12-16 09:32:04.000000000 +0000
+++ vlc-1.3.0-git/modules/services_discovery/upnp.cpp   2011-12-16 09:34:14.000000000 +0000
@@ -348,7 +348,7 @@ void MediaServer::parseDeviceDescription
     const char* psz_base_url = p_location;

     /* Try to extract baseURL */
-    IXML_NodeList* p_url_list = ixmlDocument_getElementsByTagName( p_doc, "baseURL" );
+    IXML_NodeList* p_url_list = ixmlDocument_getElementsByTagName( p_doc, "URLBase" );
     if ( p_url_list )
     {


-- 
The more I see, the more I know. The more I know, the less I understand. Changing Man - Paul Weller


From chris2553 at googlemail.com  Sun Dec 18 00:30:08 2011
From: chris2553 at googlemail.com (Chris Clayton)
Date: Sat, 17 Dec 2011 23:30:08 +0000
Subject: [vlc-devel] Patch: compatibility with upnp Mediaserver v2
Message-ID: <201112172330.08252.chris2553@googlemail.com>

Hi,

I own a Panasonic Blu-Ray recorder (model DMR-BW880), which also acts as DLNA Mediaserver for video 
or photos stored on its hard drive. The Mediaserver announces itself as being version 2, so the 
content directory service is also version 2.

The patch below (against the 20111216 snapshot) allows VLC to play content from the server. The 
Panasonic is the only mediaserver v2 device that I have, so I can't test whether the patch will 
work with other devices. Perhaps owners of other v2 mediaservers could try it out. At the least, my 
patch can be the basis for development by someone who knows more about upnp/dlna.

--- vlc-1.3.0-git/modules/services_discovery/upnp.hpp~	2011-12-16 10:04:39.000000000 +0000
+++ vlc-1.3.0-git/modules/services_discovery/upnp.hpp	2011-12-16 10:05:37.000000000 +0000
@@ -87,6 +87,7 @@ private:
     std::string _content_directory_control_url;
 
     int _i_subscription_timeout;
+    int _i_content_directory_service_version;
     Upnp_SID _subscription_id;
 };
 
--- vlc-1.3.0-git/modules/services_discovery/upnp.cpp~	2011-12-16 09:35:44.000000000 +0000
+++ vlc-1.3.0-git/modules/services_discovery/upnp.cpp	2011-12-16 10:09:12.000000000 +0000
@@ -380,7 +380,8 @@ void MediaServer::parseDeviceDescription
                 continue;
             }
 
-            if ( strcmp( MEDIA_SERVER_DEVICE_TYPE, psz_device_type ) != 0 )
+            if ( strncmp( MEDIA_SERVER_DEVICE_TYPE, psz_device_type,
+        		    strlen( MEDIA_SERVER_DEVICE_TYPE ) - 1 ) != 0 )
                 continue;
 
             const char* psz_udn = xml_getChildElementValue( p_device_element, "UDN" );
@@ -437,10 +438,14 @@ void MediaServer::parseDeviceDescription
                         continue;
                     }
 
-                    if ( strcmp( CONTENT_DIRECTORY_SERVICE_TYPE,
-                                psz_service_type ) != 0 )
+                    int k = strlen( CONTENT_DIRECTORY_SERVICE_TYPE ) - 1;
+                    if ( strncmp( CONTENT_DIRECTORY_SERVICE_TYPE,
+                                psz_service_type, k ) != 0 )
                         continue;
 
+		    p_server->_i_content_directory_service_version =
+			psz_service_type[k];
+
                     const char* psz_event_sub_url =
                         xml_getChildElementValue( p_service_element,
                                                   "eventSubURL" );
@@ -509,6 +514,7 @@ MediaServer::MediaServer( const char* ps
 
     _p_contents = NULL;
     _p_input_item = NULL;
+    _i_content_directory_service_version = 1;
 }
 
 MediaServer::~MediaServer()
@@ -597,6 +603,9 @@ IXML_Document* MediaServer::_browseActio
 
     char* psz_service_type = strdup( CONTENT_DIRECTORY_SERVICE_TYPE );
 
+    psz_service_type[strlen( psz_service_type ) - 1] =
+	_i_content_directory_service_version;
+
     int i_res;
 
     i_res = UpnpAddToAction( &p_action, "Browse",


-- 
The more I see, the more I know. The more I know, the less I understand. Changing Man - Paul Weller


From alexey at alexeysokolov.co.cc  Sun Dec 18 06:14:45 2011
From: alexey at alexeysokolov.co.cc (Alexey Sokolov)
Date: Sun, 18 Dec 2011 12:14:45 +0700
Subject: [vlc-devel] [PATCH 1/2] Build the iomx glue code as a separate
 shared library
In-Reply-To: <1324072028-6485-7-git-send-email-martin@martin.st>
References: <1324072028-6485-1-git-send-email-martin@martin.st>
	<1324072028-6485-7-git-send-email-martin@martin.st>
Message-ID: <4EED76C5.4090406@alexeysokolov.co.cc>

17.12.2011 04:47, Martin Storsj? ?????:
> diff --git a/vlc-android/src/org/videolan/vlc/android/LibVLC.java b/vlc-android/src/org/videolan/vlc/android/LibVLC.java
> index 8cd436e..8f9b67a 100644
> --- a/vlc-android/src/org/videolan/vlc/android/LibVLC.java
> +++ b/vlc-android/src/org/videolan/vlc/android/LibVLC.java
> @@ -30,6 +30,11 @@ public class LibVLC {
>      /* Load library before object instantiation */
>      static {
>          try {
> +            System.loadLibrary("iomx");
> +        } catch (Throwable t) {
> +            t.printStackTrace();
Why t.printStackTrace() instead of Log.e()?
> +        }
> +        try {
>              System.loadLibrary("vlcjni");
>          } catch (UnsatisfiedLinkError ule) {
>              Log.e(TAG, "Can't load vlcjni library: " + ule);


-- 
Best regards,
Alexey "DarthGandalf" Sokolov


-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 900 bytes
Desc: OpenPGP digital signature
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111218/7b2229c7/attachment.sig>

From alexey at alexeysokolov.co.cc  Sun Dec 18 06:18:14 2011
From: alexey at alexeysokolov.co.cc (Alexey Sokolov)
Date: Sun, 18 Dec 2011 12:18:14 +0700
Subject: [vlc-devel] [PATCH 1/2] Build the iomx glue code as a separate
 shared library
In-Reply-To: <4EED76C5.4090406@alexeysokolov.co.cc>
References: <1324072028-6485-1-git-send-email-martin@martin.st>
	<1324072028-6485-7-git-send-email-martin@martin.st>
	<4EED76C5.4090406@alexeysokolov.co.cc>
Message-ID: <4EED7796.7000204@alexeysokolov.co.cc>

18.12.2011 12:14, Alexey Sokolov ?????:
> 17.12.2011 04:47, Martin Storsj? ?????:
>> diff --git a/vlc-android/src/org/videolan/vlc/android/LibVLC.java b/vlc-android/src/org/videolan/vlc/android/LibVLC.java
>> index 8cd436e..8f9b67a 100644
>> --- a/vlc-android/src/org/videolan/vlc/android/LibVLC.java
>> +++ b/vlc-android/src/org/videolan/vlc/android/LibVLC.java
>> @@ -30,6 +30,11 @@ public class LibVLC {
>>      /* Load library before object instantiation */
>>      static {
>>          try {
>> +            System.loadLibrary("iomx");
>> +        } catch (Throwable t) {
>> +            t.printStackTrace();
> Why t.printStackTrace() instead of Log.e()?
Uhm, and why catching Throwable instead of only errors producable by
System.loadLibrary()?
>> +        }
>> +        try {
>>              System.loadLibrary("vlcjni");
>>          } catch (UnsatisfiedLinkError ule) {
>>              Log.e(TAG, "Can't load vlcjni library: " + ule);
>
>
>
> _______________________________________________
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
> http://mailman.videolan.org/listinfo/vlc-devel


-- 
Best regards,
Alexey "DarthGandalf" Sokolov

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111218/a92508b2/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 900 bytes
Desc: OpenPGP digital signature
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111218/a92508b2/attachment.sig>

From remi at remlab.net  Sun Dec 18 11:35:44 2011
From: remi at remlab.net (=?iso-8859-1?q?R=E9mi?= Denis-Courmont)
Date: Sun, 18 Dec 2011 12:35:44 +0200
Subject: [vlc-devel] [PATCH 1/2] Contribs: update live555 to latest
	snapshot
In-Reply-To: <1323992761-22393-2-git-send-email-jb@videolan.org>
References: <1323992761-22393-1-git-send-email-jb@videolan.org>
	<1323992761-22393-2-git-send-email-jb@videolan.org>
Message-ID: <201112181235.45371.remi@remlab.net>

Le vendredi 16 d?cembre 2011 01:46:00 Jean-Baptiste Kempf, vous avez ?crit :
> ---
>  contrib/src/live555/SHA512SUMS |    2 +-
>  contrib/src/live555/rules.mak  |   11 +----------
>  2 files changed, 2 insertions(+), 11 deletions(-)
> 
> diff --git a/contrib/src/live555/SHA512SUMS
> b/contrib/src/live555/SHA512SUMS index da74cf1..9e416a4 100644
> --- a/contrib/src/live555/SHA512SUMS
> +++ b/contrib/src/live555/SHA512SUMS
> @@ -1 +1 @@
> -42e424e8743cdbce2d6c87b6ceeb182a8868e870a0b233df987004a99e97e80c216e6c9967
> 74e936be40b1c3397a7c95ac8aa74eb6dcd8fac42a81dfcb4fd767 
> live.2011.11.08.tar.gz
> +00578686c1e154cea7638d19a6ab7675fa79d7da8f1848e9765954e3b02b82c54fd05e448
> 0e7ffcdc399c7ad3a0ad26aff0a8ce97222e26e66e345fe82906e02 
> live.2011.12.02.tar.gz diff --git a/contrib/src/live555/rules.mak
> b/contrib/src/live555/rules.mak index a267981..f5bb7c1 100644
> --- a/contrib/src/live555/rules.mak
> +++ b/contrib/src/live555/rules.mak
> @@ -1,7 +1,7 @@
>  # live555
> 
>  #LIVEDOTCOM_URL :=
> http://live555.com/liveMedia/public/live555-latest.tar.gz -LIVE555_FILE :=
> live.2011.11.08.tar.gz
> +LIVE555_FILE := live.2011.12.02.tar.gz
>  LIVEDOTCOM_URL :=
> http://live555sourcecontrol.googlecode.com/files/$(LIVE555_FILE)
> 
>  PKGS += live555
> @@ -15,15 +15,6 @@ live555: $(LIVE555_FILE) .sum-live555
>  	rm -Rf live
>  	$(UNPACK)
>  	chmod -R u+w live
> -	patch -p0 < $(SRC)/live555/live-uselocale.patch
> -	patch -p0 < $(SRC)/live555/live-inet_ntop.patch
> -	patch -p0 < $(SRC)/live555/live-intptr.patch
> -ifndef HAVE_WIN32
> -ifndef HAVE_WINCE
> -	patch -p0 < $(SRC)/live555/live-getaddrinfo.patch
> -endif
> -endif
> -	patch -p0 < $(SRC)/live555/live-cloexec.patch
>  	mv live $@
>  	touch $@

If it were that simple I would already have done it. You need to set the 
correct CPP macros if you remove the patches. Otherwise you basically just 
reintroduce the old bugs.

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From cheng.sun at ymail.com  Sun Dec 18 11:37:30 2011
From: cheng.sun at ymail.com (Cheng Sun)
Date: Sun, 18 Dec 2011 10:37:30 +0000
Subject: [vlc-devel] [PATCH] Fix QToolButtonExt edge cases
In-Reply-To: <20111218005139.GA13403@videolan.org>
References: <4EED0B9D.6030703@ymail.com> <20111218005139.GA13403@videolan.org>
Message-ID: <4EEDC26A.6030200@ymail.com>

On 18/12/11 00:51, Jean-Baptiste Kempf wrote:
> On Sat, Dec 17, 2011 at 09:37:33PM +0000, Cheng Sun wrote :
>> Firstly, once the user presses down on a QToolButtonExt there is no
>> way to cancel the action; either a short or a long click will be
>> generated. Compare this to a normal button, which can be cancelled by
>> releasing the mouse outside of the button area.
> Agreed.
>
>> Secondly, with the mouse button held down, moving the mouse in and out
>> of the button area will generate multiple short click events, when no
>> event at all is desired.
> Agreed too.
>
>> -    bool longClick;
>> +    bool releasedIsDown;
>> +    bool ignoreNextClick;
> Very badly chosen variables, if I could say.
>
True.

Attached revised patch which is slightly more verbose (and
understandable, hopefully)

-- Cheng
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-Fix-QToolButtonExt-edge-cases.patch
Type: text/x-patch
Size: 3907 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111218/2733d9f9/attachment.bin>

From remi at remlab.net  Sun Dec 18 11:39:42 2011
From: remi at remlab.net (=?utf-8?q?R=C3=A9mi?= Denis-Courmont)
Date: Sun, 18 Dec 2011 12:39:42 +0200
Subject: [vlc-devel] [PATCH 1/2] Contribs: update live555 to latest
	snapshot
In-Reply-To: <20111215191750.7449df23@tok>
References: <1323992761-22393-1-git-send-email-jb@videolan.org>
	<1323992761-22393-2-git-send-email-jb@videolan.org>
	<20111215191750.7449df23@tok>
Message-ID: <201112181239.42813.remi@remlab.net>

Le vendredi 16 d?cembre 2011 02:17:50 Rafa?l Carr?, vous avez ?crit :
> Le Fri, 16 Dec 2011 00:46:00 +0100,
> 
> Jean-Baptiste Kempf <jb at videolan.org> a ?crit :
> > ---
> > 
> >  contrib/src/live555/SHA512SUMS |    2 +-
> >  contrib/src/live555/rules.mak  |   11 +----------
> >  2 files changed, 2 insertions(+), 11 deletions(-)
> > 
> > diff --git a/contrib/src/live555/SHA512SUMS
> > b/contrib/src/live555/SHA512SUMS index da74cf1..9e416a4 100644
> > --- a/contrib/src/live555/SHA512SUMS
> > +++ b/contrib/src/live555/SHA512SUMS
> > @@ -1 +1 @@
> > -42e424e8743cdbce2d6c87b6ceeb182a8868e870a0b233df987004a99e97e80c216e6c99
> > 6774e936be40b1c3397a7c95ac8aa74eb6dcd8fac42a81dfcb4fd767 
> > live.2011.11.08.tar.gz
> > +00578686c1e154cea7638d19a6ab7675fa79d7da8f1848e9765954e3b02b82c54fd05e4
> > 480e7ffcdc399c7ad3a0ad26aff0a8ce97222e26e66e345fe82906e02 
> > live.2011.12.02.tar.gz diff --git a/contrib/src/live555/rules.mak
> > b/contrib/src/live555/rules.mak index a267981..f5bb7c1 100644
> > --- a/contrib/src/live555/rules.mak
> > +++ b/contrib/src/live555/rules.mak
> > @@ -1,7 +1,7 @@
> > 
> >  # live555
> >  
> >  #LIVEDOTCOM_URL :=
> >  http://live555.com/liveMedia/public/live555-latest.tar.gz
> > 
> > -LIVE555_FILE := live.2011.11.08.tar.gz
> > +LIVE555_FILE := live.2011.12.02.tar.gz
> > 
> >  LIVEDOTCOM_URL :=
> >  http://live555sourcecontrol.googlecode.com/files/$(LIVE555_FILE)
> >  
> >  PKGS += live555
> > 
> > @@ -15,15 +15,6 @@ live555: $(LIVE555_FILE) .sum-live555
> > 
> >  	rm -Rf live
> >  	$(UNPACK)
> >  	chmod -R u+w live
> > 
> > -	patch -p0 < $(SRC)/live555/live-uselocale.patch
> > -	patch -p0 < $(SRC)/live555/live-inet_ntop.patch
> > -	patch -p0 < $(SRC)/live555/live-intptr.patch
> > -ifndef HAVE_WIN32
> > -ifndef HAVE_WINCE
> > -	patch -p0 < $(SRC)/live555/live-getaddrinfo.patch
> > -endif
> > -endif
> > -	patch -p0 < $(SRC)/live555/live-cloexec.patch
> 
> don't forget to git rm those patches when you'll commit

The uselocale patch is still needed on Linux (but probably not on MacOS) as 
far as I can tell. So the next configure.ac patch is wrong.

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From martin at martin.st  Sun Dec 18 11:43:24 2011
From: martin at martin.st (=?ISO-8859-15?Q?Martin_Storsj=F6?=)
Date: Sun, 18 Dec 2011 12:43:24 +0200 (EET)
Subject: [vlc-devel] [PATCH 1/2] Build the iomx glue code as a separate
 shared library
In-Reply-To: <4EED7796.7000204@alexeysokolov.co.cc>
References: <1324072028-6485-1-git-send-email-martin@martin.st>
	<1324072028-6485-7-git-send-email-martin@martin.st>
	<4EED76C5.4090406@alexeysokolov.co.cc>
	<4EED7796.7000204@alexeysokolov.co.cc>
Message-ID: <alpine.DEB.2.00.1112181239080.10431@cone.martin.st>

Hi,

On Sun, 18 Dec 2011, Alexey Sokolov wrote:

> 17.12.2011 04:47, Martin Storsj? ?????:
> 
> diff --git a/vlc-android/src/org/videolan/vlc/android/LibVLC.java b/vlc-andr
> oid/src/org/videolan/vlc/android/LibVLC.java
> index 8cd436e..8f9b67a 100644
> --- a/vlc-android/src/org/videolan/vlc/android/LibVLC.java
> +++ b/vlc-android/src/org/videolan/vlc/android/LibVLC.java
> @@ -30,6 +30,11 @@ public class LibVLC {
>      /* Load library before object instantiation */
>      static {
>          try {
> +            System.loadLibrary("iomx");
> +        } catch (Throwable t) {
> +            t.printStackTrace();
> 
> Why t.printStackTrace() instead of Log.e()?

No particular reason, I can change it to a Log.e().

> Uhm, and why catching Throwable instead of only errors producable by
> System.loadLibrary()?

This also catches any RuntimeError exception. Since this is a library that 
does things outside of the public API/ABI, I tend to be extra careful. If 
this fails to load for any reason, the rest of the normal libvlc will 
still work fine, so any error from this should be caught, not just the 
normal ones. Not that I really know if any such exception really can be 
thrown here anyway.

// Martin

From remi at remlab.net  Sun Dec 18 13:14:40 2011
From: remi at remlab.net (=?iso-8859-1?q?R=E9mi?= Denis-Courmont)
Date: Sun, 18 Dec 2011 14:14:40 +0200
Subject: [vlc-devel] VLC Support for RTCP extensions anb playback buffer
	management
In-Reply-To: <20111216194119.21193386eadn5qkv@wm.upv.es>
References: <20111216194119.21193386eadn5qkv@wm.upv.es>
Message-ID: <201112181414.41436.remi@remlab.net>

	Hello,

Le vendredi 16 d?cembre 2011 20:41:19 Mario Montagud Climent, vous avez ?crit 
:
> I know that the ?netsync? module in VLC can be used to keep several
> clients synchronised on a single VLC stream (using UDP datagrams, not
> using RTP/RTCP). It can also be used to synchronise several computers
> playing parts of a video wall.

The netsync module can be used for any combination of media sources, so long 
as all sources have consistent time stamps on all nodes (master and slaves).

It was mainly designed to synchronize playback of identical local files, but 
it can be used with multicast RTP too.

> Our goal is to synchronize a RTP network stream across multiple
> separated clients. We use for that purpose the global time instant
> (64-bit wall clock time, eg. NTP) at which a specific RTP packet has
> been received (or played out) by each sync client.
> 
> For that purpose, first, I need to implement the new RTCP packet types
> and block reports defined in [Bra11]. I think it should be done in
> live555 library.

RTCP SR is currently done in live555, yes.

> I think the most difficult task will be the management of the playout
> processes in each media client to get in sync (e.g. to report on
> presentation time for a specific RTP packet ?e.g. the last one
> encapsulating a video frame-, and to manage the necessary playback
> adjustments (playout speed variation? Skips/pauses?).

On a typical media player, the playback speed is typically determined by the 
audio DAC buffer speed. Alternatively, it could be determined by the vertical 
refresh of the display. There is generally no way to tune the speed of the DAC 
or the display, so this cannot work for synchronized playback and/or for a 
live media source.

Thus VLC always synchronizes based the media source/input, not the 
sinks/outputs. If the media is not intrinsically paced, the local RTC is used 
as fallback (this would not the case with RTP though). Either way, this can 
always drift from the audio output sample rate and the display refresh rate, 
so VLC resamples on-the-fly to keep everything in synchronization.

> Do you think VLC is a good choice for implementing our RTP/RTCP-based
> IDMS solution? Can I access to the playback buffer to know about
> presentation times of a specific video frame and the current playback
> buffer occupancy?

There is an input clock, that gets driven by the input as outlined above, 
and/or by the netsync module. Its job is to correlate the input playback 
timestamps to the local RTC timestamps.

Audio samples are pushed to the hardware as early as possible to minimize 
underruns and maximize energy efficiency. The hardware audio device driver 
will provide a latency estimate for audio playback (latency may be longer than 
buffer time). Then the VLC audio plugin will upsample or downsample if latency 
is higher or lower than expected, so that it keeps up with the input intended 
speed. There is no feedback from the output to the input. It is all done 
within the audio output subsystem.

Similarly, the video output subsystem will drop frames if it notices that the 
computer is unable to render all of them fast enough. 

> Could the "netsync" module be adapted for our IDMS purposes?

I am not very familiar with netsync but there is no real answer to that 
question. Coding will be needed anyway.

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From finlayson at live555.com  Sun Dec 18 15:44:12 2011
From: finlayson at live555.com (Ross Finlayson)
Date: Sun, 18 Dec 2011 06:44:12 -0800
Subject: [vlc-devel] [PATCH 1/2] Contribs: update live555 to latest
	snapshot
In-Reply-To: <201112181239.42813.remi@remlab.net>
References: <1323992761-22393-1-git-send-email-jb@videolan.org>
	<1323992761-22393-2-git-send-email-jb@videolan.org>
	<20111215191750.7449df23@tok> <201112181239.42813.remi@remlab.net>
Message-ID: <EDEF108C-C9C6-489C-BDEB-F707F7D78AE9@live555.com>

> The uselocale patch is still needed on Linux (but probably not on MacOS) as 
> far as I can tell.

I thought that I had updated the LIVE555 libraries so that (in recent versions) none of the "live.*" patches are needed anymore.  That should be true for the 'uselocale' patch as well (including on Linux).  Why do you think otherwise?


Ross Finlayson
Live Networks, Inc.
http://www.live555.com/

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111218/9f269372/attachment.html>

From remi at remlab.net  Sun Dec 18 16:04:23 2011
From: remi at remlab.net (=?iso-8859-15?q?R=E9mi?= Denis-Courmont)
Date: Sun, 18 Dec 2011 17:04:23 +0200
Subject: [vlc-devel] [PATCH 1/2] Contribs: update live555 to latest
	snapshot
In-Reply-To: <EDEF108C-C9C6-489C-BDEB-F707F7D78AE9@live555.com>
References: <1323992761-22393-1-git-send-email-jb@videolan.org>
	<201112181239.42813.remi@remlab.net>
	<EDEF108C-C9C6-489C-BDEB-F707F7D78AE9@live555.com>
Message-ID: <201112181704.24485.remi@remlab.net>

Le dimanche 18 d?cembre 2011 16:44:12 Ross Finlayson, vous avez ?crit :
> > The uselocale patch is still needed on Linux (but probably not on MacOS)
> > as far as I can tell.
> 
> I thought that I had updated the LIVE555 libraries so that (in recent
> versions) none of the "live.*" patches are needed anymore.  That should be
> true for the 'uselocale' patch as well (including on Linux).  Why do you
> think otherwise?

As a matter of _fact_, the Linux build still uses setlocale().

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From cheng.sun at ymail.com  Mon Dec 19 00:18:58 2011
From: cheng.sun at ymail.com (Cheng Sun)
Date: Sun, 18 Dec 2011 23:18:58 +0000
Subject: [vlc-devel] [PATCH] Fix mono audio filter condition check
Message-ID: <4EEE74E2.3090605@ymail.com>

Patch attached fixes an obvious mistake in the condition check of the
mono audio filter.

-- Cheng
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-Fix-mono-audio-filter-condition-check.patch
Type: text/x-patch
Size: 1667 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111218/20a28e78/attachment.bin>

From linkfanel at yahoo.fr  Mon Dec 19 00:36:07 2011
From: linkfanel at yahoo.fr (Pierre Ynard)
Date: Mon, 19 Dec 2011 00:36:07 +0100
Subject: [vlc-devel] [PATCH] sap: handle SAP_Add() errors
Message-ID: <20111218233607.GA25347@via.ecp.fr>

Fixes #5640


diff --git a/src/stream_output/announce.c b/src/stream_output/announce.c
index 95dd81a..e7f39e7 100644
--- a/src/stream_output/announce.c
+++ b/src/stream_output/announce.c
@@ -92,7 +92,14 @@ sout_AnnounceRegisterSDP( vlc_object_t *obj, const char *psz_sdp,
         goto error;
 
     msg_Dbg (obj, "adding SAP session");
-    SAP_Add (p_sap, p_session );
+    if (SAP_Add (p_sap, p_session))
+    {
+        vlc_mutex_lock (&sap_mutex);
+        vlc_object_release ((vlc_object_t *)p_sap);
+        vlc_mutex_unlock (&sap_mutex);
+        goto error;
+    }
+
     return p_session;
 
 error:


Regards,

-- 
Pierre Ynard
"Une ?me dans un corps, c'est comme un dessin sur une feuille de papier."


From linkfanel at yahoo.fr  Mon Dec 19 02:10:57 2011
From: linkfanel at yahoo.fr (Pierre Ynard)
Date: Mon, 19 Dec 2011 02:10:57 +0100
Subject: [vlc-devel] [vlc-commits] VLM: use custom reference count
 instead of object destructor
In-Reply-To: <20110705151004.8F24413C49B@albiero.videolan.org>
Message-ID: <20111219011057.GA26106@via.ecp.fr>

> vlc | branch: master | R?mi Denis-Courmont <remi at remlab.net> | Tue
> Jul  5 18:09:04 2011 +0300| [71adf301e50aef185ef903d542896071f78323ff] |
> committer: R?mi Denis-Courmont
> 
> VLM: use custom reference count instead of object destructor
> 
> This should fix a crash whereby VLM is deleted before its child
> VOD server object.

> +    if( --p_vlm->users == 0 )
> +    {
> +        assert( libvlc_priv(p_vlm->p_libvlc)->p_vlm = p_vlm );
> +        libvlc_priv(p_vlm->p_libvlc)->p_vlm = NULL;
> +    }

I think this triggers the assertion in InputEvent(), because it is
called during the media clean-up which is now done after setting p_vlm
to NULL.

-- 
Pierre Ynard
"Une ?me dans un corps, c'est comme un dessin sur une feuille de papier."


From jb at videolan.org  Mon Dec 19 02:11:09 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Mon, 19 Dec 2011 02:11:09 +0100
Subject: [vlc-devel] [PATCH] Fix QToolButtonExt edge cases
In-Reply-To: <4EEDC26A.6030200@ymail.com>
References: <4EED0B9D.6030703@ymail.com> <20111218005139.GA13403@videolan.org>
	<4EEDC26A.6030200@ymail.com>
Message-ID: <20111219011109.GA30389@videolan.org>

On Sun, Dec 18, 2011 at 10:37:30AM +0000, Cheng Sun wrote :
> On 18/12/11 00:51, Jean-Baptiste Kempf wrote:
> > On Sat, Dec 17, 2011 at 09:37:33PM +0000, Cheng Sun wrote :
> >> Firstly, once the user presses down on a QToolButtonExt there is no
> >> way to cancel the action; either a short or a long click will be
> >> generated. Compare this to a normal button, which can be cancelled by
> >> releasing the mouse outside of the button area.
> > Agreed.
> >
> >> Secondly, with the mouse button held down, moving the mouse in and out
> >> of the button area will generate multiple short click events, when no
> >> event at all is desired.
> > Agreed too.
> >
> >> -    bool longClick;
> >> +    bool releasedIsDown;
> >> +    bool ignoreNextClick;
> > Very badly chosen variables, if I could say.
> >
> True.
> 
> Attached revised patch which is slightly more verbose (and
> understandable, hopefully)

Applied.


-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Mon Dec 19 02:44:48 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Mon, 19 Dec 2011 02:44:48 +0100
Subject: [vlc-devel] [PATCH 0/5] android/iomx as a separate shared
 library
In-Reply-To: <1324072028-6485-1-git-send-email-martin@martin.st>
References: <1324072028-6485-1-git-send-email-martin@martin.st>
Message-ID: <20111219014448.GA5307@videolan.org>

On Fri, Dec 16, 2011 at 11:47:01PM +0200, Martin Storsj? wrote :
> Martin Storsj? (5):
>   omxil: Make the iomx wrapper not depend on the omxil utils.c file
>   omxil: Build the iomx wrapper code as a separate shared library
>   omxil: Remove the header check for the iomx module
>   android: Don't add private header include directories when building
>     the full VLC tree
>   omxil: Use a fake library name when loading IOMX

Applied.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Mon Dec 19 02:45:09 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Mon, 19 Dec 2011 02:45:09 +0100
Subject: [vlc-devel] [PATCH] Fix stack overflow in ExecuteCommand
In-Reply-To: <20111217110938.28a79c2c@tok>
References: <4EEA2B02.1030201@gmail.com> <201112151938.26981.remi@remlab.net>
	<4EEA34FC.2080103@ymail.com> <4EEA36F3.2020100@ymail.com>
	<4EEA39D0.7090106@ymail.com> <20111216223627.1c1d0b43@tok>
	<4EEC73B5.2070806@ymail.com> <20111217110938.28a79c2c@tok>
Message-ID: <20111219014509.GB5307@videolan.org>

On Sat, Dec 17, 2011 at 11:09:38AM -0500, Rafa?l Carr? wrote :
> > Revised patch attached.
> 
> looks ok

Applied.


-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Mon Dec 19 02:48:34 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Mon, 19 Dec 2011 02:48:34 +0100
Subject: [vlc-devel] Patch: fix  a tag name in MediaServer xml parsing
In-Reply-To: <201112172300.47189.chris2553@googlemail.com>
References: <201112172300.47189.chris2553@googlemail.com>
Message-ID: <20111219014834.GC5307@videolan.org>

On Sat, Dec 17, 2011 at 11:00:47PM +0000, Chris Clayton wrote :
> According to http://upnp.org/specs/av/UPnP-av-MediaServer-v1-Device.pdf, the tag name of the element 
> that provides the base URL for relative URLs is "URLBase".  The same is true for a v2 MediaServer. 
> Make it so in MediaServer:: parseDeviceDescription(). The patch is against the 20111216 snapshot.

Applied.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Mon Dec 19 03:04:04 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Mon, 19 Dec 2011 03:04:04 +0100
Subject: [vlc-devel] [PATCH] Add filename sort option to playlist
 dialogue
In-Reply-To: <4EEC1F81.6090705@compdigitec.com>
References: <4EE945DF.2020504@compdigitec.com>
	<4EEC1F81.6090705@compdigitec.com>
Message-ID: <20111219020404.GA7172@videolan.org>

On Fri, Dec 16, 2011 at 11:50:09PM -0500, Edward Wang wrote :
> The attached patch will add a requested feature - the option to sort
> by filename (in addition to the URI) in the playlist dialogue (see
> http://trac.videolan.org/vlc/ticket/5455).

Seems fine for me. But I am not sure... Anyone?

You should send patches with git send-email and not reply to an older
thread, btw.

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From mirsal at videolan.org  Mon Dec 19 04:30:34 2011
From: mirsal at videolan.org (Mirsal Ennaime)
Date: Mon, 19 Dec 2011 04:30:34 +0100
Subject: [vlc-devel] Patch: compatibility with upnp Mediaserver v2
In-Reply-To: <201112172330.08252.chris2553@googlemail.com>
References: <201112172330.08252.chris2553@googlemail.com>
Message-ID: <1324265434.24865.66.camel@mirsal-laptop1.mirsal.fr>

Hello Chris,

On Sat, 2011-12-17 at 23:30 +0000, Chris Clayton wrote:
> I own a Panasonic Blu-Ray recorder (model DMR-BW880), which also acts as DLNA Mediaserver for video 
> or photos stored on its hard drive. The Mediaserver announces itself as being version 2, so the 
> content directory service is also version 2.
> 
> The patch below (against the 20111216 snapshot) allows VLC to play content from the server. The 
> Panasonic is the only mediaserver v2 device that I have, so I can't test whether the patch will 
> work with other devices. Perhaps owners of other v2 mediaservers could try it out. At the least, my 
> patch can be the basis for development by someone who knows more about upnp/dlna.

Applied with minor editing (whitespace, commit log), thanks.

Best regards,

-- 
mirsal





From remi at remlab.net  Mon Dec 19 08:07:14 2011
From: remi at remlab.net (=?UTF-8?Q?R=C3=A9mi_Denis-Courmont?=)
Date: Mon, 19 Dec 2011 08:07:14 +0100
Subject: [vlc-devel] [vlc-commits] configure: sort of fix speexdsp
In-Reply-To: <20111219023150.C7AFC140DB9@albiero.videolan.org>
References: <20111219023150.C7AFC140DB9@albiero.videolan.org>
Message-ID: <0202c590dac6528dd9ce447fef3f0ecc@chewa.net>

On Mon, 19 Dec 2011 03:31:50 +0100 (CET), git at videolan.org (Pierre Ynard)
wrote:
> vlc | branch: master | Pierre Ynard <linkfanel at yahoo.fr> | Mon Dec 19
> 03:30:20 2011 +0100| [c618458095cf74610053e9fe53e0a12ca7fc81c6] |
> committer: Pierre Ynard
> 
> configure: sort of fix speexdsp
> 
> $enable_speex can't be empty at this point so this check makes no sense

I think the bug is somewhere. And besides, VLC really needs a resampler so
failing when speexdsp is absent seems fine to me.

-- 
R?mi Denis-Courmont
http://www.remlab.net/


From martin at martin.st  Mon Dec 19 15:10:41 2011
From: martin at martin.st (=?UTF-8?q?Martin=20Storsj=C3=B6?=)
Date: Mon, 19 Dec 2011 16:10:41 +0200
Subject: [vlc-devel] [PATCH] androidsurface: Update the surfaceflinger video
	output for ICS
Message-ID: <1324303841-18493-1-git-send-email-martin@martin.st>

---
Tested on both the emulator and on an early cyanogenmod alpha rom
for nexus one.

 modules/video_output/androidsurface.c |   18 +++++++++++++++---
 1 files changed, 15 insertions(+), 3 deletions(-)

diff --git a/modules/video_output/androidsurface.c b/modules/video_output/androidsurface.c
index 0ea858f..af2cdc1 100644
--- a/modules/video_output/androidsurface.c
+++ b/modules/video_output/androidsurface.c
@@ -36,6 +36,9 @@
 #ifndef ANDROID_SYM_S_LOCK
 # define ANDROID_SYM_S_LOCK "_ZN7android7Surface4lockEPNS0_11SurfaceInfoEb"
 #endif
+#ifndef ANDROID_SYM_S_LOCK2
+# define ANDROID_SYM_S_LOCK2 "_ZN7android7Surface4lockEPNS0_11SurfaceInfoEPNS_6RegionE"
+#endif
 #ifndef ANDROID_SYM_S_UNLOCK
 # define ANDROID_SYM_S_UNLOCK "_ZN7android7Surface13unlockAndPostEv"
 #endif
@@ -67,6 +70,8 @@ extern void  jni_SetAndroidSurfaceSize(int width, int height);
 
 // _ZN7android7Surface4lockEPNS0_11SurfaceInfoEb
 typedef void (*Surface_lock)(void *, void *, int);
+// _ZN7android7Surface4lockEPNS0_11SurfaceInfoEPNS_6RegionE
+typedef void (*Surface_lock2)(void *, void *, void *);
 // _ZN7android7Surface13unlockAndPostEv
 typedef void (*Surface_unlockAndPost)(void *);
 
@@ -83,6 +88,7 @@ struct vout_display_sys_t {
     picture_pool_t *pool;
     void *p_library;
     Surface_lock s_lock;
+    Surface_lock2 s_lock2;
     Surface_unlockAndPost s_unlockAndPost;
 
     picture_resource_t resource;
@@ -115,9 +121,10 @@ static inline void *LoadSurface(const char *psz_lib, vout_display_sys_t *sys) {
     void *p_library = dlopen(psz_lib, RTLD_NOW);
     if (p_library) {
         sys->s_lock = (Surface_lock)(dlsym(p_library, ANDROID_SYM_S_LOCK));
+        sys->s_lock2 = (Surface_lock2)(dlsym(p_library, ANDROID_SYM_S_LOCK2));
         sys->s_unlockAndPost =
             (Surface_unlockAndPost)(dlsym(p_library, ANDROID_SYM_S_UNLOCK));
-        if (sys->s_lock && sys->s_unlockAndPost) {
+        if ((sys->s_lock || sys->s_lock2) && sys->s_unlockAndPost) {
             return p_library;
         }
         dlclose(p_library);
@@ -129,6 +136,8 @@ static void *InitLibrary(vout_display_sys_t *sys) {
     void *p_library;
     if ((p_library = LoadSurface("libsurfaceflinger_client.so", sys)))
         return p_library;
+    if ((p_library = LoadSurface("libgui.so", sys)))
+        return p_library;
     return LoadSurface("libui.so", sys);
 }
 
@@ -154,7 +163,7 @@ static int Open(vlc_object_t *p_this) {
     sys->p_library = p_library = InitLibrary(sys);
     if (!p_library) {
         free(sys);
-        msg_Err(vd, "Could not initialize libui.so/libsurfaceflinger_client.so!");
+        msg_Err(vd, "Could not initialize libui.so/libgui.so/libsurfaceflinger_client.so!");
         vlc_mutex_unlock(&single_instance);
         return VLC_EGENERIC;
     }
@@ -253,7 +262,10 @@ static int  AndroidLockSurface(picture_t *picture) {
         return VLC_EGENERIC;
     }
 
-    sys->s_lock(surf, info, 1);
+    if (sys->s_lock)
+        sys->s_lock(surf, info, 1);
+    else
+        sys->s_lock2(surf, info, NULL);
 
     // input size doesn't match the surface size,
     // request a resize
-- 
1.7.2.5



From fatbull at web.de  Mon Dec 19 15:47:44 2011
From: fatbull at web.de (=?ISO-8859-1?Q?Tobias_G=FCntner?=)
Date: Mon, 19 Dec 2011 15:47:44 +0100
Subject: [vlc-devel] [PATCH] Fix assertions
Message-ID: <4EEF4E90.8050909@web.de>

Hello!

I assume == was intended.

Regards,
Tobias
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: 0001-Fix-assertions.patch
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111219/2be9cf3a/attachment.ksh>

From martin at martin.st  Mon Dec 19 16:02:59 2011
From: martin at martin.st (=?UTF-8?q?Martin=20Storsj=C3=B6?=)
Date: Mon, 19 Dec 2011 17:02:59 +0200
Subject: [vlc-devel] [PATCH 1/6] omxil: Ignore nSliceHeight on Galaxy S II
Message-ID: <1324306984-2576-1-git-send-email-martin@martin.st>

---
 modules/codec/omxil/omxil.c |   12 ++++++++++++
 1 files changed, 12 insertions(+), 0 deletions(-)

diff --git a/modules/codec/omxil/omxil.c b/modules/codec/omxil/omxil.c
index dde8126..d12d09c 100644
--- a/modules/codec/omxil/omxil.c
+++ b/modules/codec/omxil/omxil.c
@@ -445,6 +445,18 @@ static OMX_ERRORTYPE GetPortDefinition(decoder_t *p_dec, OmxPort *p_port,
                     strlen("OMX.qcom.video.decoder")))
             def->format.video.eColorFormat = OMX_QCOM_COLOR_FormatYVU420SemiPlanar;
 
+        /* Hack: Galaxy S II (stock firmware) gives a slice height larger
+         * than the video height, but this doesn't imply padding between
+         * the video planes. Nexus S also has a slice height larger than
+         * the video height, but there it actually is real padding, thus
+         * Galaxy S II is the buggy one. The Galaxy S II decoder is
+         * named OMX.SEC.avcdec while the one on Nexus S is
+         * OMX.SEC.AVC.Decoder. Thus do this for any OMX.SEC. that don't
+         * contain the string ".Decoder". */
+        if(!strncmp(p_sys->psz_component, "OMX.SEC.", strlen("OMX.SEC.")) &&
+           !strstr(p_sys->psz_component, ".Decoder"))
+            def->format.video.nSliceHeight = 0;
+
         if(!GetVlcVideoFormat( def->format.video.eCompressionFormat,
                                &p_fmt->i_codec, 0 ) )
         {
-- 
1.7.2.5



From martin at martin.st  Mon Dec 19 16:03:00 2011
From: martin at martin.st (=?UTF-8?q?Martin=20Storsj=C3=B6?=)
Date: Mon, 19 Dec 2011 17:03:00 +0200
Subject: [vlc-devel] [PATCH 2/6] omxil: Try to enable YUV420p mode on
	Samsung Galaxy S II
In-Reply-To: <1324306984-2576-1-git-send-email-martin@martin.st>
References: <1324306984-2576-1-git-send-email-martin@martin.st>
Message-ID: <1324306984-2576-2-git-send-email-martin@martin.st>

This is similar to the ThumbnailMode on Nexus S, except that
the index doesn't seem to be queryable dynamically
(I don't seem to find any string in the binaries that
would be the right one at least).
---
 modules/codec/omxil/omxil.c       |    8 ++++++++
 modules/codec/omxil/omxil_utils.h |    2 ++
 2 files changed, 10 insertions(+), 0 deletions(-)

diff --git a/modules/codec/omxil/omxil.c b/modules/codec/omxil/omxil.c
index d12d09c..56a0c52 100644
--- a/modules/codec/omxil/omxil.c
+++ b/modules/codec/omxil/omxil.c
@@ -692,6 +692,14 @@ static OMX_ERRORTYPE InitialiseComponent(decoder_t *p_dec,
             OMX_BOOL enable = OMX_TRUE;
             omx_error = OMX_SetConfig(omx_handle, index, &enable);
             CHECK_ERROR(omx_error, "Unable to set ThumbnailMode");
+        } else {
+            OMX_BOOL enable = OMX_TRUE;
+            /* Needed on Samsung Galaxy S II */
+            omx_error = OMX_SetConfig(omx_handle, OMX_IndexVendorSetYUV420pMode, &enable);
+            if (omx_error == OMX_ErrorNone)
+                msg_Dbg(p_dec, "Set OMX_IndexVendorSetYUV420pMode successfully");
+            else
+                msg_Dbg(p_dec, "Unable to set OMX_IndexVendorSetYUV420pMode: %x", omx_error);
         }
     }
 
diff --git a/modules/codec/omxil/omxil_utils.h b/modules/codec/omxil/omxil_utils.h
index 4294170..f5bc2b1 100644
--- a/modules/codec/omxil/omxil_utils.h
+++ b/modules/codec/omxil/omxil_utils.h
@@ -185,3 +185,5 @@ unsigned int GetAudioParamSize(OMX_INDEXTYPE index);
  * Vendor specific color formats
  *****************************************************************************/
 #define OMX_QCOM_COLOR_FormatYVU420SemiPlanar 0x7FA30C00
+
+#define OMX_IndexVendorSetYUV420pMode 0x7f000003
-- 
1.7.2.5



From martin at martin.st  Mon Dec 19 16:03:01 2011
From: martin at martin.st (=?UTF-8?q?Martin=20Storsj=C3=B6?=)
Date: Mon, 19 Dec 2011 17:03:01 +0200
Subject: [vlc-devel] [PATCH 3/6] omxil: Fix a sanity check that only is
	relevant for video streams
In-Reply-To: <1324306984-2576-1-git-send-email-martin@martin.st>
References: <1324306984-2576-1-git-send-email-martin@martin.st>
Message-ID: <1324306984-2576-3-git-send-email-martin@martin.st>

---
 modules/codec/omxil/omxil.c |    6 +++---
 1 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/modules/codec/omxil/omxil.c b/modules/codec/omxil/omxil.c
index 56a0c52..763a35e 100644
--- a/modules/codec/omxil/omxil.c
+++ b/modules/codec/omxil/omxil.c
@@ -1035,9 +1035,9 @@ static OMX_ERRORTYPE PortReconfigure(decoder_t *p_dec, OmxPort *p_port)
     definition.nPortIndex = p_port->i_port_index;
     omx_error = OMX_GetParameter(p_dec->p_sys->omx_handle, OMX_IndexParamPortDefinition,
                                  &definition);
-    if(omx_error != OMX_ErrorNone ||
-       !definition.format.video.nFrameWidth ||
-       !definition.format.video.nFrameHeight )
+    if(omx_error != OMX_ErrorNone || (p_dec->fmt_in.i_cat == VIDEO_ES &&
+       (!definition.format.video.nFrameWidth ||
+       !definition.format.video.nFrameHeight)) )
         return OMX_ErrorUndefined;
 
     omx_error = OMX_SendCommand( p_sys->omx_handle, OMX_CommandPortDisable,
-- 
1.7.2.5



From martin at martin.st  Mon Dec 19 16:03:03 2011
From: martin at martin.st (=?UTF-8?q?Martin=20Storsj=C3=B6?=)
Date: Mon, 19 Dec 2011 17:03:03 +0200
Subject: [vlc-devel] [PATCH 5/6] omxil: Handle sentinel buffers in
	DecodeAudio, too
In-Reply-To: <1324306984-2576-1-git-send-email-martin@martin.st>
References: <1324306984-2576-1-git-send-email-martin@martin.st>
Message-ID: <1324306984-2576-5-git-send-email-martin@martin.st>

These buffers were introduced to avoid a potential deadlock
when the codec has issued a PortSettingsChanged event but
not returned any buffers to the client, making the DecodeVideo
function hang waiting for an input buffer to use, before
handling the reconfigure. The same handling needs to be done
for DecodeAudio too, otherwise it will try to use the
sentinel buffer as a normal one.
---
 modules/codec/omxil/omxil.c |    5 +++++
 1 files changed, 5 insertions(+), 0 deletions(-)

diff --git a/modules/codec/omxil/omxil.c b/modules/codec/omxil/omxil.c
index 92f2be7..7be3ec6 100644
--- a/modules/codec/omxil/omxil.c
+++ b/modules/codec/omxil/omxil.c
@@ -1372,6 +1372,10 @@ aout_buffer_t *DecodeAudio ( decoder_t *p_dec, block_t **pp_block )
 
     /* Send the input buffer to the component */
     OMX_FIFO_GET(&p_sys->in.fifo, p_header);
+
+    if (p_header && p_header->nFlags & OMX_BUFFERFLAG_EOS)
+        goto reconfig;
+
     if(p_header)
     {
         p_header->nFilledLen = p_block->i_buffer;
@@ -1408,6 +1412,7 @@ aout_buffer_t *DecodeAudio ( decoder_t *p_dec, block_t **pp_block )
         *pp_block = NULL; /* Avoid being fed the same packet again */
     }
 
+reconfig:
     /* Handle the PortSettingsChanged events */
     for(i = 0; i < p_sys->ports; i++)
     {
-- 
1.7.2.5



From martin at martin.st  Mon Dec 19 16:03:02 2011
From: martin at martin.st (=?UTF-8?q?Martin=20Storsj=C3=B6?=)
Date: Mon, 19 Dec 2011 17:03:02 +0200
Subject: [vlc-devel] [PATCH 4/6] omxil: Avoid a SetParameters call that
	fails for audio codecs on Galaxy S II
In-Reply-To: <1324306984-2576-1-git-send-email-martin@martin.st>
References: <1324306984-2576-1-git-send-email-martin@martin.st>
Message-ID: <1324306984-2576-4-git-send-email-martin@martin.st>

This SetParameters call shouldn't be necessary as far as I know,
but only avoiding it for audio codecs for now.
---
 modules/codec/omxil/omxil.c |   23 +++++++++++++++++++----
 1 files changed, 19 insertions(+), 4 deletions(-)

diff --git a/modules/codec/omxil/omxil.c b/modules/codec/omxil/omxil.c
index 763a35e..92f2be7 100644
--- a/modules/codec/omxil/omxil.c
+++ b/modules/codec/omxil/omxil.c
@@ -1062,10 +1062,25 @@ static OMX_ERRORTYPE PortReconfigure(decoder_t *p_dec, OmxPort *p_port)
     /* Get the new port definition */
     omx_error = GetPortDefinition(p_dec, &p_sys->out, p_sys->out.p_fmt);
     if(omx_error != OMX_ErrorNone) goto error;
-    omx_error = OMX_SetParameter(p_dec->p_sys->omx_handle, OMX_IndexParamPortDefinition,
-                                 &definition);
-    CHECK_ERROR(omx_error, "OMX_SetParameter failed (%x : %s)",
-                omx_error, ErrorToString(omx_error));
+
+    if( p_dec->fmt_in.i_cat != AUDIO_ES )
+    {
+        /* Don't explicitly set the new parameters that we got with
+         * OMX_GetParameter above when using audio codecs.
+         * That struct hasn't been changed since, so there should be
+         * no need to set it here, unless some codec expects the
+         * SetParameter call as a trigger event for some part of
+         * the reconfiguration.
+         * This fixes using audio decoders on Samsung Galaxy S II,
+         *
+         * Only skipping this for audio codecs, to minimize the
+         * change for current working configurations for video.
+         */
+        omx_error = OMX_SetParameter(p_dec->p_sys->omx_handle, OMX_IndexParamPortDefinition,
+                                     &definition);
+        CHECK_ERROR(omx_error, "OMX_SetParameter failed (%x : %s)",
+                    omx_error, ErrorToString(omx_error));
+    }
 
     omx_error = OMX_SendCommand( p_sys->omx_handle, OMX_CommandPortEnable,
                                  p_port->i_port_index, NULL);
-- 
1.7.2.5



From martin at martin.st  Mon Dec 19 16:03:04 2011
From: martin at martin.st (=?UTF-8?q?Martin=20Storsj=C3=B6?=)
Date: Mon, 19 Dec 2011 17:03:04 +0200
Subject: [vlc-devel] [PATCH 6/6] omxil: Don't free the sentinel buffers
In-Reply-To: <1324306984-2576-1-git-send-email-martin@martin.st>
References: <1324306984-2576-1-git-send-email-martin@martin.st>
Message-ID: <1324306984-2576-6-git-send-email-martin@martin.st>

These buffers aren't known to the actual OMX codec.

Normally, these buffers are removed from the queue by
DecodeAudio/DecodeVideo, but if they're pushed multiple
times (which happens when using google sw decoders),
PortReconfigure might try to free them.
---
 modules/codec/omxil/omxil.c |    4 ++++
 1 files changed, 4 insertions(+), 0 deletions(-)

diff --git a/modules/codec/omxil/omxil.c b/modules/codec/omxil/omxil.c
index 7be3ec6..ef8982b 100644
--- a/modules/codec/omxil/omxil.c
+++ b/modules/codec/omxil/omxil.c
@@ -580,6 +580,8 @@ static OMX_ERRORTYPE DeinitialiseComponent(decoder_t *p_dec,
             for(j = 0; j < p_port->i_buffers; j++)
             {
                 OMX_FIFO_GET(&p_port->fifo, p_buffer);
+                if (p_buffer == &p_sys->sentinel_buffer)
+                    continue;
                 omx_error = OMX_FreeBuffer( omx_handle,
                                             p_port->i_port_index, p_buffer );
 
@@ -1048,6 +1050,8 @@ static OMX_ERRORTYPE PortReconfigure(decoder_t *p_dec, OmxPort *p_port)
     for(i = 0; i < p_port->i_buffers; i++)
     {
         OMX_FIFO_GET(&p_port->fifo, p_buffer);
+        if (p_buffer == &p_sys->sentinel_buffer)
+            continue;
         omx_error = OMX_FreeBuffer( p_sys->omx_handle,
                                     p_port->i_port_index, p_buffer );
 
-- 
1.7.2.5



From remi at remlab.net  Mon Dec 19 18:30:27 2011
From: remi at remlab.net (=?iso-8859-15?q?R=E9mi?= Denis-Courmont)
Date: Mon, 19 Dec 2011 19:30:27 +0200
Subject: [vlc-devel] [PATCH] Fix assertions
In-Reply-To: <4EEF4E90.8050909@web.de>
References: <4EEF4E90.8050909@web.de>
Message-ID: <201112191930.27986.remi@remlab.net>

Le lundi 19 d?cembre 2011 16:47:44 Tobias G?ntner, vous avez ?crit :
> I assume == was intended.

Merged, thanks.

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From remi at remlab.net  Mon Dec 19 18:35:08 2011
From: remi at remlab.net (=?iso-8859-15?q?R=E9mi?= Denis-Courmont)
Date: Mon, 19 Dec 2011 19:35:08 +0200
Subject: [vlc-devel] [PATCH] Fix mono audio filter condition check
In-Reply-To: <4EEE74E2.3090605@ymail.com>
References: <4EEE74E2.3090605@ymail.com>
Message-ID: <201112191935.09035.remi@remlab.net>

Le lundi 19 d?cembre 2011 01:18:58 Cheng Sun, vous avez ?crit :
> Patch attached fixes an obvious mistake in the condition check of the
> mono audio filter.

Merged, thanks.

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From beauze.h at gmail.com  Mon Dec 19 21:36:59 2011
From: beauze.h at gmail.com (=?UTF-8?Q?Hugo_Beauz=C3=A9e=2DLuyssen?=)
Date: Mon, 19 Dec 2011 21:36:59 +0100
Subject: [vlc-devel] [Patch] Dash: Don't try to do something without a
 valid MPD manager
In-Reply-To: <CAK-C75TtOYuwx2d=WQX4XUvqY6bjON-P=X2zbPFnig331usRXA@mail.gmail.com>
References: <CAK-C75TOrnUGWki6HDVk0iJ0_xE1ZO4Swd8FGAoCraupTXQW8Q@mail.gmail.com>
	<20111214180641.GA23803@videolan.org>
	<CAK-C75TtOYuwx2d=WQX4XUvqY6bjON-P=X2zbPFnig331usRXA@mail.gmail.com>
Message-ID: <CAK-C75QU-kdb8JyJd_rRF99kEzNJghyGYVwn7kEqdnAcfMFcZw@mail.gmail.com>

On Fri, Dec 16, 2011 at 4:42 PM, Hugo Beauz?e-Luyssen
<beauze.h at gmail.com> wrote:
> On Wed, Dec 14, 2011 at 7:06 PM, Jean-Baptiste Kempf <jb at videolan.org> wrote:
>> Why do we need a NullManager, in the end, since we have the empty
>> interface already?
>> What is the use of instantiating it, if it does not do anything?
>> If the profile is not implemented, we just quit the module, no?
>>
>> I must be missing something, but I don't see the point.
>>
>
> As far as I'm concerned, there's no much use, but it forces us to
> maintain the NullManager implementation, in case we edit the
> interface, which is a waste of time.
>
> Regards,
>

Hi,

Ping for reviews (or merge, actually, that would work to :) )!

Regards,

-- 
Hugo Beauz?e-Luyssen


From jb at videolan.org  Mon Dec 19 22:02:35 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Mon, 19 Dec 2011 22:02:35 +0100
Subject: [vlc-devel] [PATCH 1/6] omxil: Ignore nSliceHeight on Galaxy S
 II
In-Reply-To: <1324306984-2576-1-git-send-email-martin@martin.st>
References: <1324306984-2576-1-git-send-email-martin@martin.st>
Message-ID: <20111219210235.GA31890@videolan.org>

On Mon, Dec 19, 2011 at 05:02:59PM +0200, Martin Storsj? wrote :
>  modules/codec/omxil/omxil.c |   12 ++++++++++++

Applied, and all the thread.

Thanks, once again.
-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Mon Dec 19 22:02:49 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Mon, 19 Dec 2011 22:02:49 +0100
Subject: [vlc-devel] [PATCH] androidsurface: Update the surfaceflinger
 video output for ICS
In-Reply-To: <1324303841-18493-1-git-send-email-martin@martin.st>
References: <1324303841-18493-1-git-send-email-martin@martin.st>
Message-ID: <20111219210249.GB31890@videolan.org>

On Mon, Dec 19, 2011 at 04:10:41PM +0200, Martin Storsj? wrote :
> Tested on both the emulator and on an early cyanogenmod alpha rom
> for nexus one.

Applied.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Mon Dec 19 22:03:26 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Mon, 19 Dec 2011 22:03:26 +0100
Subject: [vlc-devel] [PATCH] sap: handle SAP_Add() errors
In-Reply-To: <20111218233607.GA25347@via.ecp.fr>
References: <20111218233607.GA25347@via.ecp.fr>
Message-ID: <20111219210326.GC31890@videolan.org>

On Mon, Dec 19, 2011 at 12:36:07AM +0100, Pierre Ynard wrote :
> Fixes #5640

Was applied. Automatic closing didn't work though...


-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Mon Dec 19 22:04:00 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Mon, 19 Dec 2011 22:04:00 +0100
Subject: [vlc-devel] Mobile web interface patch
In-Reply-To: <20111217033208.GA3087@videolan.org>
References: <3366F855-5D14-460F-A81E-B2094FEA5C9F@var1able.ru>
	<20111217033208.GA3087@videolan.org>
Message-ID: <20111219210400.GD31890@videolan.org>

On Sat, Dec 17, 2011 at 04:32:08AM +0100, Jean-Baptiste Kempf wrote :
> On Fri, Dec 16, 2011 at 06:00:45PM +0400, ?????????? ???????? wrote :
> > http://www.google-melange.com/gci/task/view/google/gci2011/7180436
> 
> Patch does not apply...

FYI, a fixed patch was applied.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From nkoriyama at gmail.com  Tue Dec 20 11:14:23 2011
From: nkoriyama at gmail.com (Naohiro KORIYAMA)
Date: Tue, 20 Dec 2011 19:14:23 +0900
Subject: [vlc-devel] [vlc-commits] upnp: Switch to an unlimited maximum
	content length
In-Reply-To: <20111220073910.B3409143F8C@albiero.videolan.org>
References: <20111220073910.B3409143F8C@albiero.videolan.org>
Message-ID: <CAOa_K+tD4rDt27V_ycZHiQVTQrC3GBMTz60AdFVV0o451fuosQ@mail.gmail.com>

2011/12/20 Mirsal Ennaime <git at videolan.org>:
> vlc | branch: master | Mirsal Ennaime <mirsal at videolan.org> | Tue Dec 20 08:34:45 2011 +0100| [794557eea63853456cf3120cdb1bdc88ca44ad9f] | committer: Mirsal Ennaime
>
> upnp: Switch to an unlimited maximum content length
>
> Use INT_MAX as the maximum content length as libupnp introduced 0 as an
> unlimited content length in pupnp 1.6.7, though UPNP_VERSION was not bumped.
>
>> http://git.videolan.org/gitweb.cgi/vlc.git/?a=commit;h=794557eea63853456cf3120cdb1bdc88ca44ad9f
> ---
>
> ?modules/services_discovery/upnp.cpp | ? 12 ++++++++++++
> ?1 files changed, 12 insertions(+), 0 deletions(-)
>
> diff --git a/modules/services_discovery/upnp.cpp b/modules/services_discovery/upnp.cpp
> index 075c9dd..84cf3c4 100644
> --- a/modules/services_discovery/upnp.cpp
> +++ b/modules/services_discovery/upnp.cpp
> @@ -136,6 +136,18 @@ static int Open( vlc_object_t *p_this )
> ? ? ? ? return VLC_EGENERIC;
> ? ? }
>
> + ? ?/* libupnp does not treat a maximum content length of 0 as unlimited
> + ? ? * until 64dedf (~ pupnp v1.6.7) and provides no sane way to discriminate
> + ? ? * between versions */
> + ? ?if( (i_res = UpnpSetMaxContentLength( INT_MAX )) != UPNP_E_SUCCESS )
> + ? ?{
> + ? ? ? ?msg_Err( p_sd, "Failed to set maximum content length: %s",
> + ? ? ? ? ? ? ? ?UpnpGetErrorMessage( i_res ));
> +
> + ? ? ? ?Close( (vlc_object_t*) p_sd );
> + ? ? ? ?return VLC_EGENERIC;
> + ? ?}
> +
> ? ? return VLC_SUCCESS;
> ?}
>

win32 build failed because  INT_MAX is not defined.
#include <climits>
or
#include <limits.h>
is required.


-- 
KORIYAMA, Naohiro
nkoriyama at gmail.com


From nkoriyama at gmail.com  Tue Dec 20 12:08:46 2011
From: nkoriyama at gmail.com (Naohiro KORIYAMA)
Date: Tue, 20 Dec 2011 20:08:46 +0900
Subject: [vlc-devel] [PATCH] yadif : Add SSSE3 and SSE2 support. porting
 from FFmpeg.
In-Reply-To: <CAOa_K+tBjZCS5sSve_vwCt4DbzX2qpJGa37tTiZkX-tKOaUQgA@mail.gmail.com>
References: <CAOa_K+vdRUzBj4cE9qoLyU19+NYnVuPx03UqdOEXgkT6fw60Gg@mail.gmail.com>
	<201112091941.15730.remi@remlab.net>
	<CAOa_K+shj1Wy4pm3=sFZxw4_nM+k-VCZnRauiQujeV0+LS4CGw@mail.gmail.com>
	<201112171042.02575.remi@remlab.net>
	<CAOa_K+tBjZCS5sSve_vwCt4DbzX2qpJGa37tTiZkX-tKOaUQgA@mail.gmail.com>
Message-ID: <CAOa_K+ug10ymVFtCeTTHo=TwzOzynfiON9WE-Ci4L_A_x45WzQ@mail.gmail.com>

2011/12/17 Naohiro KORIYAMA <nkoriyama at gmail.com>:
> 2011/12/17 R?mi Denis-Courmont <remi at remlab.net>:
>> Le samedi 10 d?cembre 2011 07:55:16 Naohiro KORIYAMA, vous avez ?crit :
>>> OK. I reverted that part. and used VLC_MMX and VLC_SSE.
>>> I confirmed the compilation issue is fixed!
>>
>> I cannot link it here...
>>
>> .libs/libdeinterlace_plugin_la-algo_yadif.o: In function
>> `yadif_filter_line_ssse3':
>> /home/remi/videolan/vlc/build/modules/video_filter/../../../modules/video_filter/deinterlace/yadif_template.h:249:
>> undefined reference to `_pw_1'
>> /home/remi/videolan/vlc/build/modules/video_filter/../../../modules/video_filter/deinterlace/yadif_template.h:249:
>> undefined reference to `_pb_1'
>>
>
> I didn't have Linux environment. So I just installed Ubuntu 64 bit,
> and made a building environment.
> And I just got same error.
>
> I attached a new patch, it fixed the compilation issue.
> I built and testedya the module on Linux 64bit, Mac OS X 64bit and
> WIndows 32bit.
>

With a patch, I sent 3 days ago, inline assembler part of code may not
be compiled on some environment.
Attached patch fixed the issue.

sorry for sending modification many times.

Best regards,



-- 
KORIYAMA, Naohiro
nkoriyama at gmail.com
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-yadif-Add-SSSE3-and-SSE2-support.-porting-from-FFmpe.patch
Type: application/octet-stream
Size: 30153 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111220/ec87c6d9/attachment.obj>

From komh78 at gmail.com  Tue Dec 20 12:29:44 2011
From: komh78 at gmail.com (KO Myung-Hun)
Date: Tue, 20 Dec 2011 20:29:44 +0900
Subject: [vlc-devel] OS/2 patches, season 5
Message-ID: <1324380587-225-1-git-send-email-komh@chollian.net>

Hi/2.

These are season 5 of OS/2 patches.

Review, please. ^_______^


From komh78 at gmail.com  Tue Dec 20 12:29:45 2011
From: komh78 at gmail.com (KO Myung-Hun)
Date: Tue, 20 Dec 2011 20:29:45 +0900
Subject: [vlc-devel] [PATCH 1/3] Workaround for OS/2 iconv().
In-Reply-To: <1324380587-225-1-git-send-email-komh@chollian.net>
References: <1324380587-225-1-git-send-email-komh@chollian.net>
Message-ID: <1324380587-225-2-git-send-email-komh@chollian.net>

    1. OS/2 supports UCS-2 only not UTF-16
    2. OS/2 does not treat a string as a pathname by default
---
 src/extras/libc.c |   44 ++++++++++++++++++++++++++++++++++++++++++++
 1 files changed, 44 insertions(+), 0 deletions(-)

diff --git a/src/extras/libc.c b/src/extras/libc.c
index 6df1a52..b3d6041 100644
--- a/src/extras/libc.c
+++ b/src/extras/libc.c
@@ -37,6 +37,15 @@
 #   include <iconv.h>
 #endif
 
+#if defined(__OS2__) && defined(__INNOTEK_LIBC__)
+#   include <uconv.h>
+
+typedef struct os2_iconv_t
+{
+    UconvObject from;
+} os2_iconv_t;
+#endif
+
 /*****************************************************************************
  * Local conversion routine from ISO_6937 to UTF-8 charset. Support for this
  * is still missing in libiconv, hence multiple operating systems lack it.
@@ -326,7 +335,42 @@ vlc_iconv_t vlc_iconv_open( const char *tocode, const char *fromcode )
         return (vlc_iconv_t)(-2);
 #endif
 #if defined(HAVE_ICONV)
+# if defined(__OS2__) && defined(__INNOTEK_LIBC__)
+    char tocode_ucs2[] = "UCS-2LE";
+    char fromcode_ucs2[] = "UCS-2LE";
+    os2_iconv_t *p_os2_iconv;
+
+    /* Workaround for UTF-16 because OS/2 supports UCS-2 only not UTF-16 */
+    if( !strncmp( tocode, "UTF-16", 6 ))
+    {
+        strncpy( tocode_ucs2 + 5, tocode + 6, 2 );
+        tocode = tocode_ucs2;
+    }
+
+    if( !strncmp( fromcode, "UTF-16", 6 ))
+    {
+        strncpy( fromcode_ucs2 + 5, fromcode + 6, 2 );
+        fromcode = fromcode_ucs2;
+    }
+
+    p_os2_iconv = ( os2_iconv_t * )iconv_open( tocode, fromcode );
+
+    if( p_os2_iconv != ( iconv_t )(-1))
+    {
+        /* Assume strings contain pathnames */
+        uconv_attribute_t attr;
+
+        UniQueryUconvObject( p_os2_iconv->from, &attr,
+                             sizeof( uconv_attribute_t ),
+                             NULL, NULL, NULL );
+        attr.converttype |= CVTTYPE_PATH;
+        UniSetUconvObject( p_os2_iconv->from, &attr );
+    }
+
+    return ( vlc_iconv_t )p_os2_iconv;
+# else
     return iconv_open( tocode, fromcode );
+# endif
 #else
     return (vlc_iconv_t)(-1);
 #endif
-- 
1.7.3.2



From komh78 at gmail.com  Tue Dec 20 12:29:46 2011
From: komh78 at gmail.com (KO Myung-Hun)
Date: Tue, 20 Dec 2011 20:29:46 +0900
Subject: [vlc-devel] [PATCH 2/3] Define the some macros to avoid name
	clashes on OS/2.
In-Reply-To: <1324380587-225-1-git-send-email-komh@chollian.net>
References: <1324380587-225-1-git-send-email-komh@chollian.net>
Message-ID: <1324380587-225-3-git-send-email-komh@chollian.net>

They are

    #define BITMAPINFOHEADER    VLC_BITMAPINFOHEADER
    #define PBITMAPINFOHEADER   VLC_PBITMAPINFOHEADER
    #define LPBITMAPINFOHEADER  VLC_LPBITMAPINFOHEADER
    #define BITMAPINFO          VLC_BITMAPINFO
    #define LPBITMAPINFO        VLC_LPBITMAPINFO

OS/2 has already the types of the same name, but the different member name.
---
 include/vlc_codecs.h |   10 ++++++++++
 1 files changed, 10 insertions(+), 0 deletions(-)

diff --git a/include/vlc_codecs.h b/include/vlc_codecs.h
index 6963178..85cc766 100644
--- a/include/vlc_codecs.h
+++ b/include/vlc_codecs.h
@@ -108,6 +108,16 @@ _WAVEHEADER {
 
 #if !defined(_BITMAPINFOHEADER_) && !defined(WIN32)
 #define _BITMAPINFOHEADER_
+
+#ifdef __OS2__
+/* OS/2 has the types having the same name */
+#define BITMAPINFOHEADER    VLC_BITMAPINFOHEADER
+#define PBITMAPINFOHEADER   VLC_PBITMAPINFOHEADER
+#define LPBITMAPINFOHEADER  VLC_LPBITMAPINFOHEADER
+#define BITMAPINFO          VLC_BITMAPINFO
+#define LPBITMAPINFO        VLC_LPBITMAPINFO
+#endif
+
 typedef struct
 ATTR_PACKED
 {
-- 
1.7.3.2



From komh78 at gmail.com  Tue Dec 20 12:29:47 2011
From: komh78 at gmail.com (KO Myung-Hun)
Date: Tue, 20 Dec 2011 20:29:47 +0900
Subject: [vlc-devel] [PATCH 3/3] Do not use libdvdnav's read ahead cache on
	OS/2.
In-Reply-To: <1324380587-225-1-git-send-email-komh@chollian.net>
References: <1324380587-225-1-git-send-email-komh@chollian.net>
Message-ID: <1324380587-225-4-git-send-email-komh@chollian.net>

Using it causes libdvdnav to complain about 'demux error'.
---
 modules/access/dvdnav.c |    4 ++++
 1 files changed, 4 insertions(+), 0 deletions(-)

diff --git a/modules/access/dvdnav.c b/modules/access/dvdnav.c
index 2a7d647..fad5ed6 100644
--- a/modules/access/dvdnav.c
+++ b/modules/access/dvdnav.c
@@ -97,7 +97,11 @@ vlc_module_begin ()
 vlc_module_end ()
 
 /* Shall we use libdvdnav's read ahead cache? */
+#ifdef __OS2__
+#define DVD_READ_CACHE 0
+#else
 #define DVD_READ_CACHE 1
+#endif
 
 /*****************************************************************************
  * Local prototypes
-- 
1.7.3.2



From mirsal at mirsal.fr  Tue Dec 20 13:18:09 2011
From: mirsal at mirsal.fr (mirsal)
Date: Tue, 20 Dec 2011 13:18:09 +0100
Subject: [vlc-devel] [vlc-commits] upnp: Switch to an unlimited maximum
 content length
In-Reply-To: <CAOa_K+tD4rDt27V_ycZHiQVTQrC3GBMTz60AdFVV0o451fuosQ@mail.gmail.com>
References: <20111220073910.B3409143F8C@albiero.videolan.org>
	<CAOa_K+tD4rDt27V_ycZHiQVTQrC3GBMTz60AdFVV0o451fuosQ@mail.gmail.com>
Message-ID: <1324383489.32572.27.camel@mirsal-laptop1.mirsal.fr>

On Tue, 2011-12-20 at 19:14 +0900, Naohiro KORIYAMA wrote:
> 2011/12/20 Mirsal Ennaime <git at videolan.org>:
> > vlc | branch: master | Mirsal Ennaime <mirsal at videolan.org> | Tue Dec 20 08:34:45 2011 +0100| [794557eea63853456cf3120cdb1bdc88ca44ad9f] | committer: Mirsal Ennaime
> 
> win32 build failed because  INT_MAX is not defined.
> #include <climits>
> or
> #include <limits.h>
> is required.

Yes, I'm sorry I just saw it.
Fixed in 75a40b63d0cb0ba655a80c4030ce54cb812c471f

Best,

-- 
mirsal 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 490 bytes
Desc: This is a digitally signed message part
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111220/57d13512/attachment.sig>

From jb at videolan.org  Tue Dec 20 14:10:43 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Tue, 20 Dec 2011 14:10:43 +0100
Subject: [vlc-devel] OS/2 patches, season 5
In-Reply-To: <1324380587-225-1-git-send-email-komh@chollian.net>
References: <1324380587-225-1-git-send-email-komh@chollian.net>
Message-ID: <20111220131043.GA8794@videolan.org>

On Tue, Dec 20, 2011 at 08:29:44PM +0900, KO Myung-Hun wrote :
> These are season 5 of OS/2 patches.

Already Season 5? :D


-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Tue Dec 20 14:11:32 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Tue, 20 Dec 2011 14:11:32 +0100
Subject: [vlc-devel] [PATCH 3/3] Do not use libdvdnav's read ahead cache
 on OS/2.
In-Reply-To: <1324380587-225-4-git-send-email-komh@chollian.net>
References: <1324380587-225-1-git-send-email-komh@chollian.net>
	<1324380587-225-4-git-send-email-komh@chollian.net>
Message-ID: <20111220131132.GB8794@videolan.org>

On Tue, Dec 20, 2011 at 08:29:47PM +0900, KO Myung-Hun wrote :
> Using it causes libdvdnav to complain about 'demux error'.

This is not normal, but OK, since I do not want to read libdvdnav
code...

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From exclusion at gmail.com  Tue Dec 20 11:20:30 2011
From: exclusion at gmail.com (Alexander LAW)
Date: Tue, 20 Dec 2011 14:20:30 +0400
Subject: [vlc-devel] [PATCH] VLC qt4 menu contains non-localizable Speed item
Message-ID: <4EF0616E.9040103@gmail.com>

The patch for the issue.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-Speed-menu-item-is-not-localizable.patch
Type: text/x-patch
Size: 808 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111220/37463c97/attachment.bin>

From ben at europa-network.com  Tue Dec 20 17:20:21 2011
From: ben at europa-network.com (Ben Wheway)
Date: Tue, 20 Dec 2011 17:20:21 +0100
Subject: [vlc-devel] =?iso-8859-1?q?RTSP_Amino=B4s?=
Message-ID: <090601ccbf33$45d55d30$d1801790$@com>

Hi All

 

We currently stream using VLC to Motorola STB?s. We need to do the same with
our Amino A125 STB?s. But the Amino RTSP protocol doesn?t work with VLC.

 

I have noticed many of you have created a patch / workaround to allow this.

 

Could I have access to this patch or pay someone to do this for me?

 

Thankyou

 

Ben

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111220/80862f7f/attachment.html>

From remi at remlab.net  Tue Dec 20 18:37:16 2011
From: remi at remlab.net (=?iso-8859-15?q?R=E9mi?= Denis-Courmont)
Date: Tue, 20 Dec 2011 19:37:16 +0200
Subject: [vlc-devel] =?utf-8?q?RTSP_Amino=C2=B4s?=
In-Reply-To: <090601ccbf33$45d55d30$d1801790$@com>
References: <090601ccbf33$45d55d30$d1801790$@com>
Message-ID: <201112201937.16690.remi@remlab.net>

Le mardi 20 d?cembre 2011 18:20:21 Ben Wheway, vous avez ?crit :
> I have noticed many of you have created a patch / workaround to allow this.

We do not have such a patch.

Complain to Amino for not respecting protocol specifications.


-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From cheng.sun at ymail.com  Tue Dec 20 19:33:35 2011
From: cheng.sun at ymail.com (Cheng Sun)
Date: Tue, 20 Dec 2011 18:33:35 +0000
Subject: [vlc-devel] [PATCH] Fix vorbis channel order (which led to crash
 and contradicted spec)
Message-ID: <4EF0D4FF.3000901@ymail.com>

Fixes ticket 5704.

The vorbis spec says that the physical channels are left, right, rear
left, rear right, not left, right, center, LFE as is implemented.
(http://www.xiph.org/vorbis/doc/Vorbis_I_spec.html#x1-800004.3.9)

-- Cheng
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111220/71853cd7/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-Fix-vorbis-channel-order-which-led-to-crash-and-cont.patch
Type: text/x-patch
Size: 902 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111220/71853cd7/attachment.bin>

From mans at mansr.com  Tue Dec 20 22:28:27 2011
From: mans at mansr.com (=?iso-8859-1?Q?M=E5ns_Rullg=E5rd?=)
Date: Tue, 20 Dec 2011 21:28:27 +0000
Subject: [vlc-devel] =?iso-8859-1?q?RTSP_Amino=B4s?=
References: <090601ccbf33$45d55d30$d1801790$@com>
Message-ID: <yw1xiplbq6f8.fsf@unicorn.mansr.com>

"Ben Wheway" <ben at europa-network.com> writes:

> Hi All
>
> We currently stream using VLC to Motorola STB?s. We need to do the same with
> our Amino A125 STB?s. But the Amino RTSP protocol doesn?t work with VLC.

Amino uses something almost but not quite entirely unlike RTSP.

-- 
M?ns Rullg?rd
mans at mansr.com



From jb at videolan.org  Wed Dec 21 01:25:18 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Wed, 21 Dec 2011 01:25:18 +0100
Subject: [vlc-devel] [PATCH] Fix vorbis channel order (which led to
 crash and contradicted spec)
In-Reply-To: <4EF0D4FF.3000901@ymail.com>
References: <4EF0D4FF.3000901@ymail.com>
Message-ID: <20111221002518.GA22077@videolan.org>

On Tue, Dec 20, 2011 at 06:33:35PM +0000, Cheng Sun wrote :
> Fixes ticket 5704.
> 
> The vorbis spec says that the physical channels are left, right, rear
> left, rear right, not left, right, center, LFE as is implemented.
> (http://www.xiph.org/vorbis/doc/Vorbis_I_spec.html#x1-800004.3.9)

Applied.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Wed Dec 21 01:37:22 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Wed, 21 Dec 2011 01:37:22 +0100
Subject: [vlc-devel] [PATCH 2/3] Define the some macros to avoid name
 clashes on OS/2.
In-Reply-To: <1324380587-225-3-git-send-email-komh@chollian.net>
References: <1324380587-225-1-git-send-email-komh@chollian.net>
	<1324380587-225-3-git-send-email-komh@chollian.net>
Message-ID: <20111221003722.GA22705@videolan.org>

On Tue, Dec 20, 2011 at 08:29:46PM +0900, KO Myung-Hun wrote :
> They are
> 
>     #define BITMAPINFOHEADER    VLC_BITMAPINFOHEADER
>     #define PBITMAPINFOHEADER   VLC_PBITMAPINFOHEADER
>     #define LPBITMAPINFOHEADER  VLC_LPBITMAPINFOHEADER
>     #define BITMAPINFO          VLC_BITMAPINFO
>     #define LPBITMAPINFO        VLC_LPBITMAPINFO
> 
> OS/2 has already the types of the same name, but the different member name.

I am really not at ease with that...

BITMAPINFOHEADER is quite defined by MicroSoft. Any way to undef the
OS/2 defines?


-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Wed Dec 21 01:43:46 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Wed, 21 Dec 2011 01:43:46 +0100
Subject: [vlc-devel] [PATCH] VLC qt4 menu contains non-localizable Speed
 item
In-Reply-To: <4EF0616E.9040103@gmail.com>
References: <4EF0616E.9040103@gmail.com>
Message-ID: <20111221004346.GA25483@videolan.org>

On Tue, Dec 20, 2011 at 02:20:30PM +0400, Alexander LAW wrote :
> The patch for the issue.

Applied, thanks.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Wed Dec 21 02:25:24 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Wed, 21 Dec 2011 02:25:24 +0100
Subject: [vlc-devel] [PATCH] yadif : Add SSSE3 and SSE2 support. porting
 from FFmpeg.
In-Reply-To: <CAOa_K+ug10ymVFtCeTTHo=TwzOzynfiON9WE-Ci4L_A_x45WzQ@mail.gmail.com>
References: <CAOa_K+vdRUzBj4cE9qoLyU19+NYnVuPx03UqdOEXgkT6fw60Gg@mail.gmail.com>
	<201112091941.15730.remi@remlab.net>
	<CAOa_K+shj1Wy4pm3=sFZxw4_nM+k-VCZnRauiQujeV0+LS4CGw@mail.gmail.com>
	<201112171042.02575.remi@remlab.net>
	<CAOa_K+tBjZCS5sSve_vwCt4DbzX2qpJGa37tTiZkX-tKOaUQgA@mail.gmail.com>
	<CAOa_K+ug10ymVFtCeTTHo=TwzOzynfiON9WE-Ci4L_A_x45WzQ@mail.gmail.com>
Message-ID: <20111221012523.GA1174@videolan.org>

On Tue, Dec 20, 2011 at 08:08:46PM +0900, Naohiro KORIYAMA wrote :
> With a patch, I sent 3 days ago, inline assembler part of code may not
> be compiled on some environment.
> Attached patch fixed the issue.

I have:


CC     libdeinterlace_plugin_la-algo_yadif.lo
In file included from ../../../modules/video_filter/deinterlace/yadif.h:49:0,
                 from ../../../modules/video_filter/deinterlace/algo_yadif.c:48:
../../../modules/video_filter/deinterlace/yadif_template.h: In function 'yadif_filter_line_ssse3':
../../../modules/video_filter/deinterlace/yadif_template.h:114:32: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast]
../../../modules/video_filter/deinterlace/yadif_template.h:114:20: warning: cast to pointer from integer of different size [-Wint-to-pointer-cast]
In file included from ../../../modules/video_filter/deinterlace/yadif.h:64:0,
                 from ../../../modules/video_filter/deinterlace/algo_yadif.c:48:
../../../modules/video_filter/deinterlace/yadif_template.h: In function 'yadif_filter_line_sse2':
../../../modules/video_filter/deinterlace/yadif_template.h:114:32: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast]
../../../modules/video_filter/deinterlace/yadif_template.h:114:20: warning: cast to pointer from integer of different size [-Wint-to-pointer-cast]
In file included from ../../../modules/video_filter/deinterlace/yadif.h:77:0,
                 from ../../../modules/video_filter/deinterlace/algo_yadif.c:48:
../../../modules/video_filter/deinterlace/yadif_template.h: In function 'yadif_filter_line_mmx':
../../../modules/video_filter/deinterlace/yadif_template.h:114:32: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast]
../../../modules/video_filter/deinterlace/yadif_template.h:114:20: warning: cast to pointer from integer of different size [-Wint-to-pointer-cast]

Expected?

> sorry for sending modification many times.

Please, do not excuse yourself for improving a patch.
People should excuse themselves when they don't answer to a review, not the opposite :D

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From funman at videolan.org  Wed Dec 21 04:09:33 2011
From: funman at videolan.org (=?UTF-8?q?Rafa=C3=ABl=20Carr=C3=A9?=)
Date: Tue, 20 Dec 2011 22:09:33 -0500
Subject: [vlc-devel] [PATCH] messages.c: added logging via OutputDebugString
	under windows.
Message-ID: <1324436973-30963-1-git-send-email-funman@videolan.org>

Signed-off-by: Rafa?l Carr? <funman at videolan.org>
---
 src/misc/messages.c |   38 ++++++++++++++++++++++++++++++++++++++
 1 files changed, 38 insertions(+), 0 deletions(-)

diff --git a/src/misc/messages.c b/src/misc/messages.c
index 8c0b723..e6b6aeb 100644
--- a/src/misc/messages.c
+++ b/src/misc/messages.c
@@ -132,6 +132,10 @@ void vlc_Log (vlc_object_t *obj, int type, const char *module,
 static void PrintColorMsg (void *, int, const msg_item_t *,
                            const char *, va_list);
 static void PrintMsg (void *, int, const msg_item_t *, const char *, va_list);
+#ifdef WIN32
+static void Win32DebugOutputMsg (void *, int , const msg_item_t *,
+                                 const char *, va_list);
+#endif
 
 /**
  * Emit a log message. This function is the variable argument list equivalent
@@ -230,6 +234,13 @@ void vlc_vaLog (vlc_object_t *obj, int type, const char *module,
         PrintMsg (&priv->i_verbose, type, &msg, format, ap);
     va_end (ap);
 
+#ifdef WIN32
+    va_list dol;
+    va_copy (dol, args);
+    Win32DebugOutputMsg (0, type, &msg, format, dol);
+    va_end (dol);
+#endif
+
     vlc_rwlock_rdlock (&msg_lock);
     for (msg_subscription_t *sub = msg_head; sub != NULL; sub = sub->next)
     {
@@ -303,3 +314,30 @@ static void PrintMsg (void *d, int type, const msg_item_t *p_item,
     funlockfile (stream);
     vlc_restorecancel (canc);
 }
+
+#ifdef WIN32
+static void Win32DebugOutputMsg (void* d, int type, const msg_item_t *p_item,
+                                 const char *format, va_list dol)
+{
+    const signed char *pverbose = d;
+    if (*pverbose < 0 || *pverbose < (type - VLC_MSG_ERR))
+        return;
+
+    size_t fmt_len = strlen(format);
+    char *fmt_cr = malloc(fmt_len + 1 + 1);
+    if (!fmt_cr)
+        return;
+
+    strcpy(fmt_cr, format);
+    fmt_cr[fmt_len] = '\n';
+    fmt_cr[fmt_len + 1] = '\0';
+    char *msg;
+    if (vasprintf(&msg, fmt_cr, dol))
+        msg = NULL;
+
+    if (msg)
+        OutputDebugString(msg);
+    free(fmt_cr);
+    free(msg);
+}
+#endif
-- 
1.7.7.3


From remi at remlab.net  Wed Dec 21 08:01:59 2011
From: remi at remlab.net (=?UTF-8?Q?R=C3=A9mi_Denis-Courmont?=)
Date: Wed, 21 Dec 2011 08:01:59 +0100
Subject: [vlc-devel] [PATCH] messages.c: added logging via
 OutputDebugString under windows.
In-Reply-To: <1324436973-30963-1-git-send-email-funman@videolan.org>
References: <1324436973-30963-1-git-send-email-funman@videolan.org>
Message-ID: <573ec901bb61a887e6c7678bc4d8f856@chewa.net>

On Tue, 20 Dec 2011 22:09:33 -0500, Sergey Radionov <RSATom at gmail.com>
wrote:
> Signed-off-by: Rafa?l Carr? <funman at videolan.org>
> ---
>  src/misc/messages.c |   38 ++++++++++++++++++++++++++++++++++++++
>  1 files changed, 38 insertions(+), 0 deletions(-)
> 
> diff --git a/src/misc/messages.c b/src/misc/messages.c
> index 8c0b723..e6b6aeb 100644
> --- a/src/misc/messages.c
> +++ b/src/misc/messages.c
> @@ -132,6 +132,10 @@ void vlc_Log (vlc_object_t *obj, int type, const
char
> *module,
>  static void PrintColorMsg (void *, int, const msg_item_t *,
>                             const char *, va_list);
>  static void PrintMsg (void *, int, const msg_item_t *, const char *,
>  va_list);
> +#ifdef WIN32
> +static void Win32DebugOutputMsg (void *, int , const msg_item_t *,
> +                                 const char *, va_list);
> +#endif
>  
>  /**
>   * Emit a log message. This function is the variable argument list
>   equivalent
> @@ -230,6 +234,13 @@ void vlc_vaLog (vlc_object_t *obj, int type, const
> char *module,
>          PrintMsg (&priv->i_verbose, type, &msg, format, ap);
>      va_end (ap);
>  
> +#ifdef WIN32
> +    va_list dol;
> +    va_copy (dol, args);
> +    Win32DebugOutputMsg (0, type, &msg, format, dol);

Please don't use 0 as a pointer.

> +    va_end (dol);
> +#endif
> +
>      vlc_rwlock_rdlock (&msg_lock);
>      for (msg_subscription_t *sub = msg_head; sub != NULL; sub =
sub->next)
>      {
> @@ -303,3 +314,30 @@ static void PrintMsg (void *d, int type, const
> msg_item_t *p_item,
>      funlockfile (stream);
>      vlc_restorecancel (canc);
>  }
> +
> +#ifdef WIN32
> +static void Win32DebugOutputMsg (void* d, int type, const msg_item_t
> *p_item,
> +                                 const char *format, va_list dol)
> +{
> +    const signed char *pverbose = d;
> +    if (*pverbose < 0 || *pverbose < (type - VLC_MSG_ERR))
> +        return;

Did you actually test this? It looks like *NULL to me.

> +
> +    size_t fmt_len = strlen(format);
> +    char *fmt_cr = malloc(fmt_len + 1 + 1);
> +    if (!fmt_cr)
> +        return;
> +
> +    strcpy(fmt_cr, format);
> +    fmt_cr[fmt_len] = '\n';
> +    fmt_cr[fmt_len + 1] = '\0';
> +    char *msg;
> +    if (vasprintf(&msg, fmt_cr, dol))
> +        msg = NULL;
> +
> +    if (msg)
> +        OutputDebugString(msg);
> +    free(fmt_cr);
> +    free(msg);
> +}
> +#endif

-- 
R?mi Denis-Courmont
http://www.remlab.net/


From nkoriyama at gmail.com  Wed Dec 21 09:09:12 2011
From: nkoriyama at gmail.com (Naohiro KORIYAMA)
Date: Wed, 21 Dec 2011 17:09:12 +0900
Subject: [vlc-devel] [PATCH] yadif : Add SSSE3 and SSE2 support. porting
 from FFmpeg.
In-Reply-To: <20111221012523.GA1174@videolan.org>
References: <CAOa_K+vdRUzBj4cE9qoLyU19+NYnVuPx03UqdOEXgkT6fw60Gg@mail.gmail.com>
	<201112091941.15730.remi@remlab.net>
	<CAOa_K+shj1Wy4pm3=sFZxw4_nM+k-VCZnRauiQujeV0+LS4CGw@mail.gmail.com>
	<201112171042.02575.remi@remlab.net>
	<CAOa_K+tBjZCS5sSve_vwCt4DbzX2qpJGa37tTiZkX-tKOaUQgA@mail.gmail.com>
	<CAOa_K+ug10ymVFtCeTTHo=TwzOzynfiON9WE-Ci4L_A_x45WzQ@mail.gmail.com>
	<20111221012523.GA1174@videolan.org>
Message-ID: <CAOa_K+sPqxzZY5DFimC+Ri0-jbEW5pTGLCvDcVkQyTTo1OuCRw@mail.gmail.com>

2011/12/21 Jean-Baptiste Kempf <jb at videolan.org>:
> On Tue, Dec 20, 2011 at 08:08:46PM +0900, Naohiro KORIYAMA wrote :
>> With a patch, I sent 3 days ago, inline assembler part of code may not
>> be compiled on some environment.
>> Attached patch fixed the issue.
>
> I have:
>
>
> CC ? ? libdeinterlace_plugin_la-algo_yadif.lo
> In file included from ../../../modules/video_filter/deinterlace/yadif.h:49:0,
> ? ? ? ? ? ? ? ? from ../../../modules/video_filter/deinterlace/algo_yadif.c:48:
> ../../../modules/video_filter/deinterlace/yadif_template.h: In function 'yadif_filter_line_ssse3':
> ../../../modules/video_filter/deinterlace/yadif_template.h:114:32: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast]
> ../../../modules/video_filter/deinterlace/yadif_template.h:114:20: warning: cast to pointer from integer of different size [-Wint-to-pointer-cast]
> In file included from ../../../modules/video_filter/deinterlace/yadif.h:64:0,
> ? ? ? ? ? ? ? ? from ../../../modules/video_filter/deinterlace/algo_yadif.c:48:
> ../../../modules/video_filter/deinterlace/yadif_template.h: In function 'yadif_filter_line_sse2':
> ../../../modules/video_filter/deinterlace/yadif_template.h:114:32: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast]
> ../../../modules/video_filter/deinterlace/yadif_template.h:114:20: warning: cast to pointer from integer of different size [-Wint-to-pointer-cast]
> In file included from ../../../modules/video_filter/deinterlace/yadif.h:77:0,
> ? ? ? ? ? ? ? ? from ../../../modules/video_filter/deinterlace/algo_yadif.c:48:
> ../../../modules/video_filter/deinterlace/yadif_template.h: In function 'yadif_filter_line_mmx':
> ../../../modules/video_filter/deinterlace/yadif_template.h:114:32: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast]
> ../../../modules/video_filter/deinterlace/yadif_template.h:114:20: warning: cast to pointer from integer of different size [-Wint-to-pointer-cast]
>
> Expected?

According to http://ffmpeg.org/trac/ffmpeg/ticket/25 and
http://patches.libav.org/patch/2444/, this is a hack for MS Visual
C++.
And it isn't necessary for VLC, I think.

Here is another patch for the warning. I compiled it on Linux 64 bit
gcc,  Mac OS X 64biy clang, gcc and Windows 32 bit gcc.

-- 
KORIYAMA, Naohiro
nkoriyama at gmail.com


From nkoriyama at gmail.com  Wed Dec 21 09:14:26 2011
From: nkoriyama at gmail.com (Naohiro KORIYAMA)
Date: Wed, 21 Dec 2011 17:14:26 +0900
Subject: [vlc-devel] [PATCH] yadif : Add SSSE3 and SSE2 support. porting
 from FFmpeg.
In-Reply-To: <CAOa_K+sPqxzZY5DFimC+Ri0-jbEW5pTGLCvDcVkQyTTo1OuCRw@mail.gmail.com>
References: <CAOa_K+vdRUzBj4cE9qoLyU19+NYnVuPx03UqdOEXgkT6fw60Gg@mail.gmail.com>
	<201112091941.15730.remi@remlab.net>
	<CAOa_K+shj1Wy4pm3=sFZxw4_nM+k-VCZnRauiQujeV0+LS4CGw@mail.gmail.com>
	<201112171042.02575.remi@remlab.net>
	<CAOa_K+tBjZCS5sSve_vwCt4DbzX2qpJGa37tTiZkX-tKOaUQgA@mail.gmail.com>
	<CAOa_K+ug10ymVFtCeTTHo=TwzOzynfiON9WE-Ci4L_A_x45WzQ@mail.gmail.com>
	<20111221012523.GA1174@videolan.org>
	<CAOa_K+sPqxzZY5DFimC+Ri0-jbEW5pTGLCvDcVkQyTTo1OuCRw@mail.gmail.com>
Message-ID: <CAOa_K+tO9B6VmKYBdRh2t6ASis4_wTPO8QQVbYtdpobiOoq+wQ@mail.gmail.com>

2011/12/21 Naohiro KORIYAMA <nkoriyama at gmail.com>:
> 2011/12/21 Jean-Baptiste Kempf <jb at videolan.org>:
>> On Tue, Dec 20, 2011 at 08:08:46PM +0900, Naohiro KORIYAMA wrote :
>>> With a patch, I sent 3 days ago, inline assembler part of code may not
>>> be compiled on some environment.
>>> Attached patch fixed the issue.
>>
>> I have:
>>
>>
>> CC ? ? libdeinterlace_plugin_la-algo_yadif.lo
>> In file included from ../../../modules/video_filter/deinterlace/yadif.h:49:0,
>> ? ? ? ? ? ? ? ? from ../../../modules/video_filter/deinterlace/algo_yadif.c:48:
>> ../../../modules/video_filter/deinterlace/yadif_template.h: In function 'yadif_filter_line_ssse3':
>> ../../../modules/video_filter/deinterlace/yadif_template.h:114:32: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast]
>> ../../../modules/video_filter/deinterlace/yadif_template.h:114:20: warning: cast to pointer from integer of different size [-Wint-to-pointer-cast]
>> In file included from ../../../modules/video_filter/deinterlace/yadif.h:64:0,
>> ? ? ? ? ? ? ? ? from ../../../modules/video_filter/deinterlace/algo_yadif.c:48:
>> ../../../modules/video_filter/deinterlace/yadif_template.h: In function 'yadif_filter_line_sse2':
>> ../../../modules/video_filter/deinterlace/yadif_template.h:114:32: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast]
>> ../../../modules/video_filter/deinterlace/yadif_template.h:114:20: warning: cast to pointer from integer of different size [-Wint-to-pointer-cast]
>> In file included from ../../../modules/video_filter/deinterlace/yadif.h:77:0,
>> ? ? ? ? ? ? ? ? from ../../../modules/video_filter/deinterlace/algo_yadif.c:48:
>> ../../../modules/video_filter/deinterlace/yadif_template.h: In function 'yadif_filter_line_mmx':
>> ../../../modules/video_filter/deinterlace/yadif_template.h:114:32: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast]
>> ../../../modules/video_filter/deinterlace/yadif_template.h:114:20: warning: cast to pointer from integer of different size [-Wint-to-pointer-cast]
>>
>> Expected?
>
> According to http://ffmpeg.org/trac/ffmpeg/ticket/25 and
> http://patches.libav.org/patch/2444/, this is a hack for MS Visual
> C++.
> And it isn't necessary for VLC, I think.
>
> Here is another patch for the warning. I compiled it on Linux 64 bit
> gcc, ?Mac OS X 64biy clang, gcc and Windows 32 bit gcc.

I forget the attachment...

-- 
KORIYAMA, Naohiro
nkoriyama at gmail.com
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-yadif-Add-SSSE3-and-SSE2-support.-porting-from-FFmpe.patch
Type: application/octet-stream
Size: 30300 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111221/3e8a9750/attachment.obj>

From funman at videolan.org  Wed Dec 21 10:16:19 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Wed, 21 Dec 2011 04:16:19 -0500
Subject: [vlc-devel] [PATCH] messages.c: added logging via
 OutputDebugString under windows.
In-Reply-To: <573ec901bb61a887e6c7678bc4d8f856@chewa.net>
References: <1324436973-30963-1-git-send-email-funman@videolan.org>
	<573ec901bb61a887e6c7678bc4d8f856@chewa.net>
Message-ID: <20111221041619.61b02133@tok>

Le Wed, 21 Dec 2011 08:01:59 +0100,
R?mi Denis-Courmont <remi at remlab.net> a ?crit :

> On Tue, 20 Dec 2011 22:09:33 -0500, Sergey Radionov <RSATom at gmail.com>
> wrote:
> > Signed-off-by: Rafa?l Carr? <funman at videolan.org>
> > ---
> >  src/misc/messages.c |   38 ++++++++++++++++++++++++++++++++++++++
> >  1 files changed, 38 insertions(+), 0 deletions(-)
> > 
> > diff --git a/src/misc/messages.c b/src/misc/messages.c
> > index 8c0b723..e6b6aeb 100644
> > --- a/src/misc/messages.c
> > +++ b/src/misc/messages.c
> > @@ -132,6 +132,10 @@ void vlc_Log (vlc_object_t *obj, int type, const
> char
> > *module,
> >  static void PrintColorMsg (void *, int, const msg_item_t *,
> >                             const char *, va_list);
> >  static void PrintMsg (void *, int, const msg_item_t *, const char *,
> >  va_list);
> > +#ifdef WIN32
> > +static void Win32DebugOutputMsg (void *, int , const msg_item_t *,
> > +                                 const char *, va_list);
> > +#endif
> >  
> >  /**
> >   * Emit a log message. This function is the variable argument list
> >   equivalent
> > @@ -230,6 +234,13 @@ void vlc_vaLog (vlc_object_t *obj, int type, const
> > char *module,
> >          PrintMsg (&priv->i_verbose, type, &msg, format, ap);
> >      va_end (ap);
> >  
> > +#ifdef WIN32
> > +    va_list dol;
> > +    va_copy (dol, args);
> > +    Win32DebugOutputMsg (0, type, &msg, format, dol);
> 
> Please don't use 0 as a pointer.
> 
> > +    va_end (dol);
> > +#endif
> > +
> >      vlc_rwlock_rdlock (&msg_lock);
> >      for (msg_subscription_t *sub = msg_head; sub != NULL; sub =
> sub->next)
> >      {
> > @@ -303,3 +314,30 @@ static void PrintMsg (void *d, int type, const
> > msg_item_t *p_item,
> >      funlockfile (stream);
> >      vlc_restorecancel (canc);
> >  }
> > +
> > +#ifdef WIN32
> > +static void Win32DebugOutputMsg (void* d, int type, const msg_item_t
> > *p_item,
> > +                                 const char *format, va_list dol)
> > +{
> > +    const signed char *pverbose = d;
> > +    if (*pverbose < 0 || *pverbose < (type - VLC_MSG_ERR))
> > +        return;
> 
> Did you actually test this? It looks like *NULL to me.

Not yet, vlc exited before calling msg_* (or did it dereference NULL and 
exited without segfaulting?)

I will let Sergey send another patch.

Btw this was Sergey patch, I'm not sure how I should use git send-email with
a patch of someone else

> > +
> > +    size_t fmt_len = strlen(format);
> > +    char *fmt_cr = malloc(fmt_len + 1 + 1);
> > +    if (!fmt_cr)
> > +        return;
> > +
> > +    strcpy(fmt_cr, format);
> > +    fmt_cr[fmt_len] = '\n';
> > +    fmt_cr[fmt_len + 1] = '\0';
> > +    char *msg;
> > +    if (vasprintf(&msg, fmt_cr, dol))
> > +        msg = NULL;
> > +
> > +    if (msg)
> > +        OutputDebugString(msg);
> > +    free(fmt_cr);
> > +    free(msg);
> > +}
> > +#endif
> 



-- 
Rafa?l Carr?


From fyhuel at viotech.net  Wed Dec 21 11:02:00 2011
From: fyhuel at viotech.net (Frederic YHUEL)
Date: Wed, 21 Dec 2011 11:02:00 +0100
Subject: [vlc-devel] [PATCH] Add Smooth Streaming module
In-Reply-To: <1321548124-13651-1-git-send-email-fyhuel@viotech.net>
References: <1321548124-13651-1-git-send-email-fyhuel@viotech.net>
Message-ID: <CAKHNsC3rTfF24psDv1KrLtkyeUf3MEAfFxtrfbx0MfYKTxaL_A@mail.gmail.com>

Here is an updated version of my Smooth Streaming module, with "live"
support, bug fixes, and some cleaning.

I would maybe not call that good code, but at least the call graph is
quite simple: https://plus.google.com/photos/103704069703034617121/albums/5688256729648327185

There is still a lot of work to do:

- Replace the Python script by C code.

- Modify MP4 demuxer of VLC so that it can handle fMP4.

- Find a way to reinitialize libavcodec properly so that adaptation is possible.

- Give the user the possibility to choose / switch between streams
(e.g. choose language or subtitles). I have no idea about how to do
that, actually the module choose the first video stream available and
the first audio stream available.

Best Regards,

--
Fr?d?ric
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-Add-Smooth-Streaming-module.patch
Type: text/x-diff
Size: 47066 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111221/a3f29d5e/attachment.patch>

From cheng.sun at ymail.com  Wed Dec 21 12:01:42 2011
From: cheng.sun at ymail.com (Cheng Sun)
Date: Wed, 21 Dec 2011 11:01:42 +0000
Subject: [vlc-devel] [PATCH] Audio channel remapping filter
Message-ID: <4EF1BC96.6080800@ymail.com>

This filter allows channels to be remapped. Potential uses include
up/downmixing, or to correct non-standard channel orders.

Patch attached. Comments welcome.

-- Cheng
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-Audio-channel-remapping-filter.patch
Type: text/x-patch
Size: 16885 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111221/65de9d67/attachment.bin>

From fyhuel at viotech.net  Wed Dec 21 12:02:48 2011
From: fyhuel at viotech.net (Frederic YHUEL)
Date: Wed, 21 Dec 2011 12:02:48 +0100
Subject: [vlc-devel] [PATCH] Add Smooth Streaming module
In-Reply-To: <CAKHNsC3rTfF24psDv1KrLtkyeUf3MEAfFxtrfbx0MfYKTxaL_A@mail.gmail.com>
References: <1321548124-13651-1-git-send-email-fyhuel@viotech.net>
	<CAKHNsC3rTfF24psDv1KrLtkyeUf3MEAfFxtrfbx0MfYKTxaL_A@mail.gmail.com>
Message-ID: <CAKHNsC3CziJmtu5-CJKwEHmCNiAF+ioi_vsbsg7y4THA2drc+w@mail.gmail.com>

On Wed, Dec 21, 2011 at 11:02 AM, Frederic YHUEL <fyhuel at viotech.net> wrote:
> Here is an updated version of my Smooth Streaming module, with "live"
> support, bug fixes, and some cleaning.
>

BTW, if you to test it with Anevia server for example (see below), you
should use "libav" instead of "ffmpeg". I don't know why, but as for
me it works well with libav 0.7.2, but not with ffmpeg whatever
version.

***** Live *****
http://demo.anevia.com:3129/html/player/smooth/channel2.isml/manifest
http://demo.anevia.com:3128/html/player/smooth/channel1.isml/manifest

***** VOD *****
http://demo.anevia.com:3130/html/player/smooth/vod/content1.ism/Manifest
http://demo.anevia.com:3131/html/player/smooth/vod/IbcVoile2011.ism/Manifest

Best Regards,

--
Fr?d?ric


From remi at remlab.net  Wed Dec 21 12:03:19 2011
From: remi at remlab.net (=?UTF-8?Q?R=C3=A9mi_Denis-Courmont?=)
Date: Wed, 21 Dec 2011 12:03:19 +0100
Subject: [vlc-devel] [PATCH] messages.c: added logging via
 OutputDebugString under windows.
In-Reply-To: <20111221041619.61b02133@tok>
References: <1324436973-30963-1-git-send-email-funman@videolan.org>
	<573ec901bb61a887e6c7678bc4d8f856@chewa.net>
	<20111221041619.61b02133@tok>
Message-ID: <b8590c75149b20abd851024429f74719@chewa.net>

On Wed, 21 Dec 2011 04:16:19 -0500, Rafa?l Carr? <funman at videolan.org>
> Not yet, vlc exited before calling msg_* (or did it dereference NULL and

> exited without segfaulting?)

Sending untested patches for review is wasting other people's time.

As a retalation, I will not be reviewing this pathc series anymore.

-- 
R?mi Denis-Courmont
http://www.remlab.net/


From alexey at asokolov.org  Wed Dec 21 12:46:25 2011
From: alexey at asokolov.org (Alexey Sokolov)
Date: Wed, 21 Dec 2011 18:46:25 +0700
Subject: [vlc-devel] [PATCH] messages.c: added logging via
 OutputDebugString under windows.
In-Reply-To: <20111221041619.61b02133@tok>
References: <1324436973-30963-1-git-send-email-funman@videolan.org>
	<573ec901bb61a887e6c7678bc4d8f856@chewa.net>
	<20111221041619.61b02133@tok>
Message-ID: <4EF1C711.4090102@asokolov.org>

21.12.2011 16:16, Rafa?l Carr? ?????:
> I will let Sergey send another patch. Btw this was Sergey patch, I'm
> not sure how I should use git send-email with a patch of someone else
Just let people send their patches themselves...

-- 
Best regards,
Alexey "DarthGandalf" Sokolov


-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 900 bytes
Desc: OpenPGP digital signature
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111221/6a4bbeef/attachment.sig>

From komh78 at gmail.com  Wed Dec 21 12:49:32 2011
From: komh78 at gmail.com (KO Myung-Hun)
Date: Wed, 21 Dec 2011 20:49:32 +0900
Subject: [vlc-devel] [PATCH 2/3] Define the some macros to avoid name
	clashes on OS/2.
In-Reply-To: <20111221003722.GA22705@videolan.org>
References: <1324380587-225-1-git-send-email-komh@chollian.net>	<1324380587-225-3-git-send-email-komh@chollian.net>
	<20111221003722.GA22705@videolan.org>
Message-ID: <4EF1C7CC.1050503@chollian.net>



Jean-Baptiste Kempf wrote:
> On Tue, Dec 20, 2011 at 08:29:46PM +0900, KO Myung-Hun wrote :
>> They are
>>
>>     #define BITMAPINFOHEADER    VLC_BITMAPINFOHEADER
>>     #define PBITMAPINFOHEADER   VLC_PBITMAPINFOHEADER
>>     #define LPBITMAPINFOHEADER  VLC_LPBITMAPINFOHEADER
>>     #define BITMAPINFO          VLC_BITMAPINFO
>>     #define LPBITMAPINFO        VLC_LPBITMAPINFO
>>
>> OS/2 has already the types of the same name, but the different member name.
> 
> I am really not at ease with that...
> 
> BITMAPINFOHEADER is quite defined by MicroSoft. Any way to undef the
> OS/2 defines?
> 
> 

Unfortunately, it is also defined by IBM.

I'm sorry I don't know how to undef 'typedef'.

-- 
KO Myung-Hun

Using Mozilla SeaMonkey 2.0.14
Under OS/2 Warp 4 for Korean with FixPak #15
On AMD ThunderBird 1GHz with 512 MB RAM

Korean OS/2 User Community : http://www.ecomstation.co.kr



From komh78 at gmail.com  Wed Dec 21 12:52:07 2011
From: komh78 at gmail.com (KO Myung-Hun)
Date: Wed, 21 Dec 2011 20:52:07 +0900
Subject: [vlc-devel] OS/2 patches, season 5
In-Reply-To: <20111220131043.GA8794@videolan.org>
References: <1324380587-225-1-git-send-email-komh@chollian.net>
	<20111220131043.GA8794@videolan.org>
Message-ID: <4EF1C867.9080307@chollian.net>



Jean-Baptiste Kempf wrote:
> On Tue, Dec 20, 2011 at 08:29:44PM +0900, KO Myung-Hun wrote :
>> These are season 5 of OS/2 patches.
> 
> Already Season 5? :D
> 
> 

Yes.

But I hope that this season will be the last one.

^_______^


-- 
KO Myung-Hun

Using Mozilla SeaMonkey 2.0.14
Under OS/2 Warp 4 for Korean with FixPak #15
On AMD ThunderBird 1GHz with 512 MB RAM

Korean OS/2 User Community : http://www.ecomstation.co.kr



From edward.c.wang at compdigitec.com  Wed Dec 21 13:21:23 2011
From: edward.c.wang at compdigitec.com (Edward Wang)
Date: Wed, 21 Dec 2011 07:21:23 -0500
Subject: [vlc-devel] [PATCH] Qt4: add title to file associations dialog
Message-ID: <4EF1CF43.2080009@compdigitec.com>

Hi all:

Currently, when you click the button to set up file associations on 
Windows, the dialog presented is untitled and just displays the file 
name, "vlc". The attached patch will add add title to file associations 
dialog (see http://trac.videolan.org/vlc/ticket/5711).

Thanks,
     Edward Wang
-------------- next part --------------
A non-text attachment was scrubbed...
Name: patch5711.patch
Type: text/x-patch
Size: 531 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111221/2c8814dc/attachment.bin>

From fatbull at web.de  Wed Dec 21 14:00:18 2011
From: fatbull at web.de (=?ISO-8859-15?Q?Tobias_G=FCntner?=)
Date: Wed, 21 Dec 2011 14:00:18 +0100
Subject: [vlc-devel] [PATCH] messages.c: added logging via
 OutputDebugString under windows.
In-Reply-To: <1324436973-30963-1-git-send-email-funman@videolan.org>
References: <1324436973-30963-1-git-send-email-funman@videolan.org>
Message-ID: <4EF1D862.3070906@web.de>

Am 21.12.2011 04:09, schrieb Sergey Radionov:
> Signed-off-by: Rafa?l Carr?<funman at videolan.org>

> +#ifdef WIN32
> +    va_list dol;
> +    va_copy (dol, args);
> +    Win32DebugOutputMsg (0, type,&msg, format, dol);
> +    va_end (dol);
> +#endif

Maybe test IsDebuggerPresent() as well?

> +    size_t fmt_len = strlen(format);
> +    char *fmt_cr = malloc(fmt_len + 1 + 1);
> +    if (!fmt_cr)
> +        return;
> +
> +    strcpy(fmt_cr, format);
> +    fmt_cr[fmt_len] = '\n';
> +    fmt_cr[fmt_len + 1] = '\0';

Perhaps asprintf would be easier.

> +    char *msg;
> +    if (vasprintf(&msg, fmt_cr, dol))
> +        msg = NULL;

vasprintf returns -1 on error.

Regards,
Tobias



From etix at videolan.org  Wed Dec 21 14:58:37 2011
From: etix at videolan.org (Ludovic Fauvet)
Date: Wed, 21 Dec 2011 14:58:37 +0100
Subject: [vlc-devel] [PATCH 2/3] Qt: rework of the fullscreen controller
In-Reply-To: <1324475918-5987-1-git-send-email-etix@videolan.org>
References: <1324475918-5987-1-git-send-email-etix@videolan.org>
Message-ID: <1324475918-5987-3-git-send-email-etix@videolan.org>

- fix the logic
- simplify
- repair the ability to save the state and position
---
 modules/gui/qt4/components/controller.cpp |  128 ++++++++++++++--------------
 modules/gui/qt4/components/controller.hpp |    4 +
 2 files changed, 68 insertions(+), 64 deletions(-)

diff --git a/modules/gui/qt4/components/controller.cpp b/modules/gui/qt4/components/controller.cpp
index 9ea1f23..3d1bd78 100644
--- a/modules/gui/qt4/components/controller.cpp
+++ b/modules/gui/qt4/components/controller.cpp
@@ -761,42 +761,66 @@ FullscreenControllerWidget::FullscreenControllerWidget( intf_thread_t *_p_i, QWi
               this, setVoutList( vout_thread_t **, int ) );
 
     /* First Move */
-    QRect rect1 = getSettings()->value( "FullScreen/screen" ).toRect();
-    QPoint pos1 = getSettings()->value( "FullScreen/pos" ).toPoint();
-    int number =  var_InheritInteger( p_intf, "qt-fullscreen-screennumber" );
-    if( number == -1 || number > QApplication::desktop()->numScreens() )
-        number = QApplication::desktop()->screenNumber( p_intf->p_sys->p_mi );
-
-    QRect rect = QApplication::desktop()->screenGeometry( number );
-    if( rect == rect1 && rect.contains( pos1, true ) )
-    {
-        move( pos1 );
-        screenRes = QApplication::desktop()->screenGeometry(number);
-    }
-    else
-    {
-        centerFSC( number );
-    }
+    previousPosition = getSettings()->value( "FullScreen/pos" ).toPoint();
+    screenRes = getSettings()->value( "FullScreen/screen" ).toRect();
+    isWideFSC = getSettings()->value( "FullScreen/wide" ).toBool();
+    i_screennumber = var_InheritInteger( p_intf, "qt-fullscreen-screennumber" );
 }
 
 FullscreenControllerWidget::~FullscreenControllerWidget()
 {
-    QPoint pos1 = pos();
-    QRect rect1 = QApplication::desktop()->screenGeometry( pos1 );
-    getSettings()->setValue( "FullScreen/pos", pos1 );
-    getSettings()->setValue( "FullScreen/screen", rect1 );
+    getSettings()->setValue( "FullScreen/pos", previousPosition );
+    getSettings()->setValue( "FullScreen/screen", screenRes );
+    getSettings()->setValue( "FullScreen/wide", isWideFSC );
 
     setVoutList( NULL, 0 );
     vlc_mutex_destroy( &lock );
 }
 
+void FullscreenControllerWidget::restoreFSC()
+{
+    if( !isWideFSC )
+    {
+        /* Restore half-bar and re-centre if needed */
+        setMinimumWidth( FSC_WIDTH );
+        adjustSize();
+
+        QRect currentRes = QApplication::desktop()->screenGeometry( targetScreen() );
+
+        if( currentRes == screenRes &&
+            QApplication::desktop()->screen()->geometry().contains( previousPosition, true ) )
+        {
+            /* Restore to the last known position */
+            move( previousPosition );
+        }
+        else
+        {
+            /* FSC is out of screen or screen resolution changed */
+            msg_Dbg( p_intf, "Recentering the Fullscreen Controller" );
+            centerFSC( targetScreen() );
+            screenRes = currentRes;
+            previousPosition = pos();
+        }
+    }
+    else
+    {
+        /* Dock at the bottom of the screen */
+        updateFullwidthGeometry( targetScreen() );
+    }
+
+#ifdef Q_WS_X11
+    // Tell kwin that we do not want a shadow around the fscontroller
+    setMask( QRegion( 0, 0, width(), height() ) );
+#endif
+}
+
 void FullscreenControllerWidget::centerFSC( int number )
 {
-    screenRes = QApplication::desktop()->screenGeometry(number);
+    QRect currentRes = QApplication::desktop()->screenGeometry( number );
 
     /* screen has changed, calculate new position */
-    QPoint pos = QPoint( screenRes.x() + (screenRes.width() / 2) - (width() / 2),
-            screenRes.y() + screenRes.height() - height());
+    QPoint pos = QPoint( currentRes.x() + (currentRes.width() / 2) - (width() / 2),
+            currentRes.y() + currentRes.height() - height());
     move( pos );
 }
 
@@ -805,34 +829,12 @@ void FullscreenControllerWidget::centerFSC( int number )
  */
 void FullscreenControllerWidget::showFSC()
 {
-    int number = QApplication::desktop()->screenNumber( p_intf->p_sys->p_mi );
-
-    if( number != i_screennumber ||
-        screenRes != QApplication::desktop()->screenGeometry(number) )
-    {
-        i_screennumber = number;
-        if( !isWideFSC )
-        {
-            centerFSC( number );
-            msg_Dbg( p_intf, "Recentering the Fullscreen Controller" );
-        }
-        else
-        {
-            updateFullwidthGeometry( number );
-        }
-    }
-
-    adjustSize();
+    restoreFSC();
 
 #if HAVE_TRANSPARENCY
     setWindowOpacity( f_opacity );
 #endif
 
-#ifdef Q_WS_X11
-    // Tell kwin that we do not want a shadow around the fscontroller
-    setMask( QRegion( 0, 0, width(), height() ) );
-#endif
-
     show();
 }
 
@@ -888,30 +890,25 @@ void FullscreenControllerWidget::slowHideFSC()
 
 void FullscreenControllerWidget::updateFullwidthGeometry( int number )
 {
-    QRect screenGeometry = QApplication::desktop()->screenGeometry( i_screennumber );
+    QRect screenGeometry = QApplication::desktop()->screenGeometry( number );
     setMinimumWidth( screenGeometry.width() );
     setGeometry( screenGeometry.x(), screenGeometry.y() + screenGeometry.height() - FSC_HEIGHT, screenGeometry.width(), FSC_HEIGHT );
-}
-
-void FullscreenControllerWidget::toggleFullwidth() {
-    if( !isWideFSC ) {
-        /* Dock at the bottom of the screen */
-        updateFullwidthGeometry( i_screennumber );
-    } else {
-        /* Restore half-bar and re-centre */
-        setMinimumWidth( FSC_WIDTH );
-        centerFSC( i_screennumber );
-    }
-
     adjustSize();
+}
 
-#ifdef Q_WS_X11
-    // Update the mask to reflect the geometry change
-    setMask( QRegion( 0, 0, width(), height() ) );
-#endif
-
+void FullscreenControllerWidget::toggleFullwidth()
+{
     /* Toggle isWideFSC switch */
     isWideFSC = !isWideFSC;
+
+    restoreFSC();
+}
+
+int FullscreenControllerWidget::targetScreen()
+{
+    if( i_screennumber == -1 || i_screennumber > QApplication::desktop()->numScreens() )
+        return QApplication::desktop()->screenNumber( p_intf->p_sys->p_mi );
+    return i_screennumber;
 }
 
 /**
@@ -1004,6 +1001,9 @@ void FullscreenControllerWidget::mouseReleaseEvent( QMouseEvent *event )
     i_mouse_last_x = -1;
     i_mouse_last_y = -1;
     event->accept();
+
+    // Save the new FSC position
+    previousPosition = pos();
 }
 
 /**
diff --git a/modules/gui/qt4/components/controller.hpp b/modules/gui/qt4/components/controller.hpp
index bc1aa28..42ce504 100644
--- a/modules/gui/qt4/components/controller.hpp
+++ b/modules/gui/qt4/components/controller.hpp
@@ -261,6 +261,7 @@ public:
     void mouseChanged( vout_thread_t *, int i_mousex, int i_mousey );
     void toggleFullwidth();
     void updateFullwidthGeometry( int number );
+    int targetScreen();
 
 signals:
     void keyPressed( QKeyEvent * );
@@ -285,6 +286,7 @@ private slots:
     void planHideFSC();
     void hideFSC() { hide(); }
     void slowHideFSC();
+    void restoreFSC();
     void centerFSC( int );
 
 private:
@@ -300,6 +302,8 @@ private:
     bool b_mouse_over;
     int i_screennumber;
     QRect screenRes;
+    QRect previousScreenRes;
+    QPoint previousPosition;
 
     /* List of vouts currently tracked */
     QList<vout_thread_t *> vout;
-- 
1.7.8



From etix at videolan.org  Wed Dec 21 14:58:38 2011
From: etix at videolan.org (Ludovic Fauvet)
Date: Wed, 21 Dec 2011 14:58:38 +0100
Subject: [vlc-devel] [PATCH 3/3] Qt: constrain the position of the FSC
	within the screen bounds
In-Reply-To: <1324475918-5987-1-git-send-email-etix@videolan.org>
References: <1324475918-5987-1-git-send-email-etix@videolan.org>
Message-ID: <1324475918-5987-4-git-send-email-etix@videolan.org>

---
 modules/gui/qt4/components/controller.cpp |   30 +++++++++++++++++-----------
 modules/gui/qt4/components/controller.hpp |    2 +-
 2 files changed, 19 insertions(+), 13 deletions(-)

diff --git a/modules/gui/qt4/components/controller.cpp b/modules/gui/qt4/components/controller.cpp
index 3d1bd78..39975e4 100644
--- a/modules/gui/qt4/components/controller.cpp
+++ b/modules/gui/qt4/components/controller.cpp
@@ -707,8 +707,8 @@ FullscreenControllerWidget::FullscreenControllerWidget( intf_thread_t *_p_i, QWi
                            : AbstractController( _p_i, _parent )
 {
     RTL_UNAFFECTED_WIDGET
-    i_mouse_last_x      = -1;
-    i_mouse_last_y      = -1;
+    i_mouse_pos_x      = -1;
+    i_mouse_pos_y      = -1;
     b_mouse_over        = false;
     i_mouse_last_move_x = -1;
     i_mouse_last_move_y = -1;
@@ -970,16 +970,22 @@ void FullscreenControllerWidget::mouseMoveEvent( QMouseEvent *event )
 {
     if( event->buttons() == Qt::LeftButton )
     {
-        if( i_mouse_last_x == -1 || i_mouse_last_y == -1 )
+        if( i_mouse_pos_x == -1 || i_mouse_pos_y == -1 )
             return;
 
-        int i_moveX = event->globalX() - i_mouse_last_x;
-        int i_moveY = event->globalY() - i_mouse_last_y;
+        int i_targetX = event->globalX() - i_mouse_pos_x;
+        int i_targetY = event->globalY() - i_mouse_pos_y;
 
-        move( x() + i_moveX, y() + i_moveY );
+        QRect screen = QApplication::desktop()->screenGeometry( targetScreen() );
+        QRect pos = geometry();
 
-        i_mouse_last_x = event->globalX();
-        i_mouse_last_y = event->globalY();
+        pos.moveTo( i_targetX, i_targetY );
+        if( screen.top()    > pos.top()    )  i_targetY = screen.top();
+        if( screen.bottom() < pos.bottom() )  i_targetY = screen.bottom() - pos.height();
+        if( screen.left()   > pos.left()   )  i_targetX = screen.left();
+        if( screen.right()  < pos.right()  )  i_targetX = screen.right() - pos.width();
+
+        move( i_targetX, i_targetY );
     }
 }
 
@@ -990,16 +996,16 @@ void FullscreenControllerWidget::mouseMoveEvent( QMouseEvent *event )
 void FullscreenControllerWidget::mousePressEvent( QMouseEvent *event )
 {
     if( isWideFSC ) return;
-    i_mouse_last_x = event->globalX();
-    i_mouse_last_y = event->globalY();
+    i_mouse_pos_x = event->x();
+    i_mouse_pos_y = event->y();
     event->accept();
 }
 
 void FullscreenControllerWidget::mouseReleaseEvent( QMouseEvent *event )
 {
     if( isWideFSC ) return;
-    i_mouse_last_x = -1;
-    i_mouse_last_y = -1;
+    i_mouse_pos_x = -1;
+    i_mouse_pos_y = -1;
     event->accept();
 
     // Save the new FSC position
diff --git a/modules/gui/qt4/components/controller.hpp b/modules/gui/qt4/components/controller.hpp
index 42ce504..70d192b 100644
--- a/modules/gui/qt4/components/controller.hpp
+++ b/modules/gui/qt4/components/controller.hpp
@@ -298,7 +298,7 @@ private:
     float f_opacity;
 #endif
 
-    int i_mouse_last_x, i_mouse_last_y;
+    int i_mouse_pos_x, i_mouse_pos_y;
     bool b_mouse_over;
     int i_screennumber;
     QRect screenRes;
-- 
1.7.8



From etix at videolan.org  Wed Dec 21 14:58:35 2011
From: etix at videolan.org (Ludovic Fauvet)
Date: Wed, 21 Dec 2011 14:58:35 +0100
Subject: [vlc-devel] [PATCH 0/3] Qt: various improvements for the fullscreen
	controller
Message-ID: <1324475918-5987-1-git-send-email-etix@videolan.org>

Hello,

Please find a set of three patch for the fullscreen controller:
- The first fixes a regression
- The second is a rework of the code (mostly simplification / logic fix)
- The last one is optional, it constrains the fullscreen controller within the
  screen bounds. The main reason of doing this is that I don't get the point of
  allowing the user to put the controller on another screen since the mouse
  cursor must move on top of the vout to make it appear...

Ludovic Fauvet (3):
  Qt: use the effective size of the widget to center the FSC
  Qt: rework of the fullscreen controller
  Qt: constrain the position of the FSC within the screen bounds

 modules/gui/qt4/components/controller.cpp |  158 +++++++++++++++--------------
 modules/gui/qt4/components/controller.hpp |    6 +-
 2 files changed, 87 insertions(+), 77 deletions(-)

-- 
1.7.8



From etix at videolan.org  Wed Dec 21 14:58:36 2011
From: etix at videolan.org (Ludovic Fauvet)
Date: Wed, 21 Dec 2011 14:58:36 +0100
Subject: [vlc-devel] [PATCH 1/3] Qt: use the effective size of the widget to
	center the FSC
In-Reply-To: <1324475918-5987-1-git-send-email-etix@videolan.org>
References: <1324475918-5987-1-git-send-email-etix@videolan.org>
Message-ID: <1324475918-5987-2-git-send-email-etix@videolan.org>

---
 modules/gui/qt4/components/controller.cpp |    4 ++--
 1 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/modules/gui/qt4/components/controller.cpp b/modules/gui/qt4/components/controller.cpp
index 918615b..9ea1f23 100644
--- a/modules/gui/qt4/components/controller.cpp
+++ b/modules/gui/qt4/components/controller.cpp
@@ -795,8 +795,8 @@ void FullscreenControllerWidget::centerFSC( int number )
     screenRes = QApplication::desktop()->screenGeometry(number);
 
     /* screen has changed, calculate new position */
-    QPoint pos = QPoint( screenRes.x() + (screenRes.width() / 2) - (sizeHint().width() / 2),
-            screenRes.y() + screenRes.height() - sizeHint().height());
+    QPoint pos = QPoint( screenRes.x() + (screenRes.width() / 2) - (width() / 2),
+            screenRes.y() + screenRes.height() - height());
     move( pos );
 }
 
-- 
1.7.8



From ben at europa-network.com  Wed Dec 21 16:26:05 2011
From: ben at europa-network.com (Ben Wheway)
Date: Wed, 21 Dec 2011 16:26:05 +0100
Subject: [vlc-devel] =?utf-8?q?RTSP_Amino=C2=B4s?=
In-Reply-To: <yw1xiplbq6f8.fsf@unicorn.mansr.com>
References: <090601ccbf33$45d55d30$d1801790$@com>
	<yw1xiplbq6f8.fsf@unicorn.mansr.com>
Message-ID: <0a7d01ccbff4$dbb57710$93206530$@com>

Hi Mans

I have seen that a few people have created a workaround with VLC to allow this to work

Is this something you could do for us?

Thanks

Ben



-----Original Message-----
From: vlc-devel-bounces at videolan.org [mailto:vlc-devel-bounces at videolan.org] On Behalf Of M?ns Rullg?rd
Sent: 20 December 2011 22:28
To: vlc-devel at videolan.org
Subject: Re: [vlc-devel] RTSP Amino?s

"Ben Wheway" <ben at europa-network.com> writes:

> Hi All
>
> We currently stream using VLC to Motorola STB?s. We need to do the same with
> our Amino A125 STB?s. But the Amino RTSP protocol doesn?t work with VLC.

Amino uses something almost but not quite entirely unlike RTSP.

-- 
M?ns Rullg?rd
mans at mansr.com

_______________________________________________
vlc-devel mailing list
To unsubscribe or modify your subscription options:
http://mailman.videolan.org/listinfo/vlc-devel



From cheng.sun at ymail.com  Wed Dec 21 17:38:52 2011
From: cheng.sun at ymail.com (Cheng Sun)
Date: Wed, 21 Dec 2011 16:38:52 +0000
Subject: [vlc-devel] [PATCH] Audio channel remapping filter
In-Reply-To: <4EF1BC96.6080800@ymail.com>
References: <4EF1BC96.6080800@ymail.com>
Message-ID: <4EF20B9C.2090809@ymail.com>

On 21/12/11 11:01, Cheng Sun wrote:
> This filter allows channels to be remapped. Potential uses include
> up/downmixing, or to correct non-standard channel orders.
>
> Patch attached. Comments welcome.
>
> -- Cheng
Revised patch, minor formatting changes and corrected usage of var_Inherit

-- Cheng
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-Audio-channel-remapping-filter.patch
Type: text/x-patch
Size: 16871 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111221/f8e8ec9d/attachment.bin>

From cheng.sun at ymail.com  Wed Dec 21 17:59:35 2011
From: cheng.sun at ymail.com (Cheng Sun)
Date: Wed, 21 Dec 2011 16:59:35 +0000
Subject: [vlc-devel] [PATCH] Audio channel remapping filter
In-Reply-To: <4EF20B9C.2090809@ymail.com>
References: <4EF1BC96.6080800@ymail.com> <4EF20B9C.2090809@ymail.com>
Message-ID: <4EF21077.6000600@ymail.com>

On 21/12/11 16:38, Cheng Sun wrote:
> On 21/12/11 11:01, Cheng Sun wrote:
>> This filter allows channels to be remapped. Potential uses include
>> up/downmixing, or to correct non-standard channel orders.
>>
>> Patch attached. Comments welcome.
>>
>> -- Cheng
> Revised patch, minor formatting changes and corrected usage of var_Inherit
>
> -- Cheng
>
Further revised patch. Corrected (un)signedness and size of integers,
updated variable naming, licensed as LGPL, inlining, constness, cosmetics.

-- Cheng
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-Audio-channel-remapping-filter.patch
Type: text/x-patch
Size: 17115 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111221/1c445ac2/attachment.bin>

From cheng.sun at ymail.com  Wed Dec 21 18:03:01 2011
From: cheng.sun at ymail.com (Cheng Sun)
Date: Wed, 21 Dec 2011 17:03:01 +0000
Subject: [vlc-devel] [PATCH] Audio channel remapping filter
In-Reply-To: <4EF21077.6000600@ymail.com>
References: <4EF1BC96.6080800@ymail.com> <4EF20B9C.2090809@ymail.com>
	<4EF21077.6000600@ymail.com>
Message-ID: <4EF21145.6000500@ymail.com>

On 21/12/11 16:59, Cheng Sun wrote:
> On 21/12/11 16:38, Cheng Sun wrote:
>> On 21/12/11 11:01, Cheng Sun wrote:
>>> This filter allows channels to be remapped. Potential uses include
>>> up/downmixing, or to correct non-standard channel orders.
>>>
>>> Patch attached. Comments welcome.
>>>
>>> -- Cheng
>> Revised patch, minor formatting changes and corrected usage of var_Inherit
>>
>> -- Cheng
>>
> Further revised patch. Corrected (un)signedness and size of integers,
> updated variable naming, licensed as LGPL, inlining, constness, cosmetics.
>
> -- Cheng
>
Missed a spot.

-- Cheng
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-Audio-channel-remapping-filter.patch
Type: text/x-patch
Size: 17119 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111221/19669c61/attachment.bin>

From funman at videolan.org  Wed Dec 21 18:15:28 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Wed, 21 Dec 2011 12:15:28 -0500
Subject: [vlc-devel] [PATCH] messages.c: added logging via
 OutputDebugString under windows.
In-Reply-To: <4EF1C711.4090102@asokolov.org>
References: <1324436973-30963-1-git-send-email-funman@videolan.org>
	<573ec901bb61a887e6c7678bc4d8f856@chewa.net>
	<20111221041619.61b02133@tok> <4EF1C711.4090102@asokolov.org>
Message-ID: <20111221121528.5dfaf6eb@tok>

Le Wed, 21 Dec 2011 18:46:25 +0700,
Alexey Sokolov <alexey at asokolov.org> a ?crit :

> 21.12.2011 16:16, Rafa?l Carr? ?????:
> > I will let Sergey send another patch. Btw this was Sergey patch, I'm
> > not sure how I should use git send-email with a patch of someone else
> Just let people send their patches themselves...

This was my modified version

-- 
Rafa?l Carr?
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111221/61055f28/attachment.sig>

From funman at videolan.org  Wed Dec 21 18:18:10 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Wed, 21 Dec 2011 12:18:10 -0500
Subject: [vlc-devel] [PATCH] messages.c: added logging via
 OutputDebugString under windows.
In-Reply-To: <b8590c75149b20abd851024429f74719@chewa.net>
References: <1324436973-30963-1-git-send-email-funman@videolan.org>
	<573ec901bb61a887e6c7678bc4d8f856@chewa.net>
	<20111221041619.61b02133@tok>
	<b8590c75149b20abd851024429f74719@chewa.net>
Message-ID: <20111221121810.00c5711d@tok>

Le Wed, 21 Dec 2011 12:03:19 +0100,
R?mi Denis-Courmont <remi at remlab.net> a ?crit :

> On Wed, 21 Dec 2011 04:16:19 -0500, Rafa?l Carr? <funman at videolan.org>
> > Not yet, vlc exited before calling msg_* (or did it dereference NULL and
> 
> > exited without segfaulting?)
> 
> Sending untested patches for review is wasting other people's time.
> 
> As a retalation, I will not be reviewing this pathc series anymore.

ok

-- 
Rafa?l Carr?


From funman at videolan.org  Wed Dec 21 18:21:11 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Wed, 21 Dec 2011 12:21:11 -0500
Subject: [vlc-devel] [PATCH] messages.c: added logging via
 OutputDebugString under windows.
In-Reply-To: <4EF1D862.3070906@web.de>
References: <1324436973-30963-1-git-send-email-funman@videolan.org>
	<4EF1D862.3070906@web.de>
Message-ID: <20111221122111.38661071@tok>

Le Wed, 21 Dec 2011 14:00:18 +0100,
Tobias G?ntner <fatbull at web.de> a ?crit :

> Am 21.12.2011 04:09, schrieb Sergey Radionov:
> > Signed-off-by: Rafa?l Carr?<funman at videolan.org>
> 
> > +#ifdef WIN32
> > +    va_list dol;
> > +    va_copy (dol, args);
> > +    Win32DebugOutputMsg (0, type,&msg, format, dol);
> > +    va_end (dol);
> > +#endif
> 
> Maybe test IsDebuggerPresent() as well?

I don't know about this part, I let Sergey answer

> > +    size_t fmt_len = strlen(format);
> > +    char *fmt_cr = malloc(fmt_len + 1 + 1);
> > +    if (!fmt_cr)
> > +        return;
> > +
> > +    strcpy(fmt_cr, format);
> > +    fmt_cr[fmt_len] = '\n';
> > +    fmt_cr[fmt_len + 1] = '\0';
> 
> Perhaps asprintf would be easier.

Yes

> 
> > +    char *msg;
> > +    if (vasprintf(&msg, fmt_cr, dol))
> > +        msg = NULL;
> 
> vasprintf returns -1 on error.

Yes

Thanks

-- 
Rafa?l Carr?


From cheng.sun at ymail.com  Wed Dec 21 18:25:25 2011
From: cheng.sun at ymail.com (Cheng Sun)
Date: Wed, 21 Dec 2011 17:25:25 +0000
Subject: [vlc-devel] [PATCH] Audio channel remapping filter
In-Reply-To: <4EF21077.6000600@ymail.com>
References: <4EF1BC96.6080800@ymail.com> <4EF20B9C.2090809@ymail.com>
	<4EF21077.6000600@ymail.com>
Message-ID: <4EF21685.80207@ymail.com>

On 21/12/11 16:59, Cheng Sun wrote:
> On 21/12/11 16:38, Cheng Sun wrote:
>> On 21/12/11 11:01, Cheng Sun wrote:
>>> This filter allows channels to be remapped. Potential uses include
>>> up/downmixing, or to correct non-standard channel orders.
>>>
>>> Patch attached. Comments welcome.
>>>
>>> -- Cheng
>> Revised patch, minor formatting changes and corrected usage of var_Inherit
>>
>> -- Cheng
>>
> Further revised patch. Corrected (un)signedness and size of integers,
> updated variable naming, licensed as LGPL, inlining, constness, cosmetics.
>
> -- Cheng
>
Save one (1) byte of memory.

-- Cheng
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-Audio-channel-remapping-filter.patch
Type: text/x-patch
Size: 17119 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111221/8e76da8d/attachment.bin>

From jan.scheers at gmail.com  Wed Dec 21 18:26:58 2011
From: jan.scheers at gmail.com (Jan Scheers)
Date: Wed, 21 Dec 2011 18:26:58 +0100
Subject: [vlc-devel] [PATCH] gui: Replace all spaces with "%20" in URLs
Message-ID: <1324488418-21119-1-git-send-email-jan.scheers@gmail.com>

fix #3435

---
 modules/gui/macosx/open.m                  |    1 +
 modules/gui/qt4/components/open_panels.cpp |   22 ++++++----------------
 modules/gui/qt4/components/open_panels.hpp |    9 ---------
 3 files changed, 7 insertions(+), 25 deletions(-)

diff --git a/modules/gui/macosx/open.m b/modules/gui/macosx/open.m
index 5c5fb0c..df8368f 100644
--- a/modules/gui/macosx/open.m
+++ b/modules/gui/macosx/open.m
@@ -1040,6 +1040,7 @@ static VLCOpen *_o_sharedMainInstance = nil;
     else
     {
         o_mrl_string = [o_net_http_url stringValue];
+        o_mrl_string = [o_mrl_string stringByReplacingOccurrencesOfString:@" " withString:@"%20"];
     }
     [self setMRL: o_mrl_string];
 }
diff --git a/modules/gui/qt4/components/open_panels.cpp b/modules/gui/qt4/components/open_panels.cpp
index 78bb4b7..6a1f77d 100644
--- a/modules/gui/qt4/components/open_panels.cpp
+++ b/modules/gui/qt4/components/open_panels.cpp
@@ -601,8 +601,6 @@ NetOpenPanel::NetOpenPanel( QWidget *_parent, intf_thread_t *_p_intf ) :
     else
         b_recentList = false;
 
-    /* Use a simple validator for URLs */
-    ui.urlComboBox->setValidator( new UrlValidator( this ) );
     ui.urlComboBox->setFocus();
 }
 
@@ -642,6 +640,12 @@ void NetOpenPanel::updateMRL()
 {
     QString url = ui.urlComboBox->lineEdit()->text();
 
+    int i;
+    while( (i = url.indexOf( ' ' )) != -1 )
+    {
+        url.replace( i, 1, "%20" );
+    };
+
     if( url.isEmpty() )
         return;
 
@@ -652,20 +656,6 @@ void NetOpenPanel::updateMRL()
     emit mrlUpdated( qsl, "" );
 }
 
-QValidator::State UrlValidator::validate( QString& str, int& ) const
-{
-    if( str.contains( ' ' ) )
-        return QValidator::Invalid;
-    if( !str.contains( "://" ) )
-        return QValidator::Intermediate;
-    return QValidator::Acceptable;
-}
-
-void UrlValidator::fixup( QString& str ) const
-{
-    str = str.trimmed();
-}
-
 /**************************************************************************
  * Open Capture device ( DVB, PVR, V4L, and similar )                     *
  **************************************************************************/
diff --git a/modules/gui/qt4/components/open_panels.hpp b/modules/gui/qt4/components/open_panels.hpp
index 85257ce..0f7c216 100644
--- a/modules/gui/qt4/components/open_panels.hpp
+++ b/modules/gui/qt4/components/open_panels.hpp
@@ -149,15 +149,6 @@ public slots:
     virtual void updateMRL();
 };
 
-class UrlValidator : public QValidator
-{
-   Q_OBJECT
-public:
-   UrlValidator( QObject *parent ) : QValidator( parent ) { }
-   void fixup( QString& ) const;
-   QValidator::State validate( QString&, int& ) const;
-};
-
 class DiscOpenPanel: public OpenPanel
 {
     Q_OBJECT
-- 
1.7.8



From remi at remlab.net  Wed Dec 21 18:47:28 2011
From: remi at remlab.net (=?iso-8859-15?q?R=E9mi?= Denis-Courmont)
Date: Wed, 21 Dec 2011 19:47:28 +0200
Subject: [vlc-devel] [PATCH] Audio channel remapping filter
In-Reply-To: <4EF21685.80207@ymail.com>
References: <4EF1BC96.6080800@ymail.com> <4EF21077.6000600@ymail.com>
	<4EF21685.80207@ymail.com>
Message-ID: <201112211947.29144.remi@remlab.net>

diff --git a/modules/audio_filter/Modules.am b/modules/audio_filter/Modules.am
index 8f9e5ef..d787dbb 100644
--- a/modules/audio_filter/Modules.am
+++ b/modules/audio_filter/Modules.am
@@ -31,11 +31,13 @@ SOURCES_simple_channel_mixer = channel_mixer/simple.c
 SOURCES_headphone_channel_mixer = channel_mixer/headphone.c
 SOURCES_dolby_surround_decoder = channel_mixer/dolby.c
 SOURCES_mono = channel_mixer/mono.c
+SOURCES_remap = channel_mixer/remap.c
 
 libvlc_LTLIBRARIES += \
 	libdolby_surround_decoder_plugin.la \
 	libheadphone_channel_mixer_plugin.la \
 	libmono_plugin.la \
+	libremap_plugin.la \
 	libsimple_channel_mixer_plugin.la \
 	libtrivial_channel_mixer_plugin.la
 
diff --git a/modules/audio_filter/channel_mixer/remap.c 
b/modules/audio_filter/channel_mixer/remap.c
new file mode 100644
index 0000000..dcdefe2
--- /dev/null
+++ b/modules/audio_filter/channel_mixer/remap.c
@@ -0,0 +1,422 @@
+/*****************************************************************************
+ * remap.c : simple channel remapper plug-in
+ 
*****************************************************************************
+ * Copyright (C) 2011 the VideoLAN team
+ *
+ * Authors: Cheng Sun <chengsun9 at gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU Lesser General Public License as published by
+ * the Free Software Foundation; either version 2.1 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public License
+ * along with this program; if not, write to the Free Software Foundation,
+ * Inc., 51 Franklin Street, Fifth Floor, Boston MA 02110-1301, USA.
+ 
*****************************************************************************/
+
+/*****************************************************************************
+ * Preamble
+ 
*****************************************************************************/
+#ifdef HAVE_CONFIG_H
+# include "config.h"
+#endif
+
+#include <vlc_common.h>
+#include <vlc_plugin.h>
+#include <vlc_aout.h>
+#include <vlc_filter.h>
+#include <vlc_block.h>
+#include <assert.h>
+
+/*****************************************************************************
+ * Module descriptor
+ 
*****************************************************************************/
+static int  OpenFilter( vlc_object_t * );
+static void CloseFilter( vlc_object_t * );
+
+#define REMAP_CFG "aout-remap-"
+

Where do those magical values come from?

+static const uint8_t channel_wg4idx[] =
+{
+    0, 7, 1,
+    4, 6, 5,
+    2, 3, 8
+};
+static const int channel_wg4idx_len = sizeof( channel_wg4idx )/sizeof( 
channel_wg4idx[0] );

It should be more efficient to expand the value directly where needed.

+static const uint8_t channel_idx[] =
+{
+    0, 1, 2,
+    3, 4, 5,
+    6, 7, 8
+};

It would be clearer that this is simply [i] = i on a single line.

+
+static const char *const channel_name[] =
+{
+    REMAP_CFG "channel-left", REMAP_CFG "channel-center", REMAP_CFG "channel-
right",
+    REMAP_CFG "channel-rearleft", REMAP_CFG "channel-rearcenter", REMAP_CFG 
"channel-rearright",
+    REMAP_CFG "channel-middleleft", REMAP_CFG "channel-middleright", 
REMAP_CFG "channel-lfe"
+};

Too long lines.

+
+static const char *const channel_desc[] =
+{
+    N_( "Left" ), N_( "Center" ), N_( "Right" ),
+    N_( "Rear left" ), N_( "Rear center" ), N_( "Rear right" ),
+    N_( "Side left" ), N_( "Side right" ), N_( "Low-frequency effects" )
+};
+
+static const int channel_flag[] =
+{
+    AOUT_CHAN_LEFT, AOUT_CHAN_CENTER, AOUT_CHAN_RIGHT,
+    AOUT_CHAN_REARLEFT, AOUT_CHAN_REARCENTER, AOUT_CHAN_REARRIGHT,
+    AOUT_CHAN_MIDDLELEFT, AOUT_CHAN_MIDDLERIGHT, AOUT_CHAN_LFE
+};
+
+vlc_module_begin ()
+    set_description( N_("Audio channel remapper") )
+    set_capability( "audio filter", 0 )
+    set_category( CAT_AUDIO )
+    set_subcategory( SUBCAT_AUDIO_AFILTER )
+    set_callbacks( OpenFilter, CloseFilter )
+    set_shortname( "Remap" )
+
+#define CHANNEL( idx ) \
+    add_integer( channel_name[idx], idx, channel_desc[idx], \
+            channel_desc[idx], false) \
+        change_integer_list( channel_idx, channel_desc )
+    CHANNEL(0) CHANNEL(1) CHANNEL(2)
+    CHANNEL(3) CHANNEL(4) CHANNEL(5)
+    CHANNEL(6) CHANNEL(7) CHANNEL(8)
+#undef CHANNEL
+
+    add_bool( REMAP_CFG "normalise", true, "Normalise channels",

Please use American English for the C locale. You can use British spelling in 
en_GB.po if you so fancy.

+            "When mapping more than one channel to a single output channel, "
+            "normalise the output accordingly", false )

Missing full point.

+
+    set_callbacks( OpenFilter, CloseFilter )
+
+vlc_module_end ()
+
+/*****************************************************************************
+ * Local prototypes
+ 
*****************************************************************************/
+
+static block_t *Remap( filter_t *, block_t * );
+
+typedef void (*remap_fun_t)( filter_t *, filter_sys_t *, const void *, void 
*,
+                             int, int, unsigned, unsigned);
+
+struct filter_sys_t
+{
+    remap_fun_t pf_remap;
+    int nb_in_ch[AOUT_CHAN_MAX];
+    uint8_t map_ch[AOUT_CHAN_MAX];
+    bool b_normalise;
+};
+
+static inline uint8_t IndexOf(const uint8_t *arr, const size_t arrlen, 
uint8_t elem)
+{
+    for( size_t i = 0; i < arrlen; i++ )
+    {
+        if( arr[i] == elem )
+            return i;
+    }
+    return arrlen;
+}

You reinvented memchr() or rather index() here.

+static const uint32_t valid_channels[] = {
+/* list taken from aout_FormatPrintChannels */
+    AOUT_CHAN_LEFT,
+    AOUT_CHAN_RIGHT,
+    AOUT_CHAN_CENTER,
+    AOUT_CHAN_LEFT | AOUT_CHAN_RIGHT,
+    AOUT_CHAN_LEFT | AOUT_CHAN_RIGHT | AOUT_CHAN_CENTER,
+    AOUT_CHAN_LEFT | AOUT_CHAN_RIGHT | AOUT_CHAN_REARCENTER,
+    AOUT_CHAN_LEFT | AOUT_CHAN_RIGHT | AOUT_CHAN_CENTER
+          | AOUT_CHAN_REARCENTER,
+    AOUT_CHAN_LEFT | AOUT_CHAN_RIGHT
+          | AOUT_CHAN_REARLEFT | AOUT_CHAN_REARRIGHT,
+    AOUT_CHAN_LEFT | AOUT_CHAN_RIGHT
+          | AOUT_CHAN_MIDDLELEFT | AOUT_CHAN_MIDDLERIGHT,
+    AOUT_CHAN_LEFT | AOUT_CHAN_RIGHT | AOUT_CHAN_CENTER
+          | AOUT_CHAN_REARLEFT | AOUT_CHAN_REARRIGHT,
+    AOUT_CHAN_LEFT | AOUT_CHAN_RIGHT | AOUT_CHAN_CENTER
+          | AOUT_CHAN_MIDDLELEFT | AOUT_CHAN_MIDDLERIGHT,
+    AOUT_CHAN_CENTER | AOUT_CHAN_LFE,
+    AOUT_CHAN_LEFT | AOUT_CHAN_RIGHT | AOUT_CHAN_LFE,
+    AOUT_CHAN_LEFT | AOUT_CHAN_RIGHT | AOUT_CHAN_CENTER | AOUT_CHAN_LFE,
+    AOUT_CHAN_LEFT | AOUT_CHAN_RIGHT | AOUT_CHAN_REARCENTER
+          | AOUT_CHAN_LFE,
+    AOUT_CHAN_LEFT | AOUT_CHAN_RIGHT | AOUT_CHAN_CENTER
+          | AOUT_CHAN_REARCENTER | AOUT_CHAN_LFE,
+    AOUT_CHAN_LEFT | AOUT_CHAN_RIGHT
+          | AOUT_CHAN_REARLEFT | AOUT_CHAN_REARRIGHT | AOUT_CHAN_LFE,
+    AOUT_CHAN_LEFT | AOUT_CHAN_RIGHT
+          | AOUT_CHAN_MIDDLELEFT | AOUT_CHAN_MIDDLERIGHT | AOUT_CHAN_LFE,
+    AOUT_CHAN_LEFT | AOUT_CHAN_RIGHT | AOUT_CHAN_CENTER
+          | AOUT_CHAN_REARLEFT | AOUT_CHAN_REARRIGHT | AOUT_CHAN_LFE,
+    AOUT_CHAN_LEFT | AOUT_CHAN_RIGHT | AOUT_CHAN_CENTER
+          | AOUT_CHAN_MIDDLELEFT | AOUT_CHAN_MIDDLERIGHT | AOUT_CHAN_LFE,
+    AOUT_CHAN_LEFT | AOUT_CHAN_RIGHT | AOUT_CHAN_CENTER
+          | AOUT_CHAN_REARLEFT | AOUT_CHAN_REARRIGHT | AOUT_CHAN_MIDDLELEFT
+          | AOUT_CHAN_MIDDLERIGHT,
+    AOUT_CHAN_LEFT | AOUT_CHAN_RIGHT | AOUT_CHAN_CENTER
+          | AOUT_CHAN_REARLEFT | AOUT_CHAN_REARRIGHT | AOUT_CHAN_MIDDLELEFT
+          | AOUT_CHAN_MIDDLERIGHT | AOUT_CHAN_LFE,
+};
+
+static inline uint32_t CanonicaliseChannels( uint32_t i_physical_channels )
+{
+    for( unsigned i = 0; i < sizeof( valid_channels )/sizeof( 
valid_channels[0] ); i++ )
+        if( (i_physical_channels & ~valid_channels[i]) == 0 )
+            return valid_channels[i];
+
+    assert( false );
+    return 0;
+}
+
+/*****************************************************************************
+ * DoWork: convert a buffer
+ 
*****************************************************************************/
+static void RemapCopy( filter_t *p_filter, filter_sys_t *p_sys,
+                    const void *p_srcorig, void *p_destorig,
+                    int i_nb_samples, int i_bytespersample,
+                    unsigned i_nb_in_channels, unsigned i_nb_out_channels )
+{
+    VLC_UNUSED( p_filter );
+    const char *p_src = p_srcorig;
+    char *p_dest = p_destorig;
+
+    memset( p_dest, 0, i_bytespersample * i_nb_samples * i_nb_out_channels );
+
+    for( int i = 0; i < i_nb_samples; i++ )
+    {
+        for( uint8_t in_ch = 0; in_ch < i_nb_in_channels; in_ch++ )
+        {
+            uint8_t out_ch = p_sys->map_ch[ in_ch ];
+            memcpy( p_dest + out_ch * i_bytespersample,
+                    p_src  + in_ch  * i_bytespersample,
+                    i_bytespersample );
+        }
+        p_src  += i_nb_in_channels  * i_bytespersample;
+        p_dest += i_nb_out_channels * i_bytespersample;
+    }
+}
+
+#define DEFINE_REMAP_ADD( name, type ) \
+static void RemapAdd##name( filter_t *p_filter, filter_sys_t *p_sys, \
+                    const void *p_srcorig, void *p_destorig, \
+                    int i_nb_samples, int i_bytespersample, \
+                    unsigned i_nb_in_channels, unsigned i_nb_out_channels ) \
+{ \
+    VLC_UNUSED( p_filter ); VLC_UNUSED( i_bytespersample ); \
+    const type *p_src = p_srcorig; \
+    type *p_dest = p_destorig; \
+ \
+    assert( sizeof( type ) == i_bytespersample ); \
+ \
+    memset( p_dest, 0, i_bytespersample * i_nb_samples * i_nb_out_channels ); 
\
+ \
+    for( int i = 0; i < i_nb_samples; i++ ) \
+    { \
+        for( uint8_t in_ch = 0; in_ch < i_nb_in_channels; in_ch++ ) \
+        { \
+            uint8_t out_ch = p_sys->map_ch[ in_ch ]; \
+            if( p_sys->b_normalise ) \
+                p_dest[ out_ch ] += p_src[ in_ch ] / p_sys->nb_in_ch[ out_ch 
]; \
+            else \
+                p_dest[ out_ch ] += p_src[ in_ch ]; \
+        } \
+        p_src  += i_nb_in_channels; \
+        p_dest += i_nb_out_channels; \
+    } \
+}
+
+DEFINE_REMAP_ADD( S16N, int16_t  )
+DEFINE_REMAP_ADD( U16N, uint16_t )
+DEFINE_REMAP_ADD( S32N, int32_t  )
+DEFINE_REMAP_ADD( FL32, float    )
+DEFINE_REMAP_ADD( FL64, double   )
+
+static inline remap_fun_t GetRemapFun( audio_format_t *p_format, bool b_add )
+{
+    if( b_add )
+    {
+        switch( p_format->i_format )
+        {
+            case VLC_CODEC_S16N:
+                return RemapAddS16N;
+                break;
+            case VLC_CODEC_U16N:
+                return RemapAddU16N;
+                break;
+            case VLC_CODEC_S32N:
+                return RemapAddS32N;
+                break;
+            case VLC_CODEC_FL32:
+                return RemapAddFL32;
+                break;
+            case VLC_CODEC_FL64:
+                return RemapAddFL64;
+                break;
+        }
+    }
+    else
+    {
+        return RemapCopy;
+    }
+
+    return NULL;
+}
+
+
+/*****************************************************************************
+ * OpenFilter:
+ 
*****************************************************************************/
+static int OpenFilter( vlc_object_t *p_this )
+{
+    filter_t *p_filter = (filter_t *)p_this;
+    filter_sys_t *p_sys;
+
+    audio_format_t *audio_in  = &p_filter->fmt_in.audio;
+    audio_format_t *audio_out = &p_filter->fmt_out.audio;
+
+    if( ( audio_in->i_format != audio_out->i_format ) ||
+        ( audio_in->i_rate != audio_in->i_rate) ||

WTF?

+        ( audio_in->i_bitspersample != audio_in->i_bitspersample) ||

WTF?

+        ( audio_in->i_bitspersample % 8 != 0 ) )
+        return VLC_EGENERIC;
+
+    /* Allocate the memory needed to store the module's structure */
+    p_sys = p_filter->p_sys = malloc( sizeof(filter_sys_t) );
+    if( p_sys == NULL )

Missing unlikely().

+        return VLC_ENOMEM;
+
+    /* get number of and layout of input channels */
+    uint32_t i_output_physical = 0;
+    uint8_t pi_map_ch[ AOUT_CHAN_MAX ] = { 0 }; /* which out channel each in 
channel is mapped to */
+    p_sys->b_normalise = var_InheritBool( p_this, REMAP_CFG "normalise" );
+
+    for( uint8_t in_ch = 0, wg4_i = 0; in_ch < audio_in->i_channels; in_ch++, 
wg4_i++ )
+    {
+        /* explode in_channels in the right order */
+        while( ( audio_in->i_physical_channels & pi_vlc_chan_order_wg4[ wg4_i 
] ) == 0 )
+        {
+            wg4_i++;
+            assert( wg4_i < sizeof( pi_vlc_chan_order_wg4 )/sizeof( 
pi_vlc_chan_order_wg4[0] ) );
+        }
+        uint8_t chnidx = IndexOf( channel_wg4idx, channel_wg4idx_len, wg4_i 
);
+        assert( chnidx != channel_wg4idx_len );
+        uint8_t out_idx = var_InheritInteger( p_this, channel_name[chnidx] );
+        pi_map_ch[in_ch] = channel_wg4idx[ out_idx ];
+
+        i_output_physical |= channel_flag[ out_idx ];
+    }
+    i_output_physical = CanonicaliseChannels( i_output_physical );
+
+    audio_out->i_physical_channels = i_output_physical;
+    audio_out->i_channels = aout_FormatNbChannels( audio_out );

No. You should call the format preparation helper instead.

+    audio_out->i_rate = audio_in->i_rate;
+    audio_out->i_format = p_filter->fmt_out.i_codec;
+
+    /* condense out_channels */
+    uint8_t out_ch_sorted[ AOUT_CHAN_MAX ];
+    for( uint8_t i = 0, wg4_i = 0; i < audio_out->i_channels; i++, wg4_i++ )
+    {
+        while( ( audio_out->i_physical_channels & pi_vlc_chan_order_wg4[ 
wg4_i ] ) == 0 )
+        {
+            wg4_i++;
+            assert( wg4_i < sizeof( pi_vlc_chan_order_wg4 )/sizeof( 
pi_vlc_chan_order_wg4[0] ) );
+        }
+        out_ch_sorted[ i ] = wg4_i;
+    }
+    bool b_multiple = false; /* whether we need to add channels (multiple in 
mapped to an out) */
+    memset( p_sys->nb_in_ch, 0, sizeof( p_sys->nb_in_ch ) );
+    for( uint8_t i = 0; i < audio_in->i_channels; i++ )
+    {
+        uint8_t wg4_out_ch = pi_map_ch[i];
+        p_sys->map_ch[i] = IndexOf( out_ch_sorted, audio_out->i_channels, 
wg4_out_ch );
+        assert( p_sys->map_ch[i] != audio_out->i_channels );
+        if( ++p_sys->nb_in_ch[ p_sys->map_ch[i] ] > 1 )
+            b_multiple = true;
+    }
+
+    msg_Dbg( p_filter, "%s '%4.4s'->'%4.4s' %d Hz->%d Hz %s->%s",
+             "Remap filter",
+             (char *)&audio_in->i_format, (char *)&audio_out->i_format,
+             audio_in->i_rate, audio_out->i_rate,
+             aout_FormatPrintChannels( audio_in ),
+             aout_FormatPrintChannels( audio_out ) );
+
+    p_filter->pf_audio_filter = Remap;
+    p_sys->pf_remap = GetRemapFun( audio_in, b_multiple );
+    if( !p_sys->pf_remap )
+    {
+        msg_Err( p_filter, "Could not decide on a remap function (b_multiple 
= %d)", b_multiple );

Please don't use Cism in error messages.

+        free( p_sys );
+        return VLC_EGENERIC;
+    }
+
+    return 0;
+}
+
+/*****************************************************************************
+ * CloseFilter:
+ 
*****************************************************************************/
+static void CloseFilter( vlc_object_t *p_this )
+{
+    filter_t *p_filter = (filter_t *) p_this;
+    filter_sys_t *p_sys = p_filter->p_sys;
+    free( p_sys );
+}
+
+/*****************************************************************************
+ * Remap:
+ 
*****************************************************************************/
+static block_t *Remap( filter_t *p_filter, block_t *p_block )
+{
+    filter_sys_t *p_sys = (filter_sys_t *)p_filter->p_sys;
+    if( !p_block || !p_block->i_nb_samples )
+    {
+        if( p_block )
+            block_Release( p_block );
+        return NULL;
+    }
+
+    size_t i_out_size = p_block->i_nb_samples *
+        p_filter->fmt_out.audio.i_channels *
+        p_filter->fmt_out.audio.i_bitspersample / 8;

Use i_bytes_per_frames.

+
+    block_t *p_out = filter_NewAudioBuffer( p_filter, i_out_size );
+    if( !p_out )
+    {
+        msg_Warn( p_filter, "can't get output buffer" );
+        block_Release( p_block );
+        return NULL;
+    }
+
+    p_out->i_nb_samples = p_block->i_nb_samples;
+    p_out->i_dts = p_block->i_dts;
+    p_out->i_pts = p_block->i_pts;
+    p_out->i_length = p_block->i_length;
+
+    const int i_nb_samples = p_block->i_nb_samples;
+    const int i_bytespersample = p_filter->fmt_in.audio.i_bitspersample/8;
+    unsigned i_nb_in_channels = p_filter->fmt_in.audio.i_channels;
+    unsigned i_nb_out_channels = p_filter->fmt_out.audio.i_channels;
+
+    p_sys->pf_remap( p_filter, (filter_sys_t *)p_sys,
+                (const char *)p_block->p_buffer, (char *)p_out->p_buffer,
+                i_nb_samples, i_bytespersample,
+                i_nb_in_channels, i_nb_out_channels );

For performance reasons, it would be best to drop i_bytespersample and use one 
function per sample format, or at least per sample size. You already do so in 
the normalized case anyway.

+
+    block_Release( p_block );
+
+    return p_out;
+}

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From remi at remlab.net  Wed Dec 21 18:48:54 2011
From: remi at remlab.net (=?iso-8859-1?q?R=E9mi?= Denis-Courmont)
Date: Wed, 21 Dec 2011 19:48:54 +0200
Subject: [vlc-devel] [PATCH] gui: Replace all spaces with "%20" in URLs
In-Reply-To: <1324488418-21119-1-git-send-email-jan.scheers@gmail.com>
References: <1324488418-21119-1-git-send-email-jan.scheers@gmail.com>
Message-ID: <201112211948.54803.remi@remlab.net>

	Hello,

Le mercredi 21 d?cembre 2011 19:26:58 Jan Scheers, vous avez ?crit :
> fix #3435

If I see where this is going, you should probably replace all URL-unsafe 
characters, not just white spaces.

It might make sense to write a helper function.

-- 
R?mi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis


From jb at videolan.org  Wed Dec 21 21:35:29 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Wed, 21 Dec 2011 21:35:29 +0100
Subject: [vlc-devel] [PATCH 1/3] Qt: use the effective size of the
 widget to center the FSC
In-Reply-To: <1324475918-5987-2-git-send-email-etix@videolan.org>
References: <1324475918-5987-1-git-send-email-etix@videolan.org>
	<1324475918-5987-2-git-send-email-etix@videolan.org>
Message-ID: <20111221203529.GA27964@videolan.org>

On Wed, Dec 21, 2011 at 02:58:36PM +0100, Ludovic Fauvet wrote :
>  modules/gui/qt4/components/controller.cpp |    4 ++--

Applied.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Wed Dec 21 21:35:37 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Wed, 21 Dec 2011 21:35:37 +0100
Subject: [vlc-devel] [PATCH 2/3] Qt: rework of the fullscreen controller
In-Reply-To: <1324475918-5987-3-git-send-email-etix@videolan.org>
References: <1324475918-5987-1-git-send-email-etix@videolan.org>
	<1324475918-5987-3-git-send-email-etix@videolan.org>
Message-ID: <20111221203537.GB27964@videolan.org>

On Wed, Dec 21, 2011 at 02:58:37PM +0100, Ludovic Fauvet wrote :
> - fix the logic
> - simplify
> - repair the ability to save the state and position

Applied.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Wed Dec 21 21:35:49 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Wed, 21 Dec 2011 21:35:49 +0100
Subject: [vlc-devel] [PATCH] Qt4: add title to file associations dialog
In-Reply-To: <4EF1CF43.2080009@compdigitec.com>
References: <4EF1CF43.2080009@compdigitec.com>
Message-ID: <20111221203549.GC27964@videolan.org>

On Wed, Dec 21, 2011 at 07:21:23AM -0500, Edward Wang wrote :
> Currently, when you click the button to set up file associations on
> Windows, the dialog presented is untitled and just displays the file
> name, "vlc". The attached patch will add add title to file
> associations dialog (see http://trac.videolan.org/vlc/ticket/5711).

Applied, thanks

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Wed Dec 21 21:37:53 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Wed, 21 Dec 2011 21:37:53 +0100
Subject: [vlc-devel] [PATCH] yadif : Add SSSE3 and SSE2 support. porting
 from FFmpeg.
In-Reply-To: <CAOa_K+tO9B6VmKYBdRh2t6ASis4_wTPO8QQVbYtdpobiOoq+wQ@mail.gmail.com>
References: <CAOa_K+vdRUzBj4cE9qoLyU19+NYnVuPx03UqdOEXgkT6fw60Gg@mail.gmail.com>
	<201112091941.15730.remi@remlab.net>
	<CAOa_K+shj1Wy4pm3=sFZxw4_nM+k-VCZnRauiQujeV0+LS4CGw@mail.gmail.com>
	<201112171042.02575.remi@remlab.net>
	<CAOa_K+tBjZCS5sSve_vwCt4DbzX2qpJGa37tTiZkX-tKOaUQgA@mail.gmail.com>
	<CAOa_K+ug10ymVFtCeTTHo=TwzOzynfiON9WE-Ci4L_A_x45WzQ@mail.gmail.com>
	<20111221012523.GA1174@videolan.org>
	<CAOa_K+sPqxzZY5DFimC+Ri0-jbEW5pTGLCvDcVkQyTTo1OuCRw@mail.gmail.com>
	<CAOa_K+tO9B6VmKYBdRh2t6ASis4_wTPO8QQVbYtdpobiOoq+wQ@mail.gmail.com>
Message-ID: <20111221203753.GA28406@videolan.org>

On Wed, Dec 21, 2011 at 05:14:26PM +0900, Naohiro KORIYAMA wrote :
> I forget the attachment...
And that's the way we like it :D

Thanks a lot.

Btw, did you look at the 10bits yadif patch too?

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From cheng.sun at ymail.com  Wed Dec 21 21:57:33 2011
From: cheng.sun at ymail.com (Cheng Sun)
Date: Wed, 21 Dec 2011 20:57:33 +0000
Subject: [vlc-devel] [PATCH] Audio channel remapping filter
In-Reply-To: <201112211947.29144.remi@remlab.net>
References: <4EF1BC96.6080800@ymail.com> <4EF21077.6000600@ymail.com>
	<4EF21685.80207@ymail.com> <201112211947.29144.remi@remlab.net>
Message-ID: <4EF2483D.2070808@ymail.com>

On 21/12/11 17:47, R?mi Denis-Courmont wrote:
> WTF?
Nicely summed up :)

Revised patch attached, with your suggestions incorporated.

Cheers,
-- Cheng
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-Audio-channel-remapping-filter.patch
Type: text/x-patch
Size: 16701 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111221/2af608ab/attachment.bin>

From nkoriyama at gmail.com  Wed Dec 21 22:28:58 2011
From: nkoriyama at gmail.com (Naohiro KORIYAMA)
Date: Thu, 22 Dec 2011 06:28:58 +0900
Subject: [vlc-devel] [PATCH] yadif : Add SSSE3 and SSE2 support. porting
 from FFmpeg.
In-Reply-To: <20111221203753.GA28406@videolan.org>
References: <CAOa_K+vdRUzBj4cE9qoLyU19+NYnVuPx03UqdOEXgkT6fw60Gg@mail.gmail.com>
	<201112091941.15730.remi@remlab.net>
	<CAOa_K+shj1Wy4pm3=sFZxw4_nM+k-VCZnRauiQujeV0+LS4CGw@mail.gmail.com>
	<201112171042.02575.remi@remlab.net>
	<CAOa_K+tBjZCS5sSve_vwCt4DbzX2qpJGa37tTiZkX-tKOaUQgA@mail.gmail.com>
	<CAOa_K+ug10ymVFtCeTTHo=TwzOzynfiON9WE-Ci4L_A_x45WzQ@mail.gmail.com>
	<20111221012523.GA1174@videolan.org>
	<CAOa_K+sPqxzZY5DFimC+Ri0-jbEW5pTGLCvDcVkQyTTo1OuCRw@mail.gmail.com>
	<CAOa_K+tO9B6VmKYBdRh2t6ASis4_wTPO8QQVbYtdpobiOoq+wQ@mail.gmail.com>
	<20111221203753.GA28406@videolan.org>
Message-ID: <CAOa_K+t17rJWtgyrUVX3beQjPgLy2mb3smKOme=PHJi=yezwsQ@mail.gmail.com>

2011/12/22 Jean-Baptiste Kempf <jb at videolan.org>:
> On Wed, Dec 21, 2011 at 05:14:26PM +0900, Naohiro KORIYAMA wrote :
>> I forget the attachment...
> And that's the way we like it :D
>
> Thanks a lot.
>
> Btw, did you look at the 10bits yadif patch too?

I just knew about it. I'm looking vf_yadif.c and history now.

-- 
KORIYAMA, Naohiro
nkoriyama at gmail.com


From cheng.sun at ymail.com  Wed Dec 21 23:09:27 2011
From: cheng.sun at ymail.com (Cheng Sun)
Date: Wed, 21 Dec 2011 22:09:27 +0000
Subject: [vlc-devel] [PATCH] Break out early if a filter returns NULL,
 so other filters don't fail
Message-ID: <4EF25917.60904@ymail.com>

Test case: turn on all filters. Play some audio. Speed it up. Watch it
crash.

-- Cheng
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-Break-out-early-if-a-filter-returns-NULL-so-other-fi.patch
Type: text/x-patch
Size: 824 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111221/362e4a96/attachment.bin>

From cheng.sun at ymail.com  Wed Dec 21 23:25:15 2011
From: cheng.sun at ymail.com (Cheng Sun)
Date: Wed, 21 Dec 2011 22:25:15 +0000
Subject: [vlc-devel] [PATCH] Allow blendbench to bail out gracefully instead
 of crash, when base image cannot be found
Message-ID: <4EF25CCB.3000907@ymail.com>

Prevents a crash when blendbench cannot find the base image.

-- Cheng
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-Allow-blendbench-to-bail-out-gracefully-instead-of-c.patch
Type: text/x-patch
Size: 1650 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111221/9944493a/attachment.bin>

From rsatom at gmail.com  Thu Dec 22 00:26:45 2011
From: rsatom at gmail.com (Sergey Radionov)
Date: Thu, 22 Dec 2011 06:26:45 +0700
Subject: [vlc-devel] [PATCH] messages.c: added logging via
 OutputDebugString under windows.
In-Reply-To: <4EF1D862.3070906@web.de>
References: <1324436973-30963-1-git-send-email-funman@videolan.org>
	<4EF1D862.3070906@web.de>
Message-ID: <CADOtZwhg6g6wKYh2WdKK2bQCmjDZft3749UEhMOTJ3oZVJ6Qzw@mail.gmail.com>

2011/12/21 Tobias G?ntner <fatbull at web.de>:
> Am 21.12.2011 04:09, schrieb Sergey Radionov:
>>
>> Signed-off-by: Rafa?l Carr?<funman at videolan.org>
>
>
>> +#ifdef WIN32
>> + ? ?va_list dol;
>> + ? ?va_copy (dol, args);
>> + ? ?Win32DebugOutputMsg (0, type,&msg, format, dol);
>> + ? ?va_end (dol);
>> +#endif
>
>
> Maybe test IsDebuggerPresent() as well?
http://msdn.microsoft.com/en-us/library/windows/desktop/ms680345(v=vs.85).aspx
*Requirements:
Minimum supported client Windows 2000 Professional
Minimum supported server Windows 2000 Server*

What about WinNT 4.0, 3.x ? and win 95/98/Me?

>
>
>> + ? ?size_t fmt_len = strlen(format);
>> + ? ?char *fmt_cr = malloc(fmt_len + 1 + 1);
>> + ? ?if (!fmt_cr)
>> + ? ? ? ?return;
>> +
>> + ? ?strcpy(fmt_cr, format);
>> + ? ?fmt_cr[fmt_len] = '\n';
>> + ? ?fmt_cr[fmt_len + 1] = '\0';
>
>
> Perhaps asprintf would be easier.
>
>
>> + ? ?char *msg;
>> + ? ?if (vasprintf(&msg, fmt_cr, dol))
>> + ? ? ? ?msg = NULL;
>
>
> vasprintf returns -1 on error.
>
> Regards,
> Tobias
>
>
> _______________________________________________
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
> http://mailman.videolan.org/listinfo/vlc-devel


From jb at videolan.org  Thu Dec 22 00:28:50 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Thu, 22 Dec 2011 00:28:50 +0100
Subject: [vlc-devel] [PATCH] messages.c: added logging via
 OutputDebugString under windows.
In-Reply-To: <CADOtZwhg6g6wKYh2WdKK2bQCmjDZft3749UEhMOTJ3oZVJ6Qzw@mail.gmail.com>
References: <1324436973-30963-1-git-send-email-funman@videolan.org>
	<4EF1D862.3070906@web.de>
	<CADOtZwhg6g6wKYh2WdKK2bQCmjDZft3749UEhMOTJ3oZVJ6Qzw@mail.gmail.com>
Message-ID: <20111221232850.GA29569@videolan.org>

On Thu, Dec 22, 2011 at 06:26:45AM +0700, Sergey Radionov wrote :
> http://msdn.microsoft.com/en-us/library/windows/desktop/ms680345(v=vs.85).aspx
> *Requirements:
> Minimum supported client Windows 2000 Professional
> Minimum supported server Windows 2000 Server*
> 
> What about WinNT 4.0, 3.x ? and win 95/98/Me?

We do not care about them.
We do not even care about 2000.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From rsatom at gmail.com  Thu Dec 22 00:33:40 2011
From: rsatom at gmail.com (Sergey Radionov)
Date: Thu, 22 Dec 2011 06:33:40 +0700
Subject: [vlc-devel] [PATCH] messages.c: added logging via
 OutputDebugString under windows.
In-Reply-To: <20111221232850.GA29569@videolan.org>
References: <1324436973-30963-1-git-send-email-funman@videolan.org>
	<4EF1D862.3070906@web.de>
	<CADOtZwhg6g6wKYh2WdKK2bQCmjDZft3749UEhMOTJ3oZVJ6Qzw@mail.gmail.com>
	<20111221232850.GA29569@videolan.org>
Message-ID: <CADOtZwjhTWbqi4OLCVkJGXENOgMAV4iosPG05v02=DJitCTqMQ@mail.gmail.com>

2011/12/22 Jean-Baptiste Kempf <jb at videolan.org>:
> On Thu, Dec 22, 2011 at 06:26:45AM +0700, Sergey Radionov wrote :
>> http://msdn.microsoft.com/en-us/library/windows/desktop/ms680345(v=vs.85).aspx
>> *Requirements:
>> Minimum supported client Windows 2000 Professional
>> Minimum supported server Windows 2000 Server*
>>
>> What about WinNT 4.0, 3.x ? and win 95/98/Me?
>
> We do not care about them.
> We do not even care about 2000.
ok.

especially as:
http://msdn.microsoft.com/en-us/library/windows/desktop/aa363362(v=vs.85).aspx
*OutputDebugString function
Requirements
Minimum supported client Windows 2000 Professional
Minimum supported server Windows 2000 Server*

>
> Best regards,
>
> --
> Jean-Baptiste Kempf
> http://www.jbkempf.com/ - +33 672 704 734
> Sent from my Electronic Device
> _______________________________________________
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
> http://mailman.videolan.org/listinfo/vlc-devel


From rsatom at gmail.com  Thu Dec 22 01:15:21 2011
From: rsatom at gmail.com (Sergey Radionov)
Date: Thu, 22 Dec 2011 07:15:21 +0700
Subject: [vlc-devel] [PATCH] messages.c: added logging via
 OutputDebugString under windows.
In-Reply-To: <20111221122111.38661071@tok>
References: <1324436973-30963-1-git-send-email-funman@videolan.org>
	<4EF1D862.3070906@web.de> <20111221122111.38661071@tok>
Message-ID: <CADOtZwiGG1Z+72sZ5irh=aCrz2nWYLWDrJY+BR7ae2Dpoeg70A@mail.gmail.com>

2011/12/22 Rafa?l Carr? <funman at videolan.org>:
> Le Wed, 21 Dec 2011 14:00:18 +0100,
> Tobias G?ntner <fatbull at web.de> a ?crit :
>
>> Am 21.12.2011 04:09, schrieb Sergey Radionov:
>> > Signed-off-by: Rafa?l Carr?<funman at videolan.org>
>>
>> > +#ifdef WIN32
>> > + ? ?va_list dol;
>> > + ? ?va_copy (dol, args);
>> > + ? ?Win32DebugOutputMsg (0, type,&msg, format, dol);
>> > + ? ?va_end (dol);
>> > +#endif
>>
>> Maybe test IsDebuggerPresent() as well?
>
> I don't know about this part, I let Sergey answer
It seems not good Idea using IsDebuggerPresent, since
DebugView (http://technet.microsoft.com/en-us/sysinternals/bb896647)
is not debugger (I've check it).

>
>> > + ? ?size_t fmt_len = strlen(format);
>> > + ? ?char *fmt_cr = malloc(fmt_len + 1 + 1);
>> > + ? ?if (!fmt_cr)
>> > + ? ? ? ?return;
>> > +
>> > + ? ?strcpy(fmt_cr, format);
>> > + ? ?fmt_cr[fmt_len] = '\n';
>> > + ? ?fmt_cr[fmt_len + 1] = '\0';
>>
>> Perhaps asprintf would be easier.
>
> Yes
>
>>
>> > + ? ?char *msg;
>> > + ? ?if (vasprintf(&msg, fmt_cr, dol))
>> > + ? ? ? ?msg = NULL;
>>
>> vasprintf returns -1 on error.
>
> Yes
>
> Thanks
>
> --
> Rafa?l Carr?
> _______________________________________________
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
> http://mailman.videolan.org/listinfo/vlc-devel


From rsatom at gmail.com  Thu Dec 22 01:35:23 2011
From: rsatom at gmail.com (Sergey Radionov)
Date: Thu, 22 Dec 2011 07:35:23 +0700
Subject: [vlc-devel] [PATCH 1/1] messages.c: added logging via
	OutputDebugString under windows.
Message-ID: <1324514123-2344-1-git-send-email-RSATom@gmail.com>

---
 src/misc/messages.c |   40 ++++++++++++++++++++++++++++++++++++++++
 1 files changed, 40 insertions(+), 0 deletions(-)

diff --git a/src/misc/messages.c b/src/misc/messages.c
index 61080c0..830aed3 100644
--- a/src/misc/messages.c
+++ b/src/misc/messages.c
@@ -132,6 +132,10 @@ void vlc_Log (vlc_object_t *obj, int type, const char *module,
 static void PrintColorMsg (void *, int, const msg_item_t *,
                            const char *, va_list);
 static void PrintMsg (void *, int, const msg_item_t *, const char *, va_list);
+#ifdef WIN32
+static void Win32DebugOutputMsg (void *, int , const msg_item_t *,
+                                 const char *, va_list);
+#endif
 
 /**
  * Emit a log message. This function is the variable argument list equivalent
@@ -230,6 +234,13 @@ void vlc_vaLog (vlc_object_t *obj, int type, const char *module,
         PrintMsg (&priv->i_verbose, type, &msg, format, ap);
     va_end (ap);
 
+#ifdef WIN32
+    va_list dol;
+    va_copy (dol, args);
+    Win32DebugOutputMsg (&priv->i_verbose, type, &msg, format, dol);
+    va_end (dol);
+#endif
+
     vlc_rwlock_rdlock (&msg_lock);
     for (msg_subscription_t *sub = msg_head; sub != NULL; sub = sub->next)
     {
@@ -303,3 +314,32 @@ static void PrintMsg (void *d, int type, const msg_item_t *p_item,
     funlockfile (stream);
     vlc_restorecancel (canc);
 }
+
+#ifdef WIN32
+static void Win32DebugOutputMsg (void* d, int type, const msg_item_t *p_item,
+                                 const char *format, va_list dol)
+{
+    const signed char *pverbose = d;
+    if (*pverbose < 0 || *pverbose < (type - VLC_MSG_ERR))
+        return;
+
+    va_list dol2;
+    va_copy (dol2, dol);
+    int msg_len = vsnprintf(NULL, 0, format, dol2);
+    va_end (dol2);
+
+    char *msg = malloc(msg_len + 1 + 1);
+    if (!msg)
+        return;
+
+    msg_len = vsnprintf(msg, msg_len+1, format, dol);
+    if (msg_len > 0){
+        if(msg[msg_len-1] != '\n'){
+            msg[msg_len] = '\n';
+            msg[msg_len + 1] = '\0';
+        }
+        OutputDebugString(msg);
+    }
+    free(msg);
+}
+#endif
-- 
1.7.7.1.msysgit.0



From mirsal at mirsal.fr  Thu Dec 22 02:38:52 2011
From: mirsal at mirsal.fr (mirsal)
Date: Thu, 22 Dec 2011 02:38:52 +0100
Subject: [vlc-devel] [PATCH] gui: Replace all spaces with "%20" in URLs
In-Reply-To: <201112211948.54803.remi@remlab.net>
References: <1324488418-21119-1-git-send-email-jan.scheers@gmail.com>
	<201112211948.54803.remi@remlab.net>
Message-ID: <1324517932.8590.46.camel@mirsal-laptop1.mirsal.fr>

Hello,

On Wed, 2011-12-21 at 19:48 +0200, R?mi Denis-Courmont wrote:
> Le mercredi 21 d?cembre 2011 19:26:58 Jan Scheers, vous avez ?crit :
> > fix #3435
> 
> If I see where this is going, you should probably replace all URL-unsafe 
> characters, not just white spaces.
> 
> It might make sense to write a helper function.

Actually, Qt already has one:

http://developer.qt.nokia.com/doc/qt-4.8/qurl.html#id-ca5d3c1c-5962-46a8-b1ba-0883ed782a47

Best regards,

-- 
mirsal 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 490 bytes
Desc: This is a digitally signed message part
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111222/573b3ef1/attachment.sig>

From rsatom at gmail.com  Thu Dec 22 05:05:04 2011
From: rsatom at gmail.com (Sergey Radionov)
Date: Thu, 22 Dec 2011 11:05:04 +0700
Subject: [vlc-devel] Build under Windows&MSys. Problem with "make
	package-win-common"
Message-ID: <CADOtZwg0i-_VVaLnopRhvHHk6qQJCH4QKVVO_VMY-LRqiCoebw@mail.gmail.com>

Windows MessageBox appear:
"vlc-cache-gen.exe - Unable To Locate Component"
"This application has failed to start because libvlc.dll was not
found. Re-installing the application may fix this problem. "
"OK"

part of build log:
http://pastie.org/3055591

I think problem is that vlc-cache-gen.exe reside in _win32\lib\vlc,
but libvlc.dll is in _win32\bin, and they don't see each other.


From ptrtap at yahoo.com  Thu Dec 22 06:07:12 2011
From: ptrtap at yahoo.com (Peter Tap)
Date: Wed, 21 Dec 2011 21:07:12 -0800 (PST)
Subject: [vlc-devel] Deb package creation script for VLC
Message-ID: <1324530432.50852.YahooMailNeo@web125403.mail.ne1.yahoo.com>

Hello,

After I build vlc, I would like to create a deb package for installing on some other machines. I am wondering if there is already a script to create a deb package for vlc. Any pointers would be appreciated.


Thank you in advance for your help.

Regards,
Peter
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111221/44fa303a/attachment.html>

From remi at remlab.net  Thu Dec 22 07:27:05 2011
From: remi at remlab.net (=?iso-8859-1?q?R=E9mi?= Denis-Courmont)
Date: Thu, 22 Dec 2011 07:27:05 +0100
Subject: [vlc-devel] Deb package creation script for VLC
In-Reply-To: <1324530432.50852.YahooMailNeo@web125403.mail.ne1.yahoo.com>
References: <1324530432.50852.YahooMailNeo@web125403.mail.ne1.yahoo.com>
Message-ID: <201112220727.05557.remi@remlab.net>

   Hello,

On Thursday 22 December 2011, Peter Tap wrote:
> After I build vlc, I would like to create a deb package for installing on
> some other machines. I am wondering if there is already a script to create
> a deb package for vlc. Any pointers would be appreciated.

You can find that in the official Debian package sources.

-- 
R?mi Denis-Courmont
http://www.remlab.info


From enrique.osuna at gmail.com  Thu Dec 22 09:13:08 2011
From: enrique.osuna at gmail.com (Faustino Osuna)
Date: Thu, 22 Dec 2011 09:13:08 +0100
Subject: [vlc-devel] [PATCH] gui: Replace all spaces with "%20" in URLs
In-Reply-To: <1324517932.8590.46.camel@mirsal-laptop1.mirsal.fr>
References: <1324488418-21119-1-git-send-email-jan.scheers@gmail.com>
	<201112211948.54803.remi@remlab.net>
	<1324517932.8590.46.camel@mirsal-laptop1.mirsal.fr>
Message-ID: <CAKhCBsxJpaErzQ84TfP4OePFpAR_5x1QvgHvUBUTFamUkti=+w@mail.gmail.com>

So does the Mac.

http://developer.apple.com/library/mac/#documentation/Cocoa/Reference/Foundation/Classes/NSString_Class/Reference/NSString.html

Something like this maybe:  [o_mrl_string
 stringByAddingPercentEscapesUsingEncoding:NSUTF8StringEncoding];

Enrique

On Thu, Dec 22, 2011 at 2:38 AM, mirsal <mirsal at mirsal.fr> wrote:

> Hello,
>
> On Wed, 2011-12-21 at 19:48 +0200, R?mi Denis-Courmont wrote:
> > Le mercredi 21 d?cembre 2011 19:26:58 Jan Scheers, vous avez ?crit :
> > > fix #3435
> >
> > If I see where this is going, you should probably replace all URL-unsafe
> > characters, not just white spaces.
> >
> > It might make sense to write a helper function.
>
> Actually, Qt already has one:
>
>
> http://developer.qt.nokia.com/doc/qt-4.8/qurl.html#id-ca5d3c1c-5962-46a8-b1ba-0883ed782a47
>
> Best regards,
>
> --
> mirsal
>
> _______________________________________________
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
> http://mailman.videolan.org/listinfo/vlc-devel
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111222/8a1e0745/attachment.html>

From jpsaman at videolan.org  Thu Dec 22 10:45:22 2011
From: jpsaman at videolan.org (Jean-Paul Saman)
Date: Thu, 22 Dec 2011 10:45:22 +0100
Subject: [vlc-devel] =?iso-8859-1?q?RTSP_Amino=B4s?=
In-Reply-To: <0a7d01ccbff4$dbb57710$93206530$@com>
References: <090601ccbf33$45d55d30$d1801790$@com>
	<yw1xiplbq6f8.fsf@unicorn.mansr.com>
	<0a7d01ccbff4$dbb57710$93206530$@com>
Message-ID: <CAK9US3AftShNcKVYZk4sAvD3FaFuww8hBK76u921Aj0ZG=erqw@mail.gmail.com>

On Wed, Dec 21, 2011 at 4:26 PM, Ben Wheway <ben at europa-network.com> wrote:
> Hi Mans
>
> I have seen that a few people have created a workaround with VLC to allow this to work
>
> Is this something you could do for us?

Amino STBs expect audio and video muxed as MPEG TS. They cannot handle
individual ports for audio and video.

Thus for RTPS you need to use the TS muxer for Amino STBs

Kind Regards,

Jean-Paul Saman


From jeanmichel.lambert7 at gmail.com  Thu Dec 22 09:40:53 2011
From: jeanmichel.lambert7 at gmail.com (JM Lambert)
Date: Thu, 22 Dec 2011 09:40:53 +0100
Subject: [vlc-devel] [PATCH] Changes to libvlc required for plugin API
	improvement (input.meta method)
In-Reply-To: <Patch for web plugin API improvement (Part 1)>
References: <Patch for web plugin API improvement (Part 1)>
Message-ID: <1324543253-17750-1-git-send-email-jean-michel.lambert7@wanadoo.fr>

---
 include/vlc/libvlc_media_player.h |    8 ++++++++
 include/vlc_input.h               |    2 ++
 lib/libvlc.sym                    |    1 +
 lib/media_player.c                |   19 +++++++++++++++++++
 src/input/control.c               |   12 ++++++++++++
 5 files changed, 42 insertions(+), 0 deletions(-)

diff --git a/include/vlc/libvlc_media_player.h b/include/vlc/libvlc_media_player.h
index 152a95d..d361ce2 100644
--- a/include/vlc/libvlc_media_player.h
+++ b/include/vlc/libvlc_media_player.h
@@ -756,6 +756,14 @@ LIBVLC_API int libvlc_media_player_set_rate( libvlc_media_player_t *p_mi, float
 LIBVLC_API libvlc_state_t libvlc_media_player_get_state( libvlc_media_player_t *p_mi );
 
 /**
+ * Get media meta
+ *
+ * \param p_mi the Media Player
+ * \return ith elemet of meta structure
+ */
+LIBVLC_API char* libvlc_media_player_get_meta( libvlc_media_player_t *p_mi, int i);
+
+/**
  * Get movie fps rate
  *
  * \param p_mi the Media Player
diff --git a/include/vlc_input.h b/include/vlc_input.h
index 7d8320a..accb384 100644
--- a/include/vlc_input.h
+++ b/include/vlc_input.h
@@ -516,6 +516,8 @@ enum input_query_e
     /* External clock managments */
     INPUT_GET_PCR_SYSTEM,   /* arg1=mtime_t *, arg2=mtime_t *       res=can fail */
     INPUT_MODIFY_PCR_SYSTEM,/* arg1=int absolute, arg2=mtime_t      res=can fail */
+    INPUT_GET_META,/* arg1=vlc_meta_t *      res=can fail */
+
 };
 
 /** @}*/
diff --git a/lib/libvlc.sym b/lib/libvlc.sym
index 45327d5..2607f99 100644
--- a/lib/libvlc.sym
+++ b/lib/libvlc.sym
@@ -120,6 +120,7 @@ libvlc_media_player_get_agl
 libvlc_media_player_get_chapter
 libvlc_media_player_get_chapter_count
 libvlc_media_player_get_chapter_count_for_title
+libvlc_media_player_get_meta
 libvlc_media_player_get_fps
 libvlc_media_player_get_hwnd
 libvlc_media_player_get_length
diff --git a/lib/media_player.c b/lib/media_player.c
index b426637..1978e72 100644
--- a/lib/media_player.c
+++ b/lib/media_player.c
@@ -1217,6 +1217,25 @@ void libvlc_media_player_previous_chapter( libvlc_media_player_t *p_mi )
     vlc_object_release( p_input_thread );
 }
 
+char * libvlc_media_player_get_meta( libvlc_media_player_t *p_mi , int index)
+{
+    input_thread_t *p_input_thread = libvlc_get_input_thread ( p_mi );
+    vlc_meta_t *p_meta;
+
+    if( p_input_thread )
+    {
+        if( input_Control( p_input_thread, INPUT_GET_META, &p_meta ) )
+                vlc_meta_Set( p_meta, vlc_meta_Title, "unknown" );
+        vlc_object_release( p_input_thread );
+        if (vlc_meta_Get(p_meta,index)) return vlc_meta_Get(p_meta,index);
+        else return "";
+    }
+    else
+    {
+        return "";
+    }
+}
+
 float libvlc_media_player_get_fps( libvlc_media_player_t *p_mi )
 {
     input_thread_t *p_input_thread = libvlc_get_input_thread ( p_mi );
diff --git a/src/input/control.c b/src/input/control.c
index d50ea97..4e95661 100644
--- a/src/input/control.c
+++ b/src/input/control.c
@@ -65,6 +65,8 @@ int input_vaControl( input_thread_t *p_input, int i_query, va_list args )
     seekpoint_t *p_bkmk, ***ppp_bkmk;
     int i_bkmk = 0;
     int *pi_bkmk;
+    vlc_meta_t ** p_meta;
+
 
     int i_int, *pi_int;
     bool b_bool, *pb_bool;
@@ -357,6 +359,16 @@ int input_vaControl( input_thread_t *p_input, int i_query, va_list args )
             vlc_mutex_unlock( &p_input->p->p_item->lock );
             return VLC_SUCCESS;
 
+        case INPUT_GET_META:
+            p_meta = (vlc_meta_t **)va_arg( args, vlc_meta_t ** );
+
+            vlc_mutex_lock( &p_input->p->p_item->lock );
+            input_item_t *p_item = input_GetItem( p_input );
+            *p_meta = p_item->p_meta;
+            vlc_mutex_unlock( &p_input->p->p_item->lock );
+            return VLC_SUCCESS;
+
+
         case INPUT_ADD_SLAVE:
             psz = (char*)va_arg( args, char * );
             if( psz && *psz )
-- 
1.7.5.4



From rob at hobbyistsoftware.com  Thu Dec 22 12:19:30 2011
From: rob at hobbyistsoftware.com (Rob Jonson)
Date: Thu, 22 Dec 2011 11:19:30 +0000
Subject: [vlc-devel] [PATCH] gui: Replace all spaces with "%20" in URLs
In-Reply-To: <CAKhCBsxJpaErzQ84TfP4OePFpAR_5x1QvgHvUBUTFamUkti=+w@mail.gmail.com>
References: <1324488418-21119-1-git-send-email-jan.scheers@gmail.com>
	<201112211948.54803.remi@remlab.net>
	<1324517932.8590.46.camel@mirsal-laptop1.mirsal.fr>
	<CAKhCBsxJpaErzQ84TfP4OePFpAR_5x1QvgHvUBUTFamUkti=+w@mail.gmail.com>
Message-ID: <CACTOy+qPzEwCqxv9jVMX-qbr2YncC6duAafxFGdbOmX-=43NNA@mail.gmail.com>

stringByAddingPercentEscapesUsingEncoding:NSUTF8StringEncoding doesn't
quite do the job on mac.

I use this category on NSString:

-(NSString*)fullyEscapedStringForUrl
{
	NSString *newString= (NSString*)CFURLCreateStringByAddingPercentEscapes(
																  NULL,
																  (CFStringRef)self,
																  NULL,
																  (CFStringRef)@";/?:@&=+$,",
																  kCFStringEncodingUTF8
																  );

	[newString autorelease];
	return newString;
}


This is assuming you do want to encode chars like & and ? as part of a
param, rather than a param indicator.

cheers,

Rob



On 22 December 2011 08:13, Faustino Osuna <enrique.osuna at gmail.com> wrote:
> So does the Mac.
>
> http://developer.apple.com/library/mac/#documentation/Cocoa/Reference/Foundation/Classes/NSString_Class/Reference/NSString.html
>
> Something like this maybe:
> ?[o_mrl_string?stringByAddingPercentEscapesUsingEncoding:NSUTF8StringEncoding];
>
> Enrique
>
> On Thu, Dec 22, 2011 at 2:38 AM, mirsal <mirsal at mirsal.fr> wrote:
>>
>> Hello,
>>
>> On Wed, 2011-12-21 at 19:48 +0200, R?mi Denis-Courmont wrote:
>> > Le mercredi 21 d?cembre 2011 19:26:58 Jan Scheers, vous avez ?crit :
>> > > fix #3435
>> >
>> > If I see where this is going, you should probably replace all URL-unsafe
>> > characters, not just white spaces.
>> >
>> > It might make sense to write a helper function.
>>
>> Actually, Qt already has one:
>>
>>
>> http://developer.qt.nokia.com/doc/qt-4.8/qurl.html#id-ca5d3c1c-5962-46a8-b1ba-0883ed782a47
>>
>> Best regards,
>>
>> --
>> mirsal
>>
>> _______________________________________________
>> vlc-devel mailing list
>> To unsubscribe or modify your subscription options:
>> http://mailman.videolan.org/listinfo/vlc-devel
>>
>
>
> _______________________________________________
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
> http://mailman.videolan.org/listinfo/vlc-devel
>



-- 





Hobbyist Software is a trading name of Hobbyist Software Limited.
Registered office Flat 4 Eastmead Court, Bristol BS9 1HP. Registered
in England. Company no:7876492


From cheng.sun at ymail.com  Thu Dec 22 12:21:59 2011
From: cheng.sun at ymail.com (Cheng Sun)
Date: Thu, 22 Dec 2011 11:21:59 +0000
Subject: [vlc-devel] [PATCH] Allow blendbench to bail out gracefully
 instead of crash, when base image cannot be found
In-Reply-To: <4EF25CCB.3000907@ymail.com>
References: <4EF25CCB.3000907@ymail.com>
Message-ID: <4EF312D7.9020709@ymail.com>

On 21/12/11 22:25, Cheng Sun wrote:
> Prevents a crash when blendbench cannot find the base image.
>
> -- Cheng
>
Sorry, late-night patch creation = fail.

New patch which actually compiles attached.

-- Cheng
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-Allow-blendbench-to-bail-out-gracefully-instead-of-c.patch
Type: text/x-patch
Size: 1649 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111222/1b5f6ea3/attachment.bin>

From edward.c.wang at compdigitec.com  Thu Dec 22 13:19:38 2011
From: edward.c.wang at compdigitec.com (Edward Wang)
Date: Thu, 22 Dec 2011 07:19:38 -0500
Subject: [vlc-devel] [PATCH] Qt4: Add option to show system tray
 notifications when VLC is visible
Message-ID: <4EF3205A.4010509@compdigitec.com>

Hi all:

The attached patch will add an option to show system tray notifications 
when VLC is visible (see http://trac.videolan.org/vlc/ticket/5041). This 
will help some users who may wish to show popup notifications all the 
time, e.g. when using virtual desktops.

Thanks,
     Edward Wan
-------------- next part --------------
A non-text attachment was scrubbed...
Name: notifications_5041.patch
Type: text/x-patch
Size: 8487 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111222/2917c5ee/attachment.bin>

From jan.scheers at gmail.com  Thu Dec 22 14:00:00 2011
From: jan.scheers at gmail.com (Jan Scheers)
Date: Thu, 22 Dec 2011 14:00:00 +0100
Subject: [vlc-devel] [PATCH] gui:Percent encode all URL unsafe characters
Message-ID: <1324558800-2730-1-git-send-email-jan.scheers@gmail.com>

This should fix #3435.
I wanted to reply to the previous thread but I don't know how that works in mailing lists. 

---
 modules/gui/macosx/open.m                  |    2 ++
 modules/gui/qt4/components/open_panels.cpp |   18 ++----------------
 modules/gui/qt4/components/open_panels.hpp |    9 ---------
 3 files changed, 4 insertions(+), 25 deletions(-)

diff --git a/modules/gui/macosx/open.m b/modules/gui/macosx/open.m
index 5c5fb0c..48c3d2d 100644
--- a/modules/gui/macosx/open.m
+++ b/modules/gui/macosx/open.m
@@ -1040,6 +1040,8 @@ static VLCOpen *_o_sharedMainInstance = nil;
     else
     {
         o_mrl_string = [o_net_http_url stringValue];
+        o_mrl_string = [o_mrl_string stringByAddingPercentEscapesUsingEncoding:NSUTF8StringEncoding];
+        o_mrl_string = [o_mrl_string stringByReplacingOccurrencesOfString:@"%25" withString:@"%"];
     }
     [self setMRL: o_mrl_string];
 }
diff --git a/modules/gui/qt4/components/open_panels.cpp b/modules/gui/qt4/components/open_panels.cpp
index 78bb4b7..c1f06d9 100644
--- a/modules/gui/qt4/components/open_panels.cpp
+++ b/modules/gui/qt4/components/open_panels.cpp
@@ -601,8 +601,6 @@ NetOpenPanel::NetOpenPanel( QWidget *_parent, intf_thread_t *_p_intf ) :
     else
         b_recentList = false;
 
-    /* Use a simple validator for URLs */
-    ui.urlComboBox->setValidator( new UrlValidator( this ) );
     ui.urlComboBox->setFocus();
 }
 
@@ -642,6 +640,8 @@ void NetOpenPanel::updateMRL()
 {
     QString url = ui.urlComboBox->lineEdit()->text();
 
+    url = QString::fromAscii( QUrl::toPercentEncoding( url, "$&+,/:;=?@%", "" ) );
+
     if( url.isEmpty() )
         return;
 
@@ -652,20 +652,6 @@ void NetOpenPanel::updateMRL()
     emit mrlUpdated( qsl, "" );
 }
 
-QValidator::State UrlValidator::validate( QString& str, int& ) const
-{
-    if( str.contains( ' ' ) )
-        return QValidator::Invalid;
-    if( !str.contains( "://" ) )
-        return QValidator::Intermediate;
-    return QValidator::Acceptable;
-}
-
-void UrlValidator::fixup( QString& str ) const
-{
-    str = str.trimmed();
-}
-
 /**************************************************************************
  * Open Capture device ( DVB, PVR, V4L, and similar )                     *
  **************************************************************************/
diff --git a/modules/gui/qt4/components/open_panels.hpp b/modules/gui/qt4/components/open_panels.hpp
index 85257ce..0f7c216 100644
--- a/modules/gui/qt4/components/open_panels.hpp
+++ b/modules/gui/qt4/components/open_panels.hpp
@@ -149,15 +149,6 @@ public slots:
     virtual void updateMRL();
 };
 
-class UrlValidator : public QValidator
-{
-   Q_OBJECT
-public:
-   UrlValidator( QObject *parent ) : QValidator( parent ) { }
-   void fixup( QString& ) const;
-   QValidator::State validate( QString&, int& ) const;
-};
-
 class DiscOpenPanel: public OpenPanel
 {
     Q_OBJECT
-- 
1.7.8



From jb at videolan.org  Thu Dec 22 23:05:07 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Thu, 22 Dec 2011 23:05:07 +0100
Subject: [vlc-devel] [PATCH] Allow blendbench to bail out gracefully
 instead of crash, when base image cannot be found
In-Reply-To: <4EF312D7.9020709@ymail.com>
References: <4EF25CCB.3000907@ymail.com>
 <4EF312D7.9020709@ymail.com>
Message-ID: <20111222220506.GA26329@videolan.org>

On Thu, Dec 22, 2011 at 11:21:59AM +0000, Cheng Sun wrote :
> On 21/12/11 22:25, Cheng Sun wrote:
> > Prevents a crash when blendbench cannot find the base image.
> >
> > -- Cheng
> >
> Sorry, late-night patch creation = fail.
> 
> New patch which actually compiles attached.

Applied.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Fri Dec 23 02:06:50 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Fri, 23 Dec 2011 02:06:50 +0100
Subject: [vlc-devel] [PATCH] Changes to libvlc required for plugin API
 improvement (input.meta method)
In-Reply-To: <1324543253-17750-1-git-send-email-jean-michel.lambert7@wanadoo.fr>
References: <1324543253-17750-1-git-send-email-jean-michel.lambert7@wanadoo.fr>
Message-ID: <20111223010650.GA30294@videolan.org>

On Thu, Dec 22, 2011 at 09:40:53AM +0100, JM Lambert wrote :
> +char * libvlc_media_player_get_meta( libvlc_media_player_t *p_mi , int index)

Is the caller supposed to free it?

> +        else return "";

> +        return "";

Why not NULL ?

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Fri Dec 23 02:07:35 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Fri, 23 Dec 2011 02:07:35 +0100
Subject: [vlc-devel] Build under Windows&MSys. Problem with "make
 package-win-common"
In-Reply-To: <CADOtZwg0i-_VVaLnopRhvHHk6qQJCH4QKVVO_VMY-LRqiCoebw@mail.gmail.com>
References: <CADOtZwg0i-_VVaLnopRhvHHk6qQJCH4QKVVO_VMY-LRqiCoebw@mail.gmail.com>
Message-ID: <20111223010735.GB30294@videolan.org>

On Thu, Dec 22, 2011 at 11:05:04AM +0700, Sergey Radionov wrote :
> Windows MessageBox appear:
> "vlc-cache-gen.exe - Unable To Locate Component"
> "This application has failed to start because libvlc.dll was not
> found. Re-installing the application may fix this problem. "
> "OK"
> 
> part of build log:
> http://pastie.org/3055591
> 
> I think problem is that vlc-cache-gen.exe reside in _win32\lib\vlc,
> but libvlc.dll is in _win32\bin, and they don't see each other.

Force the test to false.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From zhushisongzhu at yahoo.com  Fri Dec 23 08:58:56 2011
From: zhushisongzhu at yahoo.com (zhu shi song)
Date: Thu, 22 Dec 2011 23:58:56 -0800 (PST)
Subject: [vlc-devel] How to trim/cut clips with VLC
Message-ID: <1324627136.35411.YahooMailNeo@web125705.mail.ne1.yahoo.com>

I'm using VLC-1.1.11.? I hope to cut one wmv2 video from start time to stop time into one piece of it.? Can I use transcode to do it?? Or it can , what's the arguments of vlc command??? If someone knows please kindly let me know.
thanks



From jpsaman at videolan.org  Fri Dec 23 09:55:16 2011
From: jpsaman at videolan.org (Jean-Paul Saman)
Date: Fri, 23 Dec 2011 09:55:16 +0100
Subject: [vlc-devel] How to trim/cut clips with VLC
In-Reply-To: <1324627136.35411.YahooMailNeo@web125705.mail.ne1.yahoo.com>
References: <1324627136.35411.YahooMailNeo@web125705.mail.ne1.yahoo.com>
Message-ID: <CAK9US3D5G=f8fhW68R1ypZV2KY1eUUx+7Hv515W0TarOqSmieQ@mail.gmail.com>

On Fri, Dec 23, 2011 at 8:58 AM, zhu shi song <zhushisongzhu at yahoo.com> wrote:
> I'm using VLC-1.1.11.? I hope to cut one wmv2 video from start time to stop time into one piece of it.? Can I use transcode to do it?? Or it can , what's the arguments of vlc command??? If someone knows please kindly let me know.
> thanks
>

Try out VLMC from vlmc.org. It is based upon libvlc.

Kind regards,

Jean-Paul Saman


From rsatom at gmail.com  Fri Dec 23 10:25:11 2011
From: rsatom at gmail.com (Sergey Radionov)
Date: Fri, 23 Dec 2011 16:25:11 +0700
Subject: [vlc-devel] [PATCH 1/1] messages.c: added logging via
	OutputDebugString under windows.
Message-ID: <1324632311-2620-1-git-send-email-RSATom@gmail.com>

---
 src/misc/messages.c |   43 +++++++++++++++++++++++++++++++++++++++++++
 1 files changed, 43 insertions(+), 0 deletions(-)

diff --git a/src/misc/messages.c b/src/misc/messages.c
index 8c0b723..66bd62e 100644
--- a/src/misc/messages.c
+++ b/src/misc/messages.c
@@ -132,6 +132,10 @@ void vlc_Log (vlc_object_t *obj, int type, const char *module,
 static void PrintColorMsg (void *, int, const msg_item_t *,
                            const char *, va_list);
 static void PrintMsg (void *, int, const msg_item_t *, const char *, va_list);
+#ifdef WIN32
+static void Win32DebugOutputMsg (void *, int , const msg_item_t *,
+                                 const char *, va_list);
+#endif
 
 /**
  * Emit a log message. This function is the variable argument list equivalent
@@ -230,6 +234,13 @@ void vlc_vaLog (vlc_object_t *obj, int type, const char *module,
         PrintMsg (&priv->i_verbose, type, &msg, format, ap);
     va_end (ap);
 
+#ifdef WIN32
+    va_list dol;
+    va_copy (dol, args);
+    Win32DebugOutputMsg (&priv->i_verbose, type, &msg, format, dol);
+    va_end (dol);
+#endif
+
     vlc_rwlock_rdlock (&msg_lock);
     for (msg_subscription_t *sub = msg_head; sub != NULL; sub = sub->next)
     {
@@ -303,3 +314,35 @@ static void PrintMsg (void *d, int type, const msg_item_t *p_item,
     funlockfile (stream);
     vlc_restorecancel (canc);
 }
+
+#ifdef WIN32
+static void Win32DebugOutputMsg (void* d, int type, const msg_item_t *p_item,
+                                 const char *format, va_list dol)
+{
+    const signed char *pverbose = d;
+    if (pverbose && (*pverbose < 0 || *pverbose < (type - VLC_MSG_ERR)))
+        return;
+
+    va_list dol2;
+    va_copy (dol2, dol);
+    int msg_len = vsnprintf(NULL, 0, format, dol2);
+    va_end (dol2);
+
+    if(msg_len <= 0)
+        return;
+
+    char *msg = malloc(msg_len + 1 + 1);
+    if (!msg)
+        return;
+
+    msg_len = vsnprintf(msg, msg_len+1, format, dol);
+    if (msg_len > 0){
+        if(msg[msg_len-1] != '\n'){
+            msg[msg_len] = '\n';
+            msg[msg_len + 1] = '\0';
+        }
+        OutputDebugString(msg);
+    }
+    free(msg);
+}
+#endif
-- 
1.7.7.1.msysgit.0



From rsatom at gmail.com  Fri Dec 23 11:03:10 2011
From: rsatom at gmail.com (Sergey Radionov)
Date: Fri, 23 Dec 2011 17:03:10 +0700
Subject: [vlc-devel] [PATCH] ActiveX: added Windows Manager detaching from
	libvlc on libvlc_media_player_release
Message-ID: <1324634590-1104-1-git-send-email-RSATom@gmail.com>

---
 activex/plugin.cpp |    1 +
 1 files changed, 1 insertions(+), 0 deletions(-)

diff --git a/activex/plugin.cpp b/activex/plugin.cpp
index e7e76c1..e436122 100644
--- a/activex/plugin.cpp
+++ b/activex/plugin.cpp
@@ -1249,6 +1249,7 @@ bool VLCPlugin::playlist_select( int idx )
         if( isPlaying() )
             playlist_stop();
         player_unregister_events();
+        _WindowsManager.LibVlcDetach();
         libvlc_media_player_release( _p_mplayer );
         _p_mplayer = NULL;
     }
-- 
1.7.7.1.msysgit.0



From beauze.h at gmail.com  Fri Dec 23 15:41:45 2011
From: beauze.h at gmail.com (=?UTF-8?Q?Hugo_Beauz=C3=A9e=2DLuyssen?=)
Date: Fri, 23 Dec 2011 15:41:45 +0100
Subject: [vlc-devel] [Patch] Adding an helper to convert iso8601
 durations to seconds
In-Reply-To: <CAK-C75R-sJ7AOJ9AGOh680VM2YAHdbnUTU2nFbJPSKhVpiPnFg@mail.gmail.com>
References: <CAK-C75R5BPUud6z2QxrFNNnh1wMFktu48O59oXi2bAUqZ0mYMw@mail.gmail.com>
	<CAK-C75RtmcQd0m32+3d9ue7WZKH8v9n84izFWRp=ziwKm4yCEw@mail.gmail.com>
	<CAK-C75QD08hyZ4XNBR7nYkSFyLqR7Tm887Uxw7fWqpXRGJd=Dg@mail.gmail.com>
	<201112141905.47669.remi@remlab.net>
	<CAK-C75RayZc-hHcd9-BE-cF5gXFF_JjNB6hDGCz5wxS4OfSfAg@mail.gmail.com>
	<CAK-C75R-sJ7AOJ9AGOh680VM2YAHdbnUTU2nFbJPSKhVpiPnFg@mail.gmail.com>
Message-ID: <CAK-C75QiyYQ7k0cexse9s2L17+dBab6Js4TCcRJ91EKDAzEgQA@mail.gmail.com>

On Fri, Dec 16, 2011 at 4:40 PM, Hugo Beauz?e-Luyssen
<beauze.h at gmail.com> wrote:
> On Wed, Dec 14, 2011 at 6:33 PM, Hugo Beauz?e-Luyssen
> <beauze.h at gmail.com> wrote:
>> 2011/12/14 R?mi Denis-Courmont <remi at remlab.net>:
>>> Le mercredi 14 d?cembre 2011 00:48:41 Hugo Beauz?e-Luyssen, vous avez ?crit :
>>>> On Tue, Dec 13, 2011 at 11:46 PM, Hugo Beauz?e-Luyssen
>>>>
>>>> <beauze.h at gmail.com> wrote:
>>>> > 2011/12/9 R?mi Denis-Courmont <remi at remlab.net>:
>>>> >> Function name is too long. And time_t is thecorrect type for seconds
>>>> >> IMHO.
>>>>
>>>> Oops, this one's the good one.
>>>
>>> I think this patch does not work due to typo. Please drop _to_s*. It's
>>> obviously seconds since the return type is time_t.
>>>

Hello,

If there's not objection, could this patch be applied? I need it in
the dash module in order to handle durations.

Best regards,

-- 
Hugo Beauz?e-Luyssen


From remi at remlab.net  Fri Dec 23 15:50:06 2011
From: remi at remlab.net (=?iso-8859-1?q?R=E9mi?= Denis-Courmont)
Date: Fri, 23 Dec 2011 15:50:06 +0100
Subject: [vlc-devel] [PATCH] gui: Replace all spaces with "%20" in URLs
In-Reply-To: <CACTOy+qPzEwCqxv9jVMX-qbr2YncC6duAafxFGdbOmX-=43NNA@mail.gmail.com>
References: <1324488418-21119-1-git-send-email-jan.scheers@gmail.com>
	<CAKhCBsxJpaErzQ84TfP4OePFpAR_5x1QvgHvUBUTFamUkti=+w@mail.gmail.com>
	<CACTOy+qPzEwCqxv9jVMX-qbr2YncC6duAafxFGdbOmX-=43NNA@mail.gmail.com>
Message-ID: <201112231550.06802.remi@remlab.net>

   Hello,

On Thursday 22 December 2011, Rob Jonson wrote:
> This is assuming you do want to encode chars like & and ? as part of a
> param, rather than a param indicator.

For reserved characters, there is a fundamental and thus unavoidable 
ambiguity: whether each of them is an URL component separators or an 
incorrectly (un)encoded character.

IMHO, the only reasonable policy consists of assuming reserved characters are 
separators, and leave them as is.

-- 
R?mi Denis-Courmont
http://www.remlab.info


From pdherbemont at gmail.com  Fri Dec 23 18:00:49 2011
From: pdherbemont at gmail.com (Pierre d'Herbemont)
Date: Fri, 23 Dec 2011 18:00:49 +0100
Subject: [vlc-devel] [PATCH] Changes to libvlc required for plugin API
	improvement (input.meta method)
In-Reply-To: <1324543253-17750-1-git-send-email-jean-michel.lambert7@wanadoo.fr>
References: <Patch for web plugin API improvement (Part 1)>
	<1324543253-17750-1-git-send-email-jean-michel.lambert7@wanadoo.fr>
Message-ID: <B97E3084-CD30-4907-96AB-8242E95E9826@gmail.com>

Hi,

The same functionality can be achieved by:
m = libvlc_media_player_get_media(mp);
libvlc_media_get_meta(m);

That means that, there is no need to:
1- modify libvlccore
2- require a running input thread to retrieve the meta data

Also the prototype of libvlc_media_get_meta() is better, as it takes a meta data key instead of an index.

Finally, adding a wrapper around the above function calls seems like a better approach... If you thinks such a wrapper is valuable.

Pierre.

On Dec 22, 2011, at 9:40, JM Lambert <jeanmichel.lambert7 at gmail.com> wrote:

> ---
> include/vlc/libvlc_media_player.h |    8 ++++++++
> include/vlc_input.h               |    2 ++
> lib/libvlc.sym                    |    1 +
> lib/media_player.c                |   19 +++++++++++++++++++
> src/input/control.c               |   12 ++++++++++++
> 5 files changed, 42 insertions(+), 0 deletions(-)
> 
> diff --git a/include/vlc/libvlc_media_player.h b/include/vlc/libvlc_media_player.h
> index 152a95d..d361ce2 100644
> --- a/include/vlc/libvlc_media_player.h
> +++ b/include/vlc/libvlc_media_player.h
> @@ -756,6 +756,14 @@ LIBVLC_API int libvlc_media_player_set_rate( libvlc_media_player_t *p_mi, float
> LIBVLC_API libvlc_state_t libvlc_media_player_get_state( libvlc_media_player_t *p_mi );
> 
> /**
> + * Get media meta
> + *
> + * \param p_mi the Media Player
> + * \return ith elemet of meta structure
> + */
> +LIBVLC_API char* libvlc_media_player_get_meta( libvlc_media_player_t *p_mi, int i);
> +
> +/**
>  * Get movie fps rate
>  *
>  * \param p_mi the Media Player
> diff --git a/include/vlc_input.h b/include/vlc_input.h
> index 7d8320a..accb384 100644
> --- a/include/vlc_input.h
> +++ b/include/vlc_input.h
> @@ -516,6 +516,8 @@ enum input_query_e
>     /* External clock managments */
>     INPUT_GET_PCR_SYSTEM,   /* arg1=mtime_t *, arg2=mtime_t *       res=can fail */
>     INPUT_MODIFY_PCR_SYSTEM,/* arg1=int absolute, arg2=mtime_t      res=can fail */
> +    INPUT_GET_META,/* arg1=vlc_meta_t *      res=can fail */
> +
> };
> 
> /** @}*/
> diff --git a/lib/libvlc.sym b/lib/libvlc.sym
> index 45327d5..2607f99 100644
> --- a/lib/libvlc.sym
> +++ b/lib/libvlc.sym
> @@ -120,6 +120,7 @@ libvlc_media_player_get_agl
> libvlc_media_player_get_chapter
> libvlc_media_player_get_chapter_count
> libvlc_media_player_get_chapter_count_for_title
> +libvlc_media_player_get_meta
> libvlc_media_player_get_fps
> libvlc_media_player_get_hwnd
> libvlc_media_player_get_length
> diff --git a/lib/media_player.c b/lib/media_player.c
> index b426637..1978e72 100644
> --- a/lib/media_player.c
> +++ b/lib/media_player.c
> @@ -1217,6 +1217,25 @@ void libvlc_media_player_previous_chapter( libvlc_media_player_t *p_mi )
>     vlc_object_release( p_input_thread );
> }
> 
> +char * libvlc_media_player_get_meta( libvlc_media_player_t *p_mi , int index)
> +{
> +    input_thread_t *p_input_thread = libvlc_get_input_thread ( p_mi );
> +    vlc_meta_t *p_meta;
> +
> +    if( p_input_thread )
> +    {
> +        if( input_Control( p_input_thread, INPUT_GET_META, &p_meta ) )
> +                vlc_meta_Set( p_meta, vlc_meta_Title, "unknown" );
> +        vlc_object_release( p_input_thread );
> +        if (vlc_meta_Get(p_meta,index)) return vlc_meta_Get(p_meta,index);
> +        else return "";
> +    }
> +    else
> +    {
> +        return "";
> +    }
> +}
> +
> float libvlc_media_player_get_fps( libvlc_media_player_t *p_mi )
> {
>     input_thread_t *p_input_thread = libvlc_get_input_thread ( p_mi );
> diff --git a/src/input/control.c b/src/input/control.c
> index d50ea97..4e95661 100644
> --- a/src/input/control.c
> +++ b/src/input/control.c
> @@ -65,6 +65,8 @@ int input_vaControl( input_thread_t *p_input, int i_query, va_list args )
>     seekpoint_t *p_bkmk, ***ppp_bkmk;
>     int i_bkmk = 0;
>     int *pi_bkmk;
> +    vlc_meta_t ** p_meta;
> +
> 
>     int i_int, *pi_int;
>     bool b_bool, *pb_bool;
> @@ -357,6 +359,16 @@ int input_vaControl( input_thread_t *p_input, int i_query, va_list args )
>             vlc_mutex_unlock( &p_input->p->p_item->lock );
>             return VLC_SUCCESS;
> 
> +        case INPUT_GET_META:
> +            p_meta = (vlc_meta_t **)va_arg( args, vlc_meta_t ** );
> +
> +            vlc_mutex_lock( &p_input->p->p_item->lock );
> +            input_item_t *p_item = input_GetItem( p_input );
> +            *p_meta = p_item->p_meta;
> +            vlc_mutex_unlock( &p_input->p->p_item->lock );
> +            return VLC_SUCCESS;
> +
> +
>         case INPUT_ADD_SLAVE:
>             psz = (char*)va_arg( args, char * );
>             if( psz && *psz )
> -- 
> 1.7.5.4
> 
> _______________________________________________
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
> http://mailman.videolan.org/listinfo/vlc-devel


From zhushisongzhu at yahoo.com  Fri Dec 23 19:07:52 2011
From: zhushisongzhu at yahoo.com (zhu shi song)
Date: Fri, 23 Dec 2011 10:07:52 -0800 (PST)
Subject: [vlc-devel] start-time has no effect when transcoding
Message-ID: <1324663672.94288.YahooMailNeo@web125701.mail.ne1.yahoo.com>

I use VLC to? transcode one wmv video to mp4 video.? I hope I can clip the video using start-time and stop-time.? When reaching stop-time, the transcoding ended. But start-time has no effect,? transcoding always started from 0.? Who has the experience to cut one video using start-time to stop-time transcoding? 

Thanks



From jean-michel.lambert7 at wanadoo.fr  Fri Dec 23 19:15:53 2011
From: jean-michel.lambert7 at wanadoo.fr (jean-michel Lambert)
Date: Fri, 23 Dec 2011 19:15:53 +0100
Subject: [vlc-devel] [PATCH] Changes to libvlc required for plugin API
 improvement (input.meta method)
In-Reply-To: <B97E3084-CD30-4907-96AB-8242E95E9826@gmail.com>
References: <Patch for web plugin API improvement (Part 1)>
	<1324543253-17750-1-git-send-email-jean-michel.lambert7@wanadoo.fr>
	<B97E3084-CD30-4907-96AB-8242E95E9826@gmail.com>
Message-ID: <4EF4C559.9070504@wanadoo.fr>

thanks Pierre, I was reinventing the wheel. Have recoded and will submit 
a patch for the plugin now.

cheers, JM

On 23/12/2011 18:00, Pierre d'Herbemont wrote:
> Hi,
>
> The same functionality can be achieved by:
> m = libvlc_media_player_get_media(mp);
> libvlc_media_get_meta(m);
>
> That means that, there is no need to:
> 1- modify libvlccore
> 2- require a running input thread to retrieve the meta data
>
> Also the prototype of libvlc_media_get_meta() is better, as it takes a meta data key instead of an index.
>
> Finally, adding a wrapper around the above function calls seems like a better approach... If you thinks such a wrapper is valuable.
>
> Pierre.
>
> On Dec 22, 2011, at 9:40, JM Lambert<jeanmichel.lambert7 at gmail.com>  wrote:
>
>> ---
>> include/vlc/libvlc_media_player.h |    8 ++++++++
>> include/vlc_input.h               |    2 ++
>> lib/libvlc.sym                    |    1 +
>> lib/media_player.c                |   19 +++++++++++++++++++
>> src/input/control.c               |   12 ++++++++++++
>> 5 files changed, 42 insertions(+), 0 deletions(-)
>>
>> diff --git a/include/vlc/libvlc_media_player.h b/include/vlc/libvlc_media_player.h
>> index 152a95d..d361ce2 100644
>> --- a/include/vlc/libvlc_media_player.h
>> +++ b/include/vlc/libvlc_media_player.h
>> @@ -756,6 +756,14 @@ LIBVLC_API int libvlc_media_player_set_rate( libvlc_media_player_t *p_mi, float
>> LIBVLC_API libvlc_state_t libvlc_media_player_get_state( libvlc_media_player_t *p_mi );
>>
>> /**
>> + * Get media meta
>> + *
>> + * \param p_mi the Media Player
>> + * \return ith elemet of meta structure
>> + */
>> +LIBVLC_API char* libvlc_media_player_get_meta( libvlc_media_player_t *p_mi, int i);
>> +
>> +/**
>>   * Get movie fps rate
>>   *
>>   * \param p_mi the Media Player
>> diff --git a/include/vlc_input.h b/include/vlc_input.h
>> index 7d8320a..accb384 100644
>> --- a/include/vlc_input.h
>> +++ b/include/vlc_input.h
>> @@ -516,6 +516,8 @@ enum input_query_e
>>      /* External clock managments */
>>      INPUT_GET_PCR_SYSTEM,   /* arg1=mtime_t *, arg2=mtime_t *       res=can fail */
>>      INPUT_MODIFY_PCR_SYSTEM,/* arg1=int absolute, arg2=mtime_t      res=can fail */
>> +    INPUT_GET_META,/* arg1=vlc_meta_t *      res=can fail */
>> +
>> };
>>
>> /** @}*/
>> diff --git a/lib/libvlc.sym b/lib/libvlc.sym
>> index 45327d5..2607f99 100644
>> --- a/lib/libvlc.sym
>> +++ b/lib/libvlc.sym
>> @@ -120,6 +120,7 @@ libvlc_media_player_get_agl
>> libvlc_media_player_get_chapter
>> libvlc_media_player_get_chapter_count
>> libvlc_media_player_get_chapter_count_for_title
>> +libvlc_media_player_get_meta
>> libvlc_media_player_get_fps
>> libvlc_media_player_get_hwnd
>> libvlc_media_player_get_length
>> diff --git a/lib/media_player.c b/lib/media_player.c
>> index b426637..1978e72 100644
>> --- a/lib/media_player.c
>> +++ b/lib/media_player.c
>> @@ -1217,6 +1217,25 @@ void libvlc_media_player_previous_chapter( libvlc_media_player_t *p_mi )
>>      vlc_object_release( p_input_thread );
>> }
>>
>> +char * libvlc_media_player_get_meta( libvlc_media_player_t *p_mi , int index)
>> +{
>> +    input_thread_t *p_input_thread = libvlc_get_input_thread ( p_mi );
>> +    vlc_meta_t *p_meta;
>> +
>> +    if( p_input_thread )
>> +    {
>> +        if( input_Control( p_input_thread, INPUT_GET_META,&p_meta ) )
>> +                vlc_meta_Set( p_meta, vlc_meta_Title, "unknown" );
>> +        vlc_object_release( p_input_thread );
>> +        if (vlc_meta_Get(p_meta,index)) return vlc_meta_Get(p_meta,index);
>> +        else return "";
>> +    }
>> +    else
>> +    {
>> +        return "";
>> +    }
>> +}
>> +
>> float libvlc_media_player_get_fps( libvlc_media_player_t *p_mi )
>> {
>>      input_thread_t *p_input_thread = libvlc_get_input_thread ( p_mi );
>> diff --git a/src/input/control.c b/src/input/control.c
>> index d50ea97..4e95661 100644
>> --- a/src/input/control.c
>> +++ b/src/input/control.c
>> @@ -65,6 +65,8 @@ int input_vaControl( input_thread_t *p_input, int i_query, va_list args )
>>      seekpoint_t *p_bkmk, ***ppp_bkmk;
>>      int i_bkmk = 0;
>>      int *pi_bkmk;
>> +    vlc_meta_t ** p_meta;
>> +
>>
>>      int i_int, *pi_int;
>>      bool b_bool, *pb_bool;
>> @@ -357,6 +359,16 @@ int input_vaControl( input_thread_t *p_input, int i_query, va_list args )
>>              vlc_mutex_unlock(&p_input->p->p_item->lock );
>>              return VLC_SUCCESS;
>>
>> +        case INPUT_GET_META:
>> +            p_meta = (vlc_meta_t **)va_arg( args, vlc_meta_t ** );
>> +
>> +            vlc_mutex_lock(&p_input->p->p_item->lock );
>> +            input_item_t *p_item = input_GetItem( p_input );
>> +            *p_meta = p_item->p_meta;
>> +            vlc_mutex_unlock(&p_input->p->p_item->lock );
>> +            return VLC_SUCCESS;
>> +
>> +
>>          case INPUT_ADD_SLAVE:
>>              psz = (char*)va_arg( args, char * );
>>              if( psz&&  *psz )
>> -- 
>> 1.7.5.4
>>
>> _______________________________________________
>> vlc-devel mailing list
>> To unsubscribe or modify your subscription options:
>> http://mailman.videolan.org/listinfo/vlc-devel



From fcvlcdev at free.fr  Fri Dec 23 19:24:23 2011
From: fcvlcdev at free.fr (Francois Cartegnie)
Date: Fri, 23 Dec 2011 19:24:23 +0100
Subject: [vlc-devel] (no subject)
Message-ID: <1324664664-11765-1-git-send-email-fcvlcdev@free.fr>

Just to validate the way of splitting current v4l2 decls&defs of structs
 and helpers that could be shared with a v4l2 output module.



From fcvlcdev at free.fr  Fri Dec 23 19:24:24 2011
From: fcvlcdev at free.fr (Francois Cartegnie)
Date: Fri, 23 Dec 2011 19:24:24 +0100
Subject: [vlc-devel] [PATCH] v4l2: split declarations
In-Reply-To: <1324664664-11765-1-git-send-email-fcvlcdev@free.fr>
References: <1324664664-11765-1-git-send-email-fcvlcdev@free.fr>
Message-ID: <1324664664-11765-2-git-send-email-fcvlcdev@free.fr>

---
 modules/access/Modules.am         |    3 +
 modules/access/v4l2/access.c      |    1 +
 modules/access/v4l2/controls.c    |    1 +
 modules/access/v4l2/demux.c       |    1 +
 modules/access/v4l2/v4l2.h        |   24 +-----
 modules/access/v4l2/v4l2_common.c |  162 +++++++++++++++++++++++++++++++++++++
 modules/access/v4l2/v4l2_common.h |   31 +++++++
 modules/access/v4l2/v4l2_input.h  |   46 +++++++++++
 modules/access/v4l2/video.c       |  162 ++++++-------------------------------
 9 files changed, 272 insertions(+), 159 deletions(-)
 create mode 100644 modules/access/v4l2/v4l2_common.c
 create mode 100644 modules/access/v4l2/v4l2_common.h
 create mode 100644 modules/access/v4l2/v4l2_input.h

diff --git a/modules/access/Modules.am b/modules/access/Modules.am
index bbf0c5e..9e86086 100644
--- a/modules/access/Modules.am
+++ b/modules/access/Modules.am
@@ -161,6 +161,9 @@ libv4l2_plugin_la_SOURCES = \
 	v4l2/video.c \
 	v4l2/demux.c \
 	v4l2/access.c \
+	v4l2/v4l2_common.c \
+	v4l2/v4l2_common.h \
+	v4l2/v4l2_input.h \
 	v4l2/controls.c \
 	v4l2/v4l2.h
 libv4l2_plugin_la_CFLAGS = $(AM_CFLAGS) $(LIBV4L2_CFLAGS)
diff --git a/modules/access/v4l2/access.c b/modules/access/v4l2/access.c
index 38348d2..df88c22 100644
--- a/modules/access/v4l2/access.c
+++ b/modules/access/v4l2/access.c
@@ -28,6 +28,7 @@
 #endif
 
 #include "v4l2.h"
+#include "v4l2_input.h"
 #include <vlc_access.h>
 
 #include <errno.h>
diff --git a/modules/access/v4l2/controls.c b/modules/access/v4l2/controls.c
index 2bfbabb..f4309a1 100644
--- a/modules/access/v4l2/controls.c
+++ b/modules/access/v4l2/controls.c
@@ -28,6 +28,7 @@
 #endif
 
 #include "v4l2.h"
+#include "v4l2_input.h"
 #include <ctype.h>
 #include <assert.h>
 #include <sys/ioctl.h>
diff --git a/modules/access/v4l2/demux.c b/modules/access/v4l2/demux.c
index bdbeb95..74c4076 100644
--- a/modules/access/v4l2/demux.c
+++ b/modules/access/v4l2/demux.c
@@ -28,6 +28,7 @@
 #endif
 
 #include "v4l2.h"
+#include "v4l2_input.h"
 #include <vlc_demux.h>
 #include <errno.h>
 #include <sys/ioctl.h>
diff --git a/modules/access/v4l2/v4l2.h b/modules/access/v4l2/v4l2.h
index 8b227d1..91207e6 100644
--- a/modules/access/v4l2/v4l2.h
+++ b/modules/access/v4l2/v4l2.h
@@ -69,8 +69,6 @@
 #   define v4l2_munmap munmap
 #endif
 
-#define CFG_PREFIX "v4l2-"
-
 /* TODO: remove this, use callbacks */
 typedef enum {
     IO_METHOD_READ=1,
@@ -92,7 +90,7 @@ struct demux_sys_t
     unsigned int i_nbuffers;
 #define blocksize i_nbuffers /* HACK HACK */
 
-    int i_fourcc;
+    vlc_fourcc_t i_fourcc;
     uint32_t i_block_flags;
 
     es_out_id_t *p_es;
@@ -109,23 +107,3 @@ struct buffer_t
     void *  start;
     size_t  length;
 };
-
-/* video.c */
-void ParseMRL(vlc_object_t *, const char *);
-int OpenVideo(vlc_object_t *, demux_sys_t *, bool);
-block_t* GrabVideo(vlc_object_t *, demux_sys_t *);
-
-/* demux.c */
-int DemuxOpen(vlc_object_t *);
-void DemuxClose(vlc_object_t *);
-float GetAbsoluteMaxFrameRate(vlc_object_t *, int fd, uint32_t fmt);
-void GetMaxDimensions(vlc_object_t *, int fd, uint32_t fmt, float fps_min,
-                      uint32_t *pwidth, uint32_t *pheight);
-
-/* access.c */
-int AccessOpen(vlc_object_t *);
-void AccessClose(vlc_object_t *);
-
-/* controls.c */
-vlc_v4l2_ctrl_t *ControlsInit(vlc_object_t *, int fd);
-void ControlsDeinit(vlc_object_t *, vlc_v4l2_ctrl_t *);
diff --git a/modules/access/v4l2/v4l2_common.c b/modules/access/v4l2/v4l2_common.c
new file mode 100644
index 0000000..8ece257
--- /dev/null
+++ b/modules/access/v4l2/v4l2_common.c
@@ -0,0 +1,162 @@
+/*****************************************************************************
+ * v4l2_common.c : Video4Linux2 input module for vlc
+ *****************************************************************************
+ * Copyright (C) 2002-2011 the VideoLAN team
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston MA 02110-1301, USA.
+ *****************************************************************************/
+
+#ifdef HAVE_CONFIG_H
+# include "config.h"
+#endif
+
+#include "v4l2_common.h"
+#include "v4l2.h"
+
+#include <vlc_fourcc.h>
+
+#include <sys/ioctl.h>
+#include <sys/mman.h>
+
+static const struct
+{
+    unsigned int i_v4l2;
+    vlc_fourcc_t i_fourcc;
+    int i_rmask;
+    int i_gmask;
+    int i_bmask;
+} v4l2chroma_to_fourcc[] =
+{
+    /* Raw data types */
+    { V4L2_PIX_FMT_GREY,    VLC_CODEC_GREY, 0, 0, 0 },
+    { V4L2_PIX_FMT_HI240,   VLC_FOURCC('I','2','4','0'), 0, 0, 0 },
+    { V4L2_PIX_FMT_RGB555,  VLC_CODEC_RGB15, 0x001f,0x03e0,0x7c00 },
+    { V4L2_PIX_FMT_RGB565,  VLC_CODEC_RGB16, 0x001f,0x07e0,0xf800 },
+    /* Won't work since we don't know how to handle such gmask values
+     * correctly
+    { V4L2_PIX_FMT_RGB555X, VLC_CODEC_RGB15, 0x007c,0xe003,0x1f00 },
+    { V4L2_PIX_FMT_RGB565X, VLC_CODEC_RGB16, 0x00f8,0xe007,0x1f00 },
+    */
+    { V4L2_PIX_FMT_BGR24,   VLC_CODEC_RGB24, 0xff0000,0xff00,0xff },
+    { V4L2_PIX_FMT_RGB24,   VLC_CODEC_RGB24, 0xff,0xff00,0xff0000 },
+    { V4L2_PIX_FMT_BGR32,   VLC_CODEC_RGB32, 0xff0000,0xff00,0xff },
+    { V4L2_PIX_FMT_RGB32,   VLC_CODEC_RGB32, 0xff,0xff00,0xff0000 },
+    { V4L2_PIX_FMT_YUYV,    VLC_CODEC_YUYV, 0, 0, 0 },
+    { V4L2_PIX_FMT_UYVY,    VLC_CODEC_UYVY, 0, 0, 0 },
+    { V4L2_PIX_FMT_Y41P,    VLC_FOURCC('I','4','1','N'), 0, 0, 0 },
+    { V4L2_PIX_FMT_YUV422P, VLC_CODEC_I422, 0, 0, 0 },
+    { V4L2_PIX_FMT_YVU420,  VLC_CODEC_YV12, 0, 0, 0 },
+    { V4L2_PIX_FMT_YUV411P, VLC_CODEC_I411, 0, 0, 0 },
+    { V4L2_PIX_FMT_YUV410,  VLC_CODEC_I410, 0, 0, 0 },
+
+    /* Raw data types, not in V4L2 spec but still in videodev2.h and supported
+     * by VLC */
+    { V4L2_PIX_FMT_YUV420,  VLC_CODEC_I420, 0, 0, 0 },
+    /* FIXME { V4L2_PIX_FMT_RGB444,  VLC_CODEC_RGB32 }, */
+
+    /* Compressed data types */
+    { V4L2_PIX_FMT_MJPEG,   VLC_CODEC_MJPG, 0, 0, 0 },
+    { V4L2_PIX_FMT_JPEG,    VLC_CODEC_JPEG, 0, 0, 0 },
+#if 0
+    { V4L2_PIX_FMT_DV,      VLC_FOURCC('?','?','?','?') },
+    { V4L2_PIX_FMT_MPEG,    VLC_FOURCC('?','?','?','?') },
+#endif
+    { 0, 0, 0, 0, 0 }
+};
+
+bool get_v4l2pixelformat_by_fourcc( unsigned int *found, vlc_fourcc_t i_requested_fourcc )
+{
+    for( int i = 0; v4l2chroma_to_fourcc[i].i_v4l2 != 0; i++ )
+    {
+        if( v4l2chroma_to_fourcc[i].i_fourcc == i_requested_fourcc )
+        {
+            *found = v4l2chroma_to_fourcc[i].i_v4l2;
+            return true;
+        }
+    }
+    return false;
+}
+
+bool get_fourcc_by_v4l2pixelformat( vlc_fourcc_t *i_fourcc, int *i_rmask, int *i_gmask, int *i_bmask, unsigned int pixelformat )
+{
+    for( int i = 0; v4l2chroma_to_fourcc[i].i_fourcc != 0; i++ )
+    {
+        if( v4l2chroma_to_fourcc[i].i_v4l2 == pixelformat )
+        {
+            *i_fourcc = v4l2chroma_to_fourcc[i].i_fourcc;
+            *i_rmask = v4l2chroma_to_fourcc[i].i_rmask;
+            *i_gmask = v4l2chroma_to_fourcc[i].i_gmask;
+            *i_bmask = v4l2chroma_to_fourcc[i].i_bmask;
+            return true;
+        }
+    }
+    return false;
+}
+/*****************************************************************************
+ * Helper function to initalise video IO using the mmap method
+ *****************************************************************************/
+int InitMmap( vlc_object_t *p_demux, demux_sys_t *p_sys, int i_fd )
+{
+    struct v4l2_requestbuffers req;
+
+    memset( &req, 0, sizeof(req) );
+    req.count = 4;
+    req.type = V4L2_BUF_TYPE_VIDEO_CAPTURE;
+    req.memory = V4L2_MEMORY_MMAP;
+
+    if( v4l2_ioctl( i_fd, VIDIOC_REQBUFS, &req ) < 0 )
+    {
+        msg_Err( p_demux, "device does not support mmap I/O" );
+        return -1;
+    }
+
+    if( req.count < 2 )
+    {
+        msg_Err( p_demux, "insufficient buffers" );
+        return -1;
+    }
+
+    p_sys->p_buffers = calloc( req.count, sizeof( *p_sys->p_buffers ) );
+    if( unlikely(!p_sys->p_buffers) )
+        return -1;
+
+    for( p_sys->i_nbuffers = 0; p_sys->i_nbuffers < req.count; ++p_sys->i_nbuffers )
+    {
+        struct v4l2_buffer buf;
+
+        memset( &buf, 0, sizeof(buf) );
+        buf.type = V4L2_BUF_TYPE_VIDEO_CAPTURE;
+        buf.memory = V4L2_MEMORY_MMAP;
+        buf.index = p_sys->i_nbuffers;
+
+        if( v4l2_ioctl( i_fd, VIDIOC_QUERYBUF, &buf ) < 0 )
+        {
+            msg_Err( p_demux, "VIDIOC_QUERYBUF: %m" );
+            return -1;
+        }
+
+        p_sys->p_buffers[p_sys->i_nbuffers].length = buf.length;
+        p_sys->p_buffers[p_sys->i_nbuffers].start =
+            v4l2_mmap( NULL, buf.length, PROT_READ | PROT_WRITE, MAP_SHARED, i_fd, buf.m.offset );
+
+        if( p_sys->p_buffers[p_sys->i_nbuffers].start == MAP_FAILED )
+        {
+            msg_Err( p_demux, "mmap failed: %m" );
+            return -1;
+        }
+    }
+
+    return 0;
+}
diff --git a/modules/access/v4l2/v4l2_common.h b/modules/access/v4l2/v4l2_common.h
new file mode 100644
index 0000000..c56b7b3
--- /dev/null
+++ b/modules/access/v4l2/v4l2_common.h
@@ -0,0 +1,31 @@
+/*****************************************************************************
+ * v4l2_common.h : Video4Linux2 input module for vlc
+ *****************************************************************************
+ * Copyright (C) 2002-2011 the VideoLAN team
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston MA 02110-1301, USA.
+ *****************************************************************************/
+#ifndef V4L2_COMMON_H
+#define V4L2_COMMON_H
+
+#include <vlc_common.h>
+
+int InitMmap( vlc_object_t *p_demux, demux_sys_t *p_sys, int i_fd );
+
+bool get_fourcc_by_v4l2pixelformat( vlc_fourcc_t *i_fourcc, int *i_rmask, int *i_gmask, int *i_bmask, unsigned int pixelformat );
+bool get_v4l2pixelformat_by_fourcc( unsigned int *found, vlc_fourcc_t i_requested_fourcc );
+
+
+#endif // V4L2_COMMON_H
diff --git a/modules/access/v4l2/v4l2_input.h b/modules/access/v4l2/v4l2_input.h
new file mode 100644
index 0000000..ffb102a
--- /dev/null
+++ b/modules/access/v4l2/v4l2_input.h
@@ -0,0 +1,46 @@
+/*****************************************************************************
+ * v4l2_input.h : Video4Linux2 input module for vlc
+ *****************************************************************************
+ * Copyright (C) 2002-2011 the VideoLAN team
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston MA 02110-1301, USA.
+ *****************************************************************************/
+
+#ifndef V4L2_INPUT_H
+#define V4L2_INPUT_H
+
+#define CFG_PREFIX "v4l2-"
+
+/* video.c */
+void ParseMRL(vlc_object_t *, const char *);
+int OpenVideo(vlc_object_t *, demux_sys_t *, bool);
+block_t* GrabVideo(vlc_object_t *, demux_sys_t *);
+
+/* demux.c */
+int DemuxOpen(vlc_object_t *);
+void DemuxClose(vlc_object_t *);
+float GetAbsoluteMaxFrameRate(vlc_object_t *, int fd, uint32_t fmt);
+void GetMaxDimensions(vlc_object_t *, int fd, uint32_t fmt, float fps_min,
+                      uint32_t *pwidth, uint32_t *pheight);
+
+/* access.c */
+int AccessOpen(vlc_object_t *);
+void AccessClose(vlc_object_t *);
+
+/* controls.c */
+vlc_v4l2_ctrl_t *ControlsInit(vlc_object_t *, int fd);
+void ControlsDeinit(vlc_object_t *, vlc_v4l2_ctrl_t *);
+
+#endif // V4L2_INPUT_H
diff --git a/modules/access/v4l2/video.c b/modules/access/v4l2/video.c
index 951213d..5c33934 100644
--- a/modules/access/v4l2/video.c
+++ b/modules/access/v4l2/video.c
@@ -38,6 +38,8 @@
 #endif
 
 #include "v4l2.h"
+#include "v4l2_input.h"
+#include "v4l2_common.h"
 #include <vlc_plugin.h>
 #include <vlc_fs.h>
 #include <vlc_demux.h>
@@ -460,52 +462,6 @@ vlc_module_end ()
 
 static block_t* ProcessVideoFrame( vlc_object_t *p_demux, uint8_t *p_frame, size_t );
 
-static const struct
-{
-    unsigned int i_v4l2;
-    vlc_fourcc_t i_fourcc;
-    int i_rmask;
-    int i_gmask;
-    int i_bmask;
-} v4l2chroma_to_fourcc[] =
-{
-    /* Raw data types */
-    { V4L2_PIX_FMT_GREY,    VLC_CODEC_GREY, 0, 0, 0 },
-    { V4L2_PIX_FMT_HI240,   VLC_FOURCC('I','2','4','0'), 0, 0, 0 },
-    { V4L2_PIX_FMT_RGB555,  VLC_CODEC_RGB15, 0x001f,0x03e0,0x7c00 },
-    { V4L2_PIX_FMT_RGB565,  VLC_CODEC_RGB16, 0x001f,0x07e0,0xf800 },
-    /* Won't work since we don't know how to handle such gmask values
-     * correctly
-    { V4L2_PIX_FMT_RGB555X, VLC_CODEC_RGB15, 0x007c,0xe003,0x1f00 },
-    { V4L2_PIX_FMT_RGB565X, VLC_CODEC_RGB16, 0x00f8,0xe007,0x1f00 },
-    */
-    { V4L2_PIX_FMT_BGR24,   VLC_CODEC_RGB24, 0xff0000,0xff00,0xff },
-    { V4L2_PIX_FMT_RGB24,   VLC_CODEC_RGB24, 0xff,0xff00,0xff0000 },
-    { V4L2_PIX_FMT_BGR32,   VLC_CODEC_RGB32, 0xff0000,0xff00,0xff },
-    { V4L2_PIX_FMT_RGB32,   VLC_CODEC_RGB32, 0xff,0xff00,0xff0000 },
-    { V4L2_PIX_FMT_YUYV,    VLC_CODEC_YUYV, 0, 0, 0 },
-    { V4L2_PIX_FMT_UYVY,    VLC_CODEC_UYVY, 0, 0, 0 },
-    { V4L2_PIX_FMT_Y41P,    VLC_FOURCC('I','4','1','N'), 0, 0, 0 },
-    { V4L2_PIX_FMT_YUV422P, VLC_CODEC_I422, 0, 0, 0 },
-    { V4L2_PIX_FMT_YVU420,  VLC_CODEC_YV12, 0, 0, 0 },
-    { V4L2_PIX_FMT_YUV411P, VLC_CODEC_I411, 0, 0, 0 },
-    { V4L2_PIX_FMT_YUV410,  VLC_CODEC_I410, 0, 0, 0 },
-
-    /* Raw data types, not in V4L2 spec but still in videodev2.h and supported
-     * by VLC */
-    { V4L2_PIX_FMT_YUV420,  VLC_CODEC_I420, 0, 0, 0 },
-    /* FIXME { V4L2_PIX_FMT_RGB444,  VLC_CODEC_RGB32 }, */
-
-    /* Compressed data types */
-    { V4L2_PIX_FMT_MJPEG,   VLC_CODEC_MJPG, 0, 0, 0 },
-    { V4L2_PIX_FMT_JPEG,    VLC_CODEC_JPEG, 0, 0, 0 },
-#if 0
-    { V4L2_PIX_FMT_DV,      VLC_FOURCC('?','?','?','?') },
-    { V4L2_PIX_FMT_MPEG,    VLC_FOURCC('?','?','?','?') },
-#endif
-    { 0, 0, 0, 0, 0 }
-};
-
 /**
  * List of V4L2 chromas were confident enough to use as fallbacks if the
  * user hasn't provided a --v4l2-chroma value.
@@ -673,63 +629,6 @@ static block_t* ProcessVideoFrame( vlc_object_t *p_demux, uint8_t *p_frame, size
 }
 
 /*****************************************************************************
- * Helper function to initalise video IO using the mmap method
- *****************************************************************************/
-static int InitMmap( vlc_object_t *p_demux, demux_sys_t *p_sys, int i_fd )
-{
-    struct v4l2_requestbuffers req;
-
-    memset( &req, 0, sizeof(req) );
-    req.count = 4;
-    req.type = V4L2_BUF_TYPE_VIDEO_CAPTURE;
-    req.memory = V4L2_MEMORY_MMAP;
-
-    if( v4l2_ioctl( i_fd, VIDIOC_REQBUFS, &req ) < 0 )
-    {
-        msg_Err( p_demux, "device does not support mmap I/O" );
-        return -1;
-    }
-
-    if( req.count < 2 )
-    {
-        msg_Err( p_demux, "insufficient buffers" );
-        return -1;
-    }
-
-    p_sys->p_buffers = calloc( req.count, sizeof( *p_sys->p_buffers ) );
-    if( unlikely(!p_sys->p_buffers) )
-        return -1;
-
-    for( p_sys->i_nbuffers = 0; p_sys->i_nbuffers < req.count; ++p_sys->i_nbuffers )
-    {
-        struct v4l2_buffer buf;
-
-        memset( &buf, 0, sizeof(buf) );
-        buf.type = V4L2_BUF_TYPE_VIDEO_CAPTURE;
-        buf.memory = V4L2_MEMORY_MMAP;
-        buf.index = p_sys->i_nbuffers;
-
-        if( v4l2_ioctl( i_fd, VIDIOC_QUERYBUF, &buf ) < 0 )
-        {
-            msg_Err( p_demux, "VIDIOC_QUERYBUF: %m" );
-            return -1;
-        }
-
-        p_sys->p_buffers[p_sys->i_nbuffers].length = buf.length;
-        p_sys->p_buffers[p_sys->i_nbuffers].start =
-            v4l2_mmap( NULL, buf.length, PROT_READ | PROT_WRITE, MAP_SHARED, i_fd, buf.m.offset );
-
-        if( p_sys->p_buffers[p_sys->i_nbuffers].start == MAP_FAILED )
-        {
-            msg_Err( p_demux, "mmap failed: %m" );
-            return -1;
-        }
-    }
-
-    return 0;
-}
-
-/*****************************************************************************
  * Helper function to initalise video IO using the userbuf method
  *****************************************************************************/
 static int InitUserP( vlc_object_t *p_demux, demux_sys_t *p_sys, int i_fd, unsigned int i_buffer_size )
@@ -1071,21 +970,17 @@ static int InitVideo( vlc_object_t *p_obj, int i_fd, demux_sys_t *p_sys,
             vlc_fourcc_to_char( codecs[i].pixelformat, fourcc_v4l2 );
 
             bool b_codec_supported = false;
-            for( unsigned j = 0; v4l2chroma_to_fourcc[j].i_v4l2 != 0; j++ )
-            {
-                if( v4l2chroma_to_fourcc[j].i_v4l2 == codecs[i].pixelformat )
-                {
-                    char fourcc[5];
-                    memset( fourcc, 0, sizeof( fourcc ) );
-                    vlc_fourcc_to_char( v4l2chroma_to_fourcc[j].i_fourcc,
-                                        fourcc );
-                    msg_Dbg( p_obj, "device supports chroma %4.4s [%s, %s]",
-                             fourcc, codecs[i].description, fourcc_v4l2 );
-                    b_codec_supported = true;
-                }
-            }
-            if( !b_codec_supported )
+            vlc_fourcc_t i_fourcc;
+            int i_rmask, i_gmask, i_bmask;
+            b_codec_supported = get_fourcc_by_v4l2pixelformat( &i_fourcc, &i_rmask, &i_gmask, &i_bmask, codecs[i].pixelformat );
+            if( b_codec_supported )
             {
+                char fourcc[5];
+                memset( fourcc, 0, sizeof( fourcc ) );
+                vlc_fourcc_to_char( i_fourcc, fourcc );
+                msg_Dbg( p_obj, "device supports chroma %4.4s [%s, %s]",
+                         fourcc, codecs[i].description, fourcc_v4l2 );
+            } else {
                 msg_Dbg( p_obj, "device codec %4.4s (%s) not supported",
                          fourcc_v4l2, codecs[i].description );
             }
@@ -1167,14 +1062,8 @@ static int InitVideo( vlc_object_t *p_obj, int i_fd, demux_sys_t *p_sys,
             const vlc_fourcc_t i_requested_fourcc =
                 vlc_fourcc_GetCodecFromString( VIDEO_ES, reqchroma );
 
-            for( int i = 0; v4l2chroma_to_fourcc[i].i_v4l2 != 0; i++ )
-            {
-                if( v4l2chroma_to_fourcc[i].i_fourcc == i_requested_fourcc )
-                {
-                    fmt.fmt.pix.pixelformat = v4l2chroma_to_fourcc[i].i_v4l2;
-                    break;
-                }
-            }
+            get_v4l2pixelformat_by_fourcc( &fmt.fmt.pix.pixelformat, i_requested_fourcc );
+
             /* Try and set user chroma */
             bool b_error = !IsPixelFormatSupported( codecs, ncodec,
                                                     fmt.fmt.pix.pixelformat );
@@ -1309,18 +1198,19 @@ static int InitVideo( vlc_object_t *p_obj, int i_fd, demux_sys_t *p_sys,
     }
 
     /* Look up final fourcc */
-    p_sys->i_fourcc = 0;
-    for( int i = 0; v4l2chroma_to_fourcc[i].i_fourcc != 0; i++ )
+    int i_rmask, i_gmask, i_bmask;
+    if ( get_fourcc_by_v4l2pixelformat( &p_sys->i_fourcc,
+                                        &i_rmask,
+                                        &i_gmask,
+                                        &i_bmask,
+                                        fmt.fmt.pix.pixelformat ) )
     {
-        if( v4l2chroma_to_fourcc[i].i_v4l2 == fmt.fmt.pix.pixelformat )
-        {
-            p_sys->i_fourcc = v4l2chroma_to_fourcc[i].i_fourcc;
-            es_format_Init( &es_fmt, VIDEO_ES, p_sys->i_fourcc );
-            es_fmt.video.i_rmask = v4l2chroma_to_fourcc[i].i_rmask;
-            es_fmt.video.i_gmask = v4l2chroma_to_fourcc[i].i_gmask;
-            es_fmt.video.i_bmask = v4l2chroma_to_fourcc[i].i_bmask;
-            break;
-        }
+        es_format_Init( &es_fmt, VIDEO_ES, p_sys->i_fourcc );
+        es_fmt.video.i_rmask = i_rmask;
+        es_fmt.video.i_gmask = i_gmask;
+        es_fmt.video.i_bmask = i_bmask;
+    } else {
+        p_sys->i_fourcc = 0;
     }
 
     /* Buggy driver paranoia */
-- 
1.7.6



From zhushisongzhu at yahoo.com  Fri Dec 23 19:27:41 2011
From: zhushisongzhu at yahoo.com (zhu shi song)
Date: Fri, 23 Dec 2011 10:27:41 -0800 (PST)
Subject: [vlc-devel] How to trim/cut clips with VLC
Message-ID: <1324664861.86424.YahooMailNeo@web125701.mail.ne1.yahoo.com>

I using the following command to transcode one piece of video from start-time to stop-time.

vlc hzvideo.wmv --start-time=10 --stop-time=20 --sout=#transcode{vcodec=h264, venc=x264{preset=veryfast}, vb=0,scale=0,acodec=mp4a,
ab=128,channels=2,samplerate=44100}:file{dst='11.mp4'}

when reaching stop-time, the transcoding ended.? But start-time had no effect, it always started from the beginning.? What's wrong with it?
Thanks



From typx at dinauz.org  Fri Dec 23 20:41:19 2011
From: typx at dinauz.org (Denis Charmet)
Date: Fri, 23 Dec 2011 20:41:19 +0100
Subject: [vlc-devel] [PATCH] Avoid a crash with some corrupted mkvs
Message-ID: <1324669279-29319-1-git-send-email-typx@dinauz.org>

Should close #5659
---
 modules/demux/mkv/matroska_segment.cpp |    5 +++++
 1 files changed, 5 insertions(+), 0 deletions(-)

diff --git a/modules/demux/mkv/matroska_segment.cpp b/modules/demux/mkv/matroska_segment.cpp
index 2bef45e..072dd71 100644
--- a/modules/demux/mkv/matroska_segment.cpp
+++ b/modules/demux/mkv/matroska_segment.cpp
@@ -930,6 +930,11 @@ bool matroska_segment_c::Select( mtime_t i_start_time )
                 p_tk->fmt.i_extra       = GetDWLE( &p_bih->biSize ) - sizeof( BITMAPINFOHEADER );
                 if( p_tk->fmt.i_extra > 0 )
                 {
+                    /* Very unlikely yet possible: bug #5659*/
+                    size_t maxlen = p_tk->i_extra_data - sizeof( BITMAPINFOHEADER );
+                    p_tk->fmt.i_extra = ( p_tk->fmt.i_extra < maxlen )?
+                        p_tk->fmt.i_extra : maxlen;
+
                     p_tk->fmt.p_extra = xmalloc( p_tk->fmt.i_extra );
                     memcpy( p_tk->fmt.p_extra, &p_bih[1], p_tk->fmt.i_extra );
                 }
-- 
1.7.7.3



From remi at remlab.net  Fri Dec 23 20:45:08 2011
From: remi at remlab.net (=?iso-8859-1?q?R=E9mi?= Denis-Courmont)
Date: Fri, 23 Dec 2011 20:45:08 +0100
Subject: [vlc-devel] [PATCH] Avoid a crash with some corrupted mkvs
In-Reply-To: <1324669279-29319-1-git-send-email-typx@dinauz.org>
References: <1324669279-29319-1-git-send-email-typx@dinauz.org>
Message-ID: <201112232045.08964.remi@remlab.net>

   Hello,

On Friday 23 December 2011, Denis Charmet wrote:
> Should close #5659
> ---
>  modules/demux/mkv/matroska_segment.cpp |    5 +++++
>  1 files changed, 5 insertions(+), 0 deletions(-)
> 
> diff --git a/modules/demux/mkv/matroska_segment.cpp
> b/modules/demux/mkv/matroska_segment.cpp index 2bef45e..072dd71 100644
> --- a/modules/demux/mkv/matroska_segment.cpp
> +++ b/modules/demux/mkv/matroska_segment.cpp
> @@ -930,6 +930,11 @@ bool matroska_segment_c::Select( mtime_t i_start_time
> ) p_tk->fmt.i_extra       = GetDWLE( &p_bih->biSize ) - sizeof(
> BITMAPINFOHEADER ); if( p_tk->fmt.i_extra > 0 )
>                  {
> +                    /* Very unlikely yet possible: bug #5659*/
> +                    size_t maxlen = p_tk->i_extra_data - sizeof(
> BITMAPINFOHEADER );

This is not very kosher if i_extra_data < sizeof(BITMAPINFOHEADER).

> +                    p_tk->fmt.i_extra = (
> p_tk->fmt.i_extra < maxlen )? +                        p_tk->fmt.i_extra :
> maxlen;
> +
>                      p_tk->fmt.p_extra = xmalloc( p_tk->fmt.i_extra );
>                      memcpy( p_tk->fmt.p_extra, &p_bih[1],
> p_tk->fmt.i_extra ); }
-- 
R?mi Denis-Courmont
http://www.remlab.info


From remi at remlab.net  Fri Dec 23 20:47:51 2011
From: remi at remlab.net (=?iso-8859-1?q?R=E9mi?= Denis-Courmont)
Date: Fri, 23 Dec 2011 20:47:51 +0100
Subject: [vlc-devel] [PATCH] v4l2: split declarations
In-Reply-To: <1324664664-11765-2-git-send-email-fcvlcdev@free.fr>
References: <1324664664-11765-1-git-send-email-fcvlcdev@free.fr>
	<1324664664-11765-2-git-send-email-fcvlcdev@free.fr>
Message-ID: <201112232047.52152.remi@remlab.net>

   Hello,

I don't see the point in factoring demux-specific mmap stuff. It can't work. And 
it will make my work on zero-copy V4L2 input needlessly more complicated. It's 
already bad enough with the V4L2 access that nobody seems to use.

On Friday 23 December 2011, Francois Cartegnie wrote:
> ---
>  modules/access/Modules.am         |    3 +
>  modules/access/v4l2/access.c      |    1 +
>  modules/access/v4l2/controls.c    |    1 +
>  modules/access/v4l2/demux.c       |    1 +
>  modules/access/v4l2/v4l2.h        |   24 +-----
>  modules/access/v4l2/v4l2_common.c |  162
> +++++++++++++++++++++++++++++++++++++ modules/access/v4l2/v4l2_common.h | 
>  31 +++++++
>  modules/access/v4l2/v4l2_input.h  |   46 +++++++++++
>  modules/access/v4l2/video.c       |  162
> ++++++------------------------------- 9 files changed, 272 insertions(+),
> 159 deletions(-)
>  create mode 100644 modules/access/v4l2/v4l2_common.c
>  create mode 100644 modules/access/v4l2/v4l2_common.h
>  create mode 100644 modules/access/v4l2/v4l2_input.h
> 
> diff --git a/modules/access/Modules.am b/modules/access/Modules.am
> index bbf0c5e..9e86086 100644
> --- a/modules/access/Modules.am
> +++ b/modules/access/Modules.am
> @@ -161,6 +161,9 @@ libv4l2_plugin_la_SOURCES = \
>  	v4l2/video.c \
>  	v4l2/demux.c \
>  	v4l2/access.c \
> +	v4l2/v4l2_common.c \
> +	v4l2/v4l2_common.h \
> +	v4l2/v4l2_input.h \
>  	v4l2/controls.c \
>  	v4l2/v4l2.h
>  libv4l2_plugin_la_CFLAGS = $(AM_CFLAGS) $(LIBV4L2_CFLAGS)
> diff --git a/modules/access/v4l2/access.c b/modules/access/v4l2/access.c
> index 38348d2..df88c22 100644
> --- a/modules/access/v4l2/access.c
> +++ b/modules/access/v4l2/access.c
> @@ -28,6 +28,7 @@
>  #endif
> 
>  #include "v4l2.h"
> +#include "v4l2_input.h"
>  #include <vlc_access.h>
> 
>  #include <errno.h>
> diff --git a/modules/access/v4l2/controls.c
> b/modules/access/v4l2/controls.c index 2bfbabb..f4309a1 100644
> --- a/modules/access/v4l2/controls.c
> +++ b/modules/access/v4l2/controls.c
> @@ -28,6 +28,7 @@
>  #endif
> 
>  #include "v4l2.h"
> +#include "v4l2_input.h"
>  #include <ctype.h>
>  #include <assert.h>
>  #include <sys/ioctl.h>
> diff --git a/modules/access/v4l2/demux.c b/modules/access/v4l2/demux.c
> index bdbeb95..74c4076 100644
> --- a/modules/access/v4l2/demux.c
> +++ b/modules/access/v4l2/demux.c
> @@ -28,6 +28,7 @@
>  #endif
> 
>  #include "v4l2.h"
> +#include "v4l2_input.h"
>  #include <vlc_demux.h>
>  #include <errno.h>
>  #include <sys/ioctl.h>
> diff --git a/modules/access/v4l2/v4l2.h b/modules/access/v4l2/v4l2.h
> index 8b227d1..91207e6 100644
> --- a/modules/access/v4l2/v4l2.h
> +++ b/modules/access/v4l2/v4l2.h
> @@ -69,8 +69,6 @@
>  #   define v4l2_munmap munmap
>  #endif
> 
> -#define CFG_PREFIX "v4l2-"
> -
>  /* TODO: remove this, use callbacks */
>  typedef enum {
>      IO_METHOD_READ=1,
> @@ -92,7 +90,7 @@ struct demux_sys_t
>      unsigned int i_nbuffers;
>  #define blocksize i_nbuffers /* HACK HACK */
> 
> -    int i_fourcc;
> +    vlc_fourcc_t i_fourcc;
>      uint32_t i_block_flags;
> 
>      es_out_id_t *p_es;
> @@ -109,23 +107,3 @@ struct buffer_t
>      void *  start;
>      size_t  length;
>  };
> -
> -/* video.c */
> -void ParseMRL(vlc_object_t *, const char *);
> -int OpenVideo(vlc_object_t *, demux_sys_t *, bool);
> -block_t* GrabVideo(vlc_object_t *, demux_sys_t *);
> -
> -/* demux.c */
> -int DemuxOpen(vlc_object_t *);
> -void DemuxClose(vlc_object_t *);
> -float GetAbsoluteMaxFrameRate(vlc_object_t *, int fd, uint32_t fmt);
> -void GetMaxDimensions(vlc_object_t *, int fd, uint32_t fmt, float fps_min,
> -                      uint32_t *pwidth, uint32_t *pheight);
> -
> -/* access.c */
> -int AccessOpen(vlc_object_t *);
> -void AccessClose(vlc_object_t *);
> -
> -/* controls.c */
> -vlc_v4l2_ctrl_t *ControlsInit(vlc_object_t *, int fd);
> -void ControlsDeinit(vlc_object_t *, vlc_v4l2_ctrl_t *);
> diff --git a/modules/access/v4l2/v4l2_common.c
> b/modules/access/v4l2/v4l2_common.c new file mode 100644
> index 0000000..8ece257
> --- /dev/null
> +++ b/modules/access/v4l2/v4l2_common.c
> @@ -0,0 +1,162 @@
> +/*************************************************************************
> **** + * v4l2_common.c : Video4Linux2 input module for vlc
> +
> **************************************************************************
> *** + * Copyright (C) 2002-2011 the VideoLAN team
> + *
> + * This program is free software; you can redistribute it and/or modify
> + * it under the terms of the GNU General Public License as published by
> + * the Free Software Foundation; either version 2 of the License, or
> + * (at your option) any later version.
> + *
> + * This program is distributed in the hope that it will be useful,
> + * but WITHOUT ANY WARRANTY; without even the implied warranty of
> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
> + * GNU General Public License for more details.
> + *
> + * You should have received a copy of the GNU General Public License
> + * along with this program; if not, write to the Free Software
> + * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston MA
> 02110-1301, USA. +
> **************************************************************************
> ***/ +
> +#ifdef HAVE_CONFIG_H
> +# include "config.h"
> +#endif
> +
> +#include "v4l2_common.h"
> +#include "v4l2.h"
> +
> +#include <vlc_fourcc.h>
> +
> +#include <sys/ioctl.h>
> +#include <sys/mman.h>
> +
> +static const struct
> +{
> +    unsigned int i_v4l2;
> +    vlc_fourcc_t i_fourcc;
> +    int i_rmask;
> +    int i_gmask;
> +    int i_bmask;
> +} v4l2chroma_to_fourcc[] =
> +{
> +    /* Raw data types */
> +    { V4L2_PIX_FMT_GREY,    VLC_CODEC_GREY, 0, 0, 0 },
> +    { V4L2_PIX_FMT_HI240,   VLC_FOURCC('I','2','4','0'), 0, 0, 0 },
> +    { V4L2_PIX_FMT_RGB555,  VLC_CODEC_RGB15, 0x001f,0x03e0,0x7c00 },
> +    { V4L2_PIX_FMT_RGB565,  VLC_CODEC_RGB16, 0x001f,0x07e0,0xf800 },
> +    /* Won't work since we don't know how to handle such gmask values
> +     * correctly
> +    { V4L2_PIX_FMT_RGB555X, VLC_CODEC_RGB15, 0x007c,0xe003,0x1f00 },
> +    { V4L2_PIX_FMT_RGB565X, VLC_CODEC_RGB16, 0x00f8,0xe007,0x1f00 },
> +    */
> +    { V4L2_PIX_FMT_BGR24,   VLC_CODEC_RGB24, 0xff0000,0xff00,0xff },
> +    { V4L2_PIX_FMT_RGB24,   VLC_CODEC_RGB24, 0xff,0xff00,0xff0000 },
> +    { V4L2_PIX_FMT_BGR32,   VLC_CODEC_RGB32, 0xff0000,0xff00,0xff },
> +    { V4L2_PIX_FMT_RGB32,   VLC_CODEC_RGB32, 0xff,0xff00,0xff0000 },
> +    { V4L2_PIX_FMT_YUYV,    VLC_CODEC_YUYV, 0, 0, 0 },
> +    { V4L2_PIX_FMT_UYVY,    VLC_CODEC_UYVY, 0, 0, 0 },
> +    { V4L2_PIX_FMT_Y41P,    VLC_FOURCC('I','4','1','N'), 0, 0, 0 },
> +    { V4L2_PIX_FMT_YUV422P, VLC_CODEC_I422, 0, 0, 0 },
> +    { V4L2_PIX_FMT_YVU420,  VLC_CODEC_YV12, 0, 0, 0 },
> +    { V4L2_PIX_FMT_YUV411P, VLC_CODEC_I411, 0, 0, 0 },
> +    { V4L2_PIX_FMT_YUV410,  VLC_CODEC_I410, 0, 0, 0 },
> +
> +    /* Raw data types, not in V4L2 spec but still in videodev2.h and
> supported +     * by VLC */
> +    { V4L2_PIX_FMT_YUV420,  VLC_CODEC_I420, 0, 0, 0 },
> +    /* FIXME { V4L2_PIX_FMT_RGB444,  VLC_CODEC_RGB32 }, */
> +
> +    /* Compressed data types */
> +    { V4L2_PIX_FMT_MJPEG,   VLC_CODEC_MJPG, 0, 0, 0 },
> +    { V4L2_PIX_FMT_JPEG,    VLC_CODEC_JPEG, 0, 0, 0 },
> +#if 0
> +    { V4L2_PIX_FMT_DV,      VLC_FOURCC('?','?','?','?') },
> +    { V4L2_PIX_FMT_MPEG,    VLC_FOURCC('?','?','?','?') },
> +#endif
> +    { 0, 0, 0, 0, 0 }
> +};
> +
> +bool get_v4l2pixelformat_by_fourcc( unsigned int *found, vlc_fourcc_t
> i_requested_fourcc ) +{
> +    for( int i = 0; v4l2chroma_to_fourcc[i].i_v4l2 != 0; i++ )
> +    {
> +        if( v4l2chroma_to_fourcc[i].i_fourcc == i_requested_fourcc )
> +        {
> +            *found = v4l2chroma_to_fourcc[i].i_v4l2;
> +            return true;
> +        }
> +    }
> +    return false;
> +}
> +
> +bool get_fourcc_by_v4l2pixelformat( vlc_fourcc_t *i_fourcc, int *i_rmask,
> int *i_gmask, int *i_bmask, unsigned int pixelformat ) +{
> +    for( int i = 0; v4l2chroma_to_fourcc[i].i_fourcc != 0; i++ )
> +    {
> +        if( v4l2chroma_to_fourcc[i].i_v4l2 == pixelformat )
> +        {
> +            *i_fourcc = v4l2chroma_to_fourcc[i].i_fourcc;
> +            *i_rmask = v4l2chroma_to_fourcc[i].i_rmask;
> +            *i_gmask = v4l2chroma_to_fourcc[i].i_gmask;
> +            *i_bmask = v4l2chroma_to_fourcc[i].i_bmask;
> +            return true;
> +        }
> +    }
> +    return false;
> +}
> +/*************************************************************************
> **** + * Helper function to initalise video IO using the mmap method
> +
> **************************************************************************
> ***/ +int InitMmap( vlc_object_t *p_demux, demux_sys_t *p_sys, int i_fd )
> +{
> +    struct v4l2_requestbuffers req;
> +
> +    memset( &req, 0, sizeof(req) );
> +    req.count = 4;
> +    req.type = V4L2_BUF_TYPE_VIDEO_CAPTURE;
> +    req.memory = V4L2_MEMORY_MMAP;
> +
> +    if( v4l2_ioctl( i_fd, VIDIOC_REQBUFS, &req ) < 0 )
> +    {
> +        msg_Err( p_demux, "device does not support mmap I/O" );
> +        return -1;
> +    }
> +
> +    if( req.count < 2 )
> +    {
> +        msg_Err( p_demux, "insufficient buffers" );
> +        return -1;
> +    }
> +
> +    p_sys->p_buffers = calloc( req.count, sizeof( *p_sys->p_buffers ) );
> +    if( unlikely(!p_sys->p_buffers) )
> +        return -1;
> +
> +    for( p_sys->i_nbuffers = 0; p_sys->i_nbuffers < req.count;
> ++p_sys->i_nbuffers ) +    {
> +        struct v4l2_buffer buf;
> +
> +        memset( &buf, 0, sizeof(buf) );
> +        buf.type = V4L2_BUF_TYPE_VIDEO_CAPTURE;
> +        buf.memory = V4L2_MEMORY_MMAP;
> +        buf.index = p_sys->i_nbuffers;
> +
> +        if( v4l2_ioctl( i_fd, VIDIOC_QUERYBUF, &buf ) < 0 )
> +        {
> +            msg_Err( p_demux, "VIDIOC_QUERYBUF: %m" );
> +            return -1;
> +        }
> +
> +        p_sys->p_buffers[p_sys->i_nbuffers].length = buf.length;
> +        p_sys->p_buffers[p_sys->i_nbuffers].start =
> +            v4l2_mmap( NULL, buf.length, PROT_READ | PROT_WRITE,
> MAP_SHARED, i_fd, buf.m.offset ); +
> +        if( p_sys->p_buffers[p_sys->i_nbuffers].start == MAP_FAILED )
> +        {
> +            msg_Err( p_demux, "mmap failed: %m" );
> +            return -1;
> +        }
> +    }
> +
> +    return 0;
> +}
> diff --git a/modules/access/v4l2/v4l2_common.h
> b/modules/access/v4l2/v4l2_common.h new file mode 100644
> index 0000000..c56b7b3
> --- /dev/null
> +++ b/modules/access/v4l2/v4l2_common.h
> @@ -0,0 +1,31 @@
> +/*************************************************************************
> **** + * v4l2_common.h : Video4Linux2 input module for vlc
> +
> **************************************************************************
> *** + * Copyright (C) 2002-2011 the VideoLAN team
> + *
> + * This program is free software; you can redistribute it and/or modify
> + * it under the terms of the GNU General Public License as published by
> + * the Free Software Foundation; either version 2 of the License, or
> + * (at your option) any later version.
> + *
> + * This program is distributed in the hope that it will be useful,
> + * but WITHOUT ANY WARRANTY; without even the implied warranty of
> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
> + * GNU General Public License for more details.
> + *
> + * You should have received a copy of the GNU General Public License
> + * along with this program; if not, write to the Free Software
> + * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston MA
> 02110-1301, USA. +
> **************************************************************************
> ***/ +#ifndef V4L2_COMMON_H
> +#define V4L2_COMMON_H
> +
> +#include <vlc_common.h>
> +
> +int InitMmap( vlc_object_t *p_demux, demux_sys_t *p_sys, int i_fd );
> +
> +bool get_fourcc_by_v4l2pixelformat( vlc_fourcc_t *i_fourcc, int *i_rmask,
> int *i_gmask, int *i_bmask, unsigned int pixelformat ); +bool
> get_v4l2pixelformat_by_fourcc( unsigned int *found, vlc_fourcc_t
> i_requested_fourcc ); +
> +
> +#endif // V4L2_COMMON_H
> diff --git a/modules/access/v4l2/v4l2_input.h
> b/modules/access/v4l2/v4l2_input.h new file mode 100644
> index 0000000..ffb102a
> --- /dev/null
> +++ b/modules/access/v4l2/v4l2_input.h
> @@ -0,0 +1,46 @@
> +/*************************************************************************
> **** + * v4l2_input.h : Video4Linux2 input module for vlc
> +
> **************************************************************************
> *** + * Copyright (C) 2002-2011 the VideoLAN team
> + *
> + * This program is free software; you can redistribute it and/or modify
> + * it under the terms of the GNU General Public License as published by
> + * the Free Software Foundation; either version 2 of the License, or
> + * (at your option) any later version.
> + *
> + * This program is distributed in the hope that it will be useful,
> + * but WITHOUT ANY WARRANTY; without even the implied warranty of
> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
> + * GNU General Public License for more details.
> + *
> + * You should have received a copy of the GNU General Public License
> + * along with this program; if not, write to the Free Software
> + * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston MA
> 02110-1301, USA. +
> **************************************************************************
> ***/ +
> +#ifndef V4L2_INPUT_H
> +#define V4L2_INPUT_H
> +
> +#define CFG_PREFIX "v4l2-"
> +
> +/* video.c */
> +void ParseMRL(vlc_object_t *, const char *);
> +int OpenVideo(vlc_object_t *, demux_sys_t *, bool);
> +block_t* GrabVideo(vlc_object_t *, demux_sys_t *);
> +
> +/* demux.c */
> +int DemuxOpen(vlc_object_t *);
> +void DemuxClose(vlc_object_t *);
> +float GetAbsoluteMaxFrameRate(vlc_object_t *, int fd, uint32_t fmt);
> +void GetMaxDimensions(vlc_object_t *, int fd, uint32_t fmt, float fps_min,
> +                      uint32_t *pwidth, uint32_t *pheight);
> +
> +/* access.c */
> +int AccessOpen(vlc_object_t *);
> +void AccessClose(vlc_object_t *);
> +
> +/* controls.c */
> +vlc_v4l2_ctrl_t *ControlsInit(vlc_object_t *, int fd);
> +void ControlsDeinit(vlc_object_t *, vlc_v4l2_ctrl_t *);
> +
> +#endif // V4L2_INPUT_H
> diff --git a/modules/access/v4l2/video.c b/modules/access/v4l2/video.c
> index 951213d..5c33934 100644
> --- a/modules/access/v4l2/video.c
> +++ b/modules/access/v4l2/video.c
> @@ -38,6 +38,8 @@
>  #endif
> 
>  #include "v4l2.h"
> +#include "v4l2_input.h"
> +#include "v4l2_common.h"
>  #include <vlc_plugin.h>
>  #include <vlc_fs.h>
>  #include <vlc_demux.h>
> @@ -460,52 +462,6 @@ vlc_module_end ()
> 
>  static block_t* ProcessVideoFrame( vlc_object_t *p_demux, uint8_t
> *p_frame, size_t );
> 
> -static const struct
> -{
> -    unsigned int i_v4l2;
> -    vlc_fourcc_t i_fourcc;
> -    int i_rmask;
> -    int i_gmask;
> -    int i_bmask;
> -} v4l2chroma_to_fourcc[] =
> -{
> -    /* Raw data types */
> -    { V4L2_PIX_FMT_GREY,    VLC_CODEC_GREY, 0, 0, 0 },
> -    { V4L2_PIX_FMT_HI240,   VLC_FOURCC('I','2','4','0'), 0, 0, 0 },
> -    { V4L2_PIX_FMT_RGB555,  VLC_CODEC_RGB15, 0x001f,0x03e0,0x7c00 },
> -    { V4L2_PIX_FMT_RGB565,  VLC_CODEC_RGB16, 0x001f,0x07e0,0xf800 },
> -    /* Won't work since we don't know how to handle such gmask values
> -     * correctly
> -    { V4L2_PIX_FMT_RGB555X, VLC_CODEC_RGB15, 0x007c,0xe003,0x1f00 },
> -    { V4L2_PIX_FMT_RGB565X, VLC_CODEC_RGB16, 0x00f8,0xe007,0x1f00 },
> -    */
> -    { V4L2_PIX_FMT_BGR24,   VLC_CODEC_RGB24, 0xff0000,0xff00,0xff },
> -    { V4L2_PIX_FMT_RGB24,   VLC_CODEC_RGB24, 0xff,0xff00,0xff0000 },
> -    { V4L2_PIX_FMT_BGR32,   VLC_CODEC_RGB32, 0xff0000,0xff00,0xff },
> -    { V4L2_PIX_FMT_RGB32,   VLC_CODEC_RGB32, 0xff,0xff00,0xff0000 },
> -    { V4L2_PIX_FMT_YUYV,    VLC_CODEC_YUYV, 0, 0, 0 },
> -    { V4L2_PIX_FMT_UYVY,    VLC_CODEC_UYVY, 0, 0, 0 },
> -    { V4L2_PIX_FMT_Y41P,    VLC_FOURCC('I','4','1','N'), 0, 0, 0 },
> -    { V4L2_PIX_FMT_YUV422P, VLC_CODEC_I422, 0, 0, 0 },
> -    { V4L2_PIX_FMT_YVU420,  VLC_CODEC_YV12, 0, 0, 0 },
> -    { V4L2_PIX_FMT_YUV411P, VLC_CODEC_I411, 0, 0, 0 },
> -    { V4L2_PIX_FMT_YUV410,  VLC_CODEC_I410, 0, 0, 0 },
> -
> -    /* Raw data types, not in V4L2 spec but still in videodev2.h and
> supported -     * by VLC */
> -    { V4L2_PIX_FMT_YUV420,  VLC_CODEC_I420, 0, 0, 0 },
> -    /* FIXME { V4L2_PIX_FMT_RGB444,  VLC_CODEC_RGB32 }, */
> -
> -    /* Compressed data types */
> -    { V4L2_PIX_FMT_MJPEG,   VLC_CODEC_MJPG, 0, 0, 0 },
> -    { V4L2_PIX_FMT_JPEG,    VLC_CODEC_JPEG, 0, 0, 0 },
> -#if 0
> -    { V4L2_PIX_FMT_DV,      VLC_FOURCC('?','?','?','?') },
> -    { V4L2_PIX_FMT_MPEG,    VLC_FOURCC('?','?','?','?') },
> -#endif
> -    { 0, 0, 0, 0, 0 }
> -};
> -
>  /**
>   * List of V4L2 chromas were confident enough to use as fallbacks if the
>   * user hasn't provided a --v4l2-chroma value.
> @@ -673,63 +629,6 @@ static block_t* ProcessVideoFrame( vlc_object_t
> *p_demux, uint8_t *p_frame, size }
> 
>  /*************************************************************************
> **** - * Helper function to initalise video IO using the mmap method
> -
> **************************************************************************
> ***/ -static int InitMmap( vlc_object_t *p_demux, demux_sys_t *p_sys, int
> i_fd ) -{
> -    struct v4l2_requestbuffers req;
> -
> -    memset( &req, 0, sizeof(req) );
> -    req.count = 4;
> -    req.type = V4L2_BUF_TYPE_VIDEO_CAPTURE;
> -    req.memory = V4L2_MEMORY_MMAP;
> -
> -    if( v4l2_ioctl( i_fd, VIDIOC_REQBUFS, &req ) < 0 )
> -    {
> -        msg_Err( p_demux, "device does not support mmap I/O" );
> -        return -1;
> -    }
> -
> -    if( req.count < 2 )
> -    {
> -        msg_Err( p_demux, "insufficient buffers" );
> -        return -1;
> -    }
> -
> -    p_sys->p_buffers = calloc( req.count, sizeof( *p_sys->p_buffers ) );
> -    if( unlikely(!p_sys->p_buffers) )
> -        return -1;
> -
> -    for( p_sys->i_nbuffers = 0; p_sys->i_nbuffers < req.count;
> ++p_sys->i_nbuffers ) -    {
> -        struct v4l2_buffer buf;
> -
> -        memset( &buf, 0, sizeof(buf) );
> -        buf.type = V4L2_BUF_TYPE_VIDEO_CAPTURE;
> -        buf.memory = V4L2_MEMORY_MMAP;
> -        buf.index = p_sys->i_nbuffers;
> -
> -        if( v4l2_ioctl( i_fd, VIDIOC_QUERYBUF, &buf ) < 0 )
> -        {
> -            msg_Err( p_demux, "VIDIOC_QUERYBUF: %m" );
> -            return -1;
> -        }
> -
> -        p_sys->p_buffers[p_sys->i_nbuffers].length = buf.length;
> -        p_sys->p_buffers[p_sys->i_nbuffers].start =
> -            v4l2_mmap( NULL, buf.length, PROT_READ | PROT_WRITE,
> MAP_SHARED, i_fd, buf.m.offset ); -
> -        if( p_sys->p_buffers[p_sys->i_nbuffers].start == MAP_FAILED )
> -        {
> -            msg_Err( p_demux, "mmap failed: %m" );
> -            return -1;
> -        }
> -    }
> -
> -    return 0;
> -}
> -
> -/*************************************************************************
> **** * Helper function to initalise video IO using the userbuf method
> **************************************************************************
> ***/ static int InitUserP( vlc_object_t *p_demux, demux_sys_t *p_sys, int
> i_fd, unsigned int i_buffer_size ) @@ -1071,21 +970,17 @@ static int
> InitVideo( vlc_object_t *p_obj, int i_fd, demux_sys_t *p_sys,
> vlc_fourcc_to_char( codecs[i].pixelformat, fourcc_v4l2 );
> 
>              bool b_codec_supported = false;
> -            for( unsigned j = 0; v4l2chroma_to_fourcc[j].i_v4l2 != 0; j++
> ) -            {
> -                if( v4l2chroma_to_fourcc[j].i_v4l2 ==
> codecs[i].pixelformat ) -                {
> -                    char fourcc[5];
> -                    memset( fourcc, 0, sizeof( fourcc ) );
> -                    vlc_fourcc_to_char( v4l2chroma_to_fourcc[j].i_fourcc,
> -                                        fourcc );
> -                    msg_Dbg( p_obj, "device supports chroma %4.4s [%s,
> %s]", -                             fourcc, codecs[i].description,
> fourcc_v4l2 ); -                    b_codec_supported = true;
> -                }
> -            }
> -            if( !b_codec_supported )
> +            vlc_fourcc_t i_fourcc;
> +            int i_rmask, i_gmask, i_bmask;
> +            b_codec_supported = get_fourcc_by_v4l2pixelformat( &i_fourcc,
> &i_rmask, &i_gmask, &i_bmask, codecs[i].pixelformat ); +            if(
> b_codec_supported )
>              {
> +                char fourcc[5];
> +                memset( fourcc, 0, sizeof( fourcc ) );
> +                vlc_fourcc_to_char( i_fourcc, fourcc );
> +                msg_Dbg( p_obj, "device supports chroma %4.4s [%s, %s]",
> +                         fourcc, codecs[i].description, fourcc_v4l2 );
> +            } else {
>                  msg_Dbg( p_obj, "device codec %4.4s (%s) not supported",
>                           fourcc_v4l2, codecs[i].description );
>              }
> @@ -1167,14 +1062,8 @@ static int InitVideo( vlc_object_t *p_obj, int i_fd,
> demux_sys_t *p_sys, const vlc_fourcc_t i_requested_fourcc =
>                  vlc_fourcc_GetCodecFromString( VIDEO_ES, reqchroma );
> 
> -            for( int i = 0; v4l2chroma_to_fourcc[i].i_v4l2 != 0; i++ )
> -            {
> -                if( v4l2chroma_to_fourcc[i].i_fourcc == i_requested_fourcc
> ) -                {
> -                    fmt.fmt.pix.pixelformat =
> v4l2chroma_to_fourcc[i].i_v4l2; -                    break;
> -                }
> -            }
> +            get_v4l2pixelformat_by_fourcc( &fmt.fmt.pix.pixelformat,
> i_requested_fourcc ); +
>              /* Try and set user chroma */
>              bool b_error = !IsPixelFormatSupported( codecs, ncodec,
>                                                     
> fmt.fmt.pix.pixelformat ); @@ -1309,18 +1198,19 @@ static int InitVideo(
> vlc_object_t *p_obj, int i_fd, demux_sys_t *p_sys, }
> 
>      /* Look up final fourcc */
> -    p_sys->i_fourcc = 0;
> -    for( int i = 0; v4l2chroma_to_fourcc[i].i_fourcc != 0; i++ )
> +    int i_rmask, i_gmask, i_bmask;
> +    if ( get_fourcc_by_v4l2pixelformat( &p_sys->i_fourcc,
> +                                        &i_rmask,
> +                                        &i_gmask,
> +                                        &i_bmask,
> +                                        fmt.fmt.pix.pixelformat ) )
>      {
> -        if( v4l2chroma_to_fourcc[i].i_v4l2 == fmt.fmt.pix.pixelformat )
> -        {
> -            p_sys->i_fourcc = v4l2chroma_to_fourcc[i].i_fourcc;
> -            es_format_Init( &es_fmt, VIDEO_ES, p_sys->i_fourcc );
> -            es_fmt.video.i_rmask = v4l2chroma_to_fourcc[i].i_rmask;
> -            es_fmt.video.i_gmask = v4l2chroma_to_fourcc[i].i_gmask;
> -            es_fmt.video.i_bmask = v4l2chroma_to_fourcc[i].i_bmask;
> -            break;
> -        }
> +        es_format_Init( &es_fmt, VIDEO_ES, p_sys->i_fourcc );
> +        es_fmt.video.i_rmask = i_rmask;
> +        es_fmt.video.i_gmask = i_gmask;
> +        es_fmt.video.i_bmask = i_bmask;
> +    } else {
> +        p_sys->i_fourcc = 0;
>      }
> 
>      /* Buggy driver paranoia */
-- 
R?mi Denis-Courmont
http://www.remlab.info


From remi at remlab.net  Fri Dec 23 20:52:34 2011
From: remi at remlab.net (=?iso-8859-1?q?R=E9mi?= Denis-Courmont)
Date: Fri, 23 Dec 2011 20:52:34 +0100
Subject: [vlc-devel] [PATCH] Break out early if a filter returns NULL,
	so other filters don't fail
In-Reply-To: <4EF25917.60904@ymail.com>
References: <4EF25917.60904@ymail.com>
Message-ID: <201112232052.34987.remi@remlab.net>

   Hello,

On Wednesday 21 December 2011, Cheng Sun wrote:
> Test case: turn on all filters. Play some audio. Speed it up. Watch it
> crash.

Fixed thanks.


-- 
R?mi Denis-Courmont
http://www.remlab.info


From remi at remlab.net  Fri Dec 23 20:56:57 2011
From: remi at remlab.net (=?iso-8859-1?q?R=E9mi?= Denis-Courmont)
Date: Fri, 23 Dec 2011 20:56:57 +0100
Subject: [vlc-devel] [PATCH] Audio channel remapping filter
In-Reply-To: <4EF2483D.2070808@ymail.com>
References: <4EF1BC96.6080800@ymail.com> <201112211947.29144.remi@remlab.net>
	<4EF2483D.2070808@ymail.com>
Message-ID: <201112232056.57830.remi@remlab.net>

   Hello,

On Wednesday 21 December 2011, Cheng Sun wrote:
> On 21/12/11 17:47, R?mi Denis-Courmont wrote:
> > WTF?
> 
> Nicely summed up :)

This is useless since you already check that format are identical:

+        ( audio_in->i_bitspersample != audio_out->i_bitspersample) ||

This is useless since you have a list of supported formats in GetRemapFun(). 
However, you should move GetRemapFun() earlier in probing.

+        ( audio_in->i_bitspersample % 8 != 0 ) )

+    audio_out->i_rate = audio_in->i_rate;

You are not supposed to change the rate, unless you're implementing a 
resampler.

+    audio_out->i_format = p_filter->fmt_out.i_codec;

That is also useless.

-- 
R?mi Denis-Courmont
http://www.remlab.info


From remi at remlab.net  Fri Dec 23 21:00:56 2011
From: remi at remlab.net (=?iso-8859-1?q?R=E9mi?= Denis-Courmont)
Date: Fri, 23 Dec 2011 21:00:56 +0100
Subject: [vlc-devel] [Patch] Adding an helper to convert iso8601
	durations to seconds
In-Reply-To: <CAK-C75RayZc-hHcd9-BE-cF5gXFF_JjNB6hDGCz5wxS4OfSfAg@mail.gmail.com>
References: <CAK-C75R5BPUud6z2QxrFNNnh1wMFktu48O59oXi2bAUqZ0mYMw@mail.gmail.com>
	<201112141905.47669.remi@remlab.net>
	<CAK-C75RayZc-hHcd9-BE-cF5gXFF_JjNB6hDGCz5wxS4OfSfAg@mail.gmail.com>
Message-ID: <201112232100.56625.remi@remlab.net>

   Hello,

On Wednesday 14 December 2011, Hugo Beauz?e-Luyssen wrote:
> 2011/12/14 R?mi Denis-Courmont <remi at remlab.net>:
> > Le mercredi 14 d?cembre 2011 00:48:41 Hugo Beauz?e-Luyssen, vous avez 
?crit :
> >> On Tue, Dec 13, 2011 at 11:46 PM, Hugo Beauz?e-Luyssen
> >> 
> >> <beauze.h at gmail.com> wrote:
> >> > 2011/12/9 R?mi Denis-Courmont <remi at remlab.net>:
> >> >> Function name is too long. And time_t is thecorrect type for seconds
> >> >> IMHO.
> >> 
> >> Oops, this one's the good one.
> > 
> > I think this patch does not work due to typo. Please drop _to_s*. It's
> > obviously seconds since the return type is time_t.
> 
> Hmmm yep you're right... my bad.
> Here are some fixed versions of both patches.

I hate defensive programming in internal functions. It encourages sloppy code.

While I don't know the ISO standard, I find it very suspicious to parse floating 
point values according to the system locale.

And I don't see the use for this helper, at least in that patch series...

-- 
R?mi Denis-Courmont
http://www.remlab.info


From typx at dinauz.org  Fri Dec 23 21:02:06 2011
From: typx at dinauz.org (Denis Charmet)
Date: Fri, 23 Dec 2011 21:02:06 +0100
Subject: [vlc-devel] [PATCH] Avoid a crash with some corrupted mkvs
In-Reply-To: <201112232045.08964.remi@remlab.net>
References: <1324669279-29319-1-git-send-email-typx@dinauz.org>
	<201112232045.08964.remi@remlab.net>
Message-ID: <20111223200206.GA30961@dinauz.org>

Le vendredi 23 d?cembre 2011 ? 08:45:08, R?mi Denis-Courmont a ?crit :
>    Hello,
> 
> On Friday 23 December 2011, Denis Charmet wrote:
> This is not very kosher if i_extra_data < sizeof(BITMAPINFOHEADER).

I totally agree but well... a bug was filed... 

-- 
TypX
Le mauvais esprit est un art de vivre


From remi at remlab.net  Fri Dec 23 21:02:13 2011
From: remi at remlab.net (=?iso-8859-1?q?R=E9mi?= Denis-Courmont)
Date: Fri, 23 Dec 2011 21:02:13 +0100
Subject: [vlc-devel] [PATCH 1/1] messages.c: added logging via
	OutputDebugString under windows.
In-Reply-To: <1324632311-2620-1-git-send-email-RSATom@gmail.com>
References: <1324632311-2620-1-git-send-email-RSATom@gmail.com>
Message-ID: <201112232102.13744.remi@remlab.net>

   Hello,

Merged, thanks,

-- 
R?mi Denis-Courmont
http://www.remlab.info


From jb at videolan.org  Fri Dec 23 21:09:31 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Fri, 23 Dec 2011 21:09:31 +0100
Subject: [vlc-devel] [PATCH] Avoid a crash with some corrupted mkvs
In-Reply-To: <20111223200206.GA30961@dinauz.org>
References: <1324669279-29319-1-git-send-email-typx@dinauz.org>
	<201112232045.08964.remi@remlab.net>
	<20111223200206.GA30961@dinauz.org>
Message-ID: <20111223200931.GA9535@videolan.org>

On Fri, Dec 23, 2011 at 09:02:06PM +0100, Denis Charmet wrote :
> Le vendredi 23 d?cembre 2011 ? 08:45:08, R?mi Denis-Courmont a ?crit :
> >    Hello,
> > 
> > On Friday 23 December 2011, Denis Charmet wrote:
> > This is not very kosher if i_extra_data < sizeof(BITMAPINFOHEADER).
> 
> I totally agree but well... a bug was filed... 
> 

What I understood from R?mi's mail is that you should test i_extra_data
< sizeof(BITMAPINFOHEADER) before doing the substraction.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From typx at dinauz.org  Fri Dec 23 21:12:59 2011
From: typx at dinauz.org (Denis Charmet)
Date: Fri, 23 Dec 2011 21:12:59 +0100
Subject: [vlc-devel] [PATCH] Avoid a crash with some corrupted mkvs
In-Reply-To: <20111223200931.GA9535@videolan.org>
References: <1324669279-29319-1-git-send-email-typx@dinauz.org>
	<201112232045.08964.remi@remlab.net>
	<20111223200206.GA30961@dinauz.org>
	<20111223200931.GA9535@videolan.org>
Message-ID: <20111223201259.GA32346@dinauz.org>

Le vendredi 23 d?cembre 2011 ? 09:09:31, Jean-Baptiste Kempf a ?crit :
> On Fri, Dec 23, 2011 at 09:02:06PM +0100, Denis Charmet wrote :
> > Le vendredi 23 d?cembre 2011 ? 08:45:08, R?mi Denis-Courmont a ?crit :
> > >    Hello,
> > > 
> > > On Friday 23 December 2011, Denis Charmet wrote:
> > > This is not very kosher if i_extra_data < sizeof(BITMAPINFOHEADER).
> > 
> > I totally agree but well... a bug was filed... 
> > 
> 
> What I understood from R?mi's mail is that you should test i_extra_data
> < sizeof(BITMAPINFOHEADER) before doing the substraction.
> 
Oh... my bad I didn't understand as the test is done before the patch
(line 917)

-- 
TypX
Le mauvais esprit est un art de vivre


From jb at videolan.org  Fri Dec 23 21:14:58 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Fri, 23 Dec 2011 21:14:58 +0100
Subject: [vlc-devel] [PATCH] Avoid a crash with some corrupted mkvs
In-Reply-To: <20111223201259.GA32346@dinauz.org>
References: <1324669279-29319-1-git-send-email-typx@dinauz.org>
	<201112232045.08964.remi@remlab.net>
	<20111223200206.GA30961@dinauz.org>
	<20111223200931.GA9535@videolan.org>
	<20111223201259.GA32346@dinauz.org>
Message-ID: <20111223201458.GA20183@videolan.org>

On Fri, Dec 23, 2011 at 09:12:59PM +0100, Denis Charmet wrote :
> Le vendredi 23 d?cembre 2011 ? 09:09:31, Jean-Baptiste Kempf a ?crit :
> > On Fri, Dec 23, 2011 at 09:02:06PM +0100, Denis Charmet wrote :
> > > Le vendredi 23 d?cembre 2011 ? 08:45:08, R?mi Denis-Courmont a ?crit :
> > > >    Hello,
> > > > 
> > > > On Friday 23 December 2011, Denis Charmet wrote:
> > > > This is not very kosher if i_extra_data < sizeof(BITMAPINFOHEADER).
> > > 
> > > I totally agree but well... a bug was filed... 
> > > 
> > 
> > What I understood from R?mi's mail is that you should test i_extra_data
> > < sizeof(BITMAPINFOHEADER) before doing the substraction.
> > 
> Oh... my bad I didn't understand as the test is done before the patch
> (line 917)

Maybe, also, I understood WORONG.


-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From remi at remlab.net  Fri Dec 23 21:28:37 2011
From: remi at remlab.net (=?iso-8859-1?q?R=E9mi?= Denis-Courmont)
Date: Fri, 23 Dec 2011 21:28:37 +0100
Subject: [vlc-devel] [vlc-commits] dhsow: missing white space (breaks PO
	files)
In-Reply-To: <20111223201743.42F2E145A7A@albiero.videolan.org>
References: <20111223201743.42F2E145A7A@albiero.videolan.org>
Message-ID: <201112232128.38303.remi@remlab.net>

   Hello,

On Friday 23 December 2011, R?mi Denis-Courmont wrote:
> vlc/vlc-1.2 | branch: master | R?mi Denis-Courmont <remi at remlab.net> | Fri
> Dec 23 09:05:37 2011 +0100| [78931a8390510d0eed941d02e823298968fd05d6] |
> committer: Jean-Baptiste Kempf
> 
> dhsow: missing white space (breaks PO files)
> (cherry picked from commit fbfc5966f10c2763576f6743269efe1ff3179673)
> 
> Signed-off-by: Jean-Baptiste Kempf <jb at videolan.org>

And you're going to fix the PO files ?

-- 
R?mi Denis-Courmont
http://www.remlab.info


From fcvlcdev at free.fr  Fri Dec 23 23:53:19 2011
From: fcvlcdev at free.fr (Francois Cartegnie)
Date: Fri, 23 Dec 2011 23:53:19 +0100
Subject: [vlc-devel] [PATCH] v4l2: split declarations
In-Reply-To: <201112232047.52152.remi@remlab.net>
References: <1324664664-11765-1-git-send-email-fcvlcdev@free.fr>
	<1324664664-11765-2-git-send-email-fcvlcdev@free.fr>
	<201112232047.52152.remi@remlab.net>
Message-ID: <4EF5065F.5050601@free.fr>

Le 23/12/2011 20:47, R?mi Denis-Courmont a ?crit :
>    Hello,
> 
> I don't see the point in factoring demux-specific mmap stuff. It can't work. And 
> it will make my work on zero-copy V4L2 input needlessly more complicated. It's 
> already bad enough with the V4L2 access that nobody seems to use.

Buffer allocation is exactly the same for using an output driver.
We just need to change the initmmap() to pass the buffer type.

Francois


From rsatom at gmail.com  Sat Dec 24 02:15:59 2011
From: rsatom at gmail.com (Sergey Radionov)
Date: Sat, 24 Dec 2011 08:15:59 +0700
Subject: [vlc-devel] [PATCH 1/7] npapi: added members for notification child
	classes about creation/destroying of media_player_t
Message-ID: <1324689365-2372-1-git-send-email-RSATom@gmail.com>

---
 npapi/vlcplugin_base.cpp |    2 ++
 npapi/vlcplugin_base.h   |    5 +++++
 2 files changed, 7 insertions(+), 0 deletions(-)

diff --git a/npapi/vlcplugin_base.cpp b/npapi/vlcplugin_base.cpp
index d1d9bc6..bd6db07 100644
--- a/npapi/vlcplugin_base.cpp
+++ b/npapi/vlcplugin_base.cpp
@@ -650,6 +650,7 @@ bool VlcPluginBase::playlist_select( int idx )
         if( playlist_isplaying() )
             playlist_stop();
         events.unhook_manager( this );
+        on_media_player_release();
         libvlc_media_player_release( libvlc_media_player );
         libvlc_media_player = NULL;
     }
@@ -657,6 +658,7 @@ bool VlcPluginBase::playlist_select( int idx )
     libvlc_media_player = libvlc_media_player_new_from_media( p_m );
     if( libvlc_media_player )
     {
+        on_media_player_new();
         set_player_window();
 
         libvlc_event_manager_t *p_em;
diff --git a/npapi/vlcplugin_base.h b/npapi/vlcplugin_base.h
index f731976..c9c8a59 100644
--- a/npapi/vlcplugin_base.h
+++ b/npapi/vlcplugin_base.h
@@ -293,6 +293,11 @@ public:
     void event_callback(const libvlc_event_t *, NPVariant *, uint32_t);
 
 protected:
+    // called after libvlc_media_player_new_from_media
+    virtual void on_media_player_new()     {};
+    // called before libvlc_media_player_release
+    virtual void on_media_player_release() {};
+
     bool playlist_select(int);
     virtual void set_player_window() = 0;
 
-- 
1.7.7.1.msysgit.0



From rsatom at gmail.com  Sat Dec 24 02:17:09 2011
From: rsatom at gmail.com (Sergey Radionov)
Date: Sat, 24 Dec 2011 08:17:09 +0700
Subject: [vlc-devel] [PATCH 2/7] npapi win32: added handling of destroying
	of media_player_t
Message-ID: <1324689434-3552-2-git-send-email-RSATom@gmail.com>

---
 npapi/vlcplugin_win.cpp |    4 ++++
 npapi/vlcplugin_win.h   |    3 +++
 2 files changed, 7 insertions(+), 0 deletions(-)

diff --git a/npapi/vlcplugin_win.cpp b/npapi/vlcplugin_win.cpp
index 8e54c5d..260c77e 100644
--- a/npapi/vlcplugin_win.cpp
+++ b/npapi/vlcplugin_win.cpp
@@ -177,3 +177,7 @@ bool VlcPluginWin::destroy_windows()
 
     return true;
 }
+void VlcPluginWin::on_media_player_release()
+{
+    _WindowsManager.LibVlcDetach();
+}
diff --git a/npapi/vlcplugin_win.h b/npapi/vlcplugin_win.h
index e5ede7b..249c9f4 100644
--- a/npapi/vlcplugin_win.h
+++ b/npapi/vlcplugin_win.h
@@ -62,6 +62,9 @@ public:
     void                setWindowProc(WNDPROC wndproc)
                             { pf_wndproc = wndproc; };
 
+protected:
+    virtual void on_media_player_release();
+
 private:
     void set_player_window();
 
-- 
1.7.7.1.msysgit.0



From rsatom at gmail.com  Sat Dec 24 02:17:10 2011
From: rsatom at gmail.com (Sergey Radionov)
Date: Sat, 24 Dec 2011 08:17:10 +0700
Subject: [vlc-devel] [PATCH 3/7] npapi win32: move attaching to libvlc from
	set_player_window to on_media_player_new
In-Reply-To: <1324689434-3552-2-git-send-email-RSATom@gmail.com>
References: <1324689434-3552-2-git-send-email-RSATom@gmail.com>
Message-ID: <1324689434-3552-3-git-send-email-RSATom@gmail.com>

---
 npapi/vlcplugin_win.cpp |   11 ++++++-----
 npapi/vlcplugin_win.h   |    3 ++-
 2 files changed, 8 insertions(+), 6 deletions(-)

diff --git a/npapi/vlcplugin_win.cpp b/npapi/vlcplugin_win.cpp
index 260c77e..8ff89cf 100644
--- a/npapi/vlcplugin_win.cpp
+++ b/npapi/vlcplugin_win.cpp
@@ -86,11 +86,6 @@ VlcPluginWin::~VlcPluginWin()
     destroy_windows();
 }
 
-void VlcPluginWin::set_player_window()
-{
-    _WindowsManager.LibVlcAttach(libvlc_media_player);
-}
-
 void VlcPluginWin::toggle_fullscreen()
 {
     _WindowsManager.ToggleFullScreen();
@@ -177,6 +172,12 @@ bool VlcPluginWin::destroy_windows()
 
     return true;
 }
+
+void VlcPluginWin::on_media_player_new()
+{
+    _WindowsManager.LibVlcAttach(libvlc_media_player);
+}
+
 void VlcPluginWin::on_media_player_release()
 {
     _WindowsManager.LibVlcDetach();
diff --git a/npapi/vlcplugin_win.h b/npapi/vlcplugin_win.h
index 249c9f4..5c1206d 100644
--- a/npapi/vlcplugin_win.h
+++ b/npapi/vlcplugin_win.h
@@ -63,10 +63,11 @@ public:
                             { pf_wndproc = wndproc; };
 
 protected:
+    virtual void on_media_player_new();
     virtual void on_media_player_release();
 
 private:
-    void set_player_window();
+    void set_player_window(){};
 
     unsigned int     i_width, i_height;
     unsigned int     i_tb_width, i_tb_height;
-- 
1.7.7.1.msysgit.0



From rsatom at gmail.com  Sat Dec 24 02:17:11 2011
From: rsatom at gmail.com (Sergey Radionov)
Date: Sat, 24 Dec 2011 08:17:11 +0700
Subject: [vlc-devel] [PATCH 4/7] ActiveX: added Windows Manager detaching
	from libvlc on libvlc_media_player_release
In-Reply-To: <1324689434-3552-2-git-send-email-RSATom@gmail.com>
References: <1324689434-3552-2-git-send-email-RSATom@gmail.com>
Message-ID: <1324689434-3552-4-git-send-email-RSATom@gmail.com>

---
 activex/plugin.cpp |    1 +
 1 files changed, 1 insertions(+), 0 deletions(-)

diff --git a/activex/plugin.cpp b/activex/plugin.cpp
index e4450d5..9c70ba2 100644
--- a/activex/plugin.cpp
+++ b/activex/plugin.cpp
@@ -1245,6 +1245,7 @@ bool VLCPlugin::playlist_select( int idx )
         if( isPlaying() )
             playlist_stop();
         player_unregister_events();
+        _WindowsManager.LibVlcDetach();
         libvlc_media_player_release( _p_mplayer );
         _p_mplayer = NULL;
     }
-- 
1.7.7.1.msysgit.0



From rsatom at gmail.com  Sat Dec 24 02:17:12 2011
From: rsatom at gmail.com (Sergey Radionov)
Date: Sat, 24 Dec 2011 08:17:12 +0700
Subject: [vlc-devel] [PATCH 5/7] npapi win32: move holder window resizing to
	resize_window(); 
In-Reply-To: <1324689434-3552-2-git-send-email-RSATom@gmail.com>
References: <1324689434-3552-2-git-send-email-RSATom@gmail.com>
Message-ID: <1324689434-3552-5-git-send-email-RSATom@gmail.com>

---
 npapi/vlcplugin_win.cpp |   20 ++++++--------------
 1 files changed, 6 insertions(+), 14 deletions(-)

diff --git a/npapi/vlcplugin_win.cpp b/npapi/vlcplugin_win.cpp
index 8ff89cf..2b82953 100644
--- a/npapi/vlcplugin_win.cpp
+++ b/npapi/vlcplugin_win.cpp
@@ -53,16 +53,6 @@ static LRESULT CALLBACK Manage( HWND p_hwnd, UINT i_msg, WPARAM wpar, LPARAM lpa
             EndPaint( p_hwnd, &paintstruct );
             return 0L;
         }
-        case WM_SIZE:{
-            int new_client_width = LOWORD(lpar);
-            int new_client_height = HIWORD(lpar);
-            //first child will be resized to client area
-            HWND hChildWnd = GetWindow(p_hwnd, GW_CHILD);
-            if(hChildWnd){
-                MoveWindow(hChildWnd, 0, 0, new_client_width, new_client_height, FALSE);
-            }
-            return 0L;
-        }
         case WM_LBUTTONDBLCLK:{
             p_plugin->toggle_fullscreen();
             return 0L;
@@ -145,11 +135,13 @@ bool VlcPluginWin::create_windows()
 
 bool VlcPluginWin::resize_windows()
 {
-    /* TODO */
     HWND drawable = (HWND) (getWindow().window);
-    /* Redraw window */
-    InvalidateRect( drawable, NULL, TRUE );
-    UpdateWindow( drawable );
+    RECT rect;
+    GetClientRect(drawable, &rect);
+    if(!_WindowsManager.IsFullScreen() && _WindowsManager.getHolderWnd()){
+        HWND hHolderWnd = _WindowsManager.getHolderWnd()->getHWND();
+        MoveWindow(hHolderWnd, 0, 0, rect.right - rect.left, rect.bottom - rect.top, TRUE);
+    }
     return true;
 }
 
-- 
1.7.7.1.msysgit.0



From rsatom at gmail.com  Sat Dec 24 02:17:13 2011
From: rsatom at gmail.com (Sergey Radionov)
Date: Sat, 24 Dec 2011 08:17:13 +0700
Subject: [vlc-devel] [PATCH 6/7] npapi win32: removed replacing mozilla
	window's window proc.
In-Reply-To: <1324689434-3552-2-git-send-email-RSATom@gmail.com>
References: <1324689434-3552-2-git-send-email-RSATom@gmail.com>
Message-ID: <1324689434-3552-6-git-send-email-RSATom@gmail.com>

since:
a) any painting on it is never visible, because it is under holder window.
b) mozilla window can't recive mouse input, because all mouse input catched by holder window.
---
 npapi/vlcplugin_win.cpp |   59 -----------------------------------------------
 npapi/vlcplugin_win.h   |    6 ----
 2 files changed, 0 insertions(+), 65 deletions(-)

diff --git a/npapi/vlcplugin_win.cpp b/npapi/vlcplugin_win.cpp
index 2b82953..a8f09f9 100644
--- a/npapi/vlcplugin_win.cpp
+++ b/npapi/vlcplugin_win.cpp
@@ -24,49 +24,8 @@ BOOL WINAPI DllMain(HANDLE hModule, DWORD fdwReason, LPVOID lpReserved )
 };
 
 
-static LRESULT CALLBACK Manage( HWND p_hwnd, UINT i_msg, WPARAM wpar, LPARAM lpar )
-{
-    VlcPluginWin *p_plugin = reinterpret_cast<VlcPluginWin *>(GetWindowLongPtr(p_hwnd, GWLP_USERDATA));
-
-    switch( i_msg )
-    {
-        case WM_ERASEBKGND:
-            return 1L;
-
-        case WM_PAINT:
-        {
-            PAINTSTRUCT paintstruct;
-            HDC hdc;
-            RECT rect;
-
-            hdc = BeginPaint( p_hwnd, &paintstruct );
-
-            GetClientRect( p_hwnd, &rect );
-
-            FillRect( hdc, &rect, (HBRUSH)GetStockObject(BLACK_BRUSH) );
-            SetTextColor(hdc, RGB(255, 255, 255));
-            SetBkColor(hdc, RGB(0, 0, 0));
-            if( p_plugin->psz_text )
-                DrawText( hdc, p_plugin->psz_text, strlen(p_plugin->psz_text), &rect,
-                          DT_CENTER|DT_VCENTER|DT_SINGLELINE);
-
-            EndPaint( p_hwnd, &paintstruct );
-            return 0L;
-        }
-        case WM_LBUTTONDBLCLK:{
-            p_plugin->toggle_fullscreen();
-            return 0L;
-        }
-        default:
-            /* delegate to default handler */
-            return CallWindowProc( p_plugin->getWindowProc(), p_hwnd,
-                                   i_msg, wpar, lpar );
-    }
-}
-
 VlcPluginWin::VlcPluginWin(NPP instance, NPuint16_t mode) :
     VlcPluginBase(instance, mode),
-    pf_wndproc(NULL),
     _WindowsManager(DllGetModule())
 {
 }
@@ -119,10 +78,6 @@ bool VlcPluginWin::create_windows()
     SetWindowLongPtr(drawable, GWLP_USERDATA,
                      (LONG_PTR)this);
 
-    /* install our WNDPROC */
-    setWindowProc( (WNDPROC)SetWindowLongPtr( drawable,
-                                     GWLP_WNDPROC, (LONG_PTR)Manage ) );
-
     /* change window style to our liking */
     LONG style = GetWindowLong(drawable, GWL_STYLE);
     style |= WS_CLIPCHILDREN | WS_CLIPSIBLINGS;
@@ -148,20 +103,6 @@ bool VlcPluginWin::resize_windows()
 bool VlcPluginWin::destroy_windows()
 {
     _WindowsManager.DestroyWindows();
-
-    HWND oldwin = (HWND)npwindow.window;
-    if( oldwin )
-    {
-        WNDPROC winproc = getWindowProc();
-        if( winproc )
-        {
-            /* reset WNDPROC */
-            SetWindowLongPtr( oldwin, GWLP_WNDPROC, (LONG_PTR)winproc );
-            setWindowProc(NULL);
-        }
-        npwindow.window = NULL;
-    }
-
     return true;
 }
 
diff --git a/npapi/vlcplugin_win.h b/npapi/vlcplugin_win.h
index 5c1206d..255219f 100644
--- a/npapi/vlcplugin_win.h
+++ b/npapi/vlcplugin_win.h
@@ -57,11 +57,6 @@ public:
     void update_controls();
     void popup_menu(){};
 
-    WNDPROC             getWindowProc()
-                            { return pf_wndproc; };
-    void                setWindowProc(WNDPROC wndproc)
-                            { pf_wndproc = wndproc; };
-
 protected:
     virtual void on_media_player_new();
     virtual void on_media_player_release();
@@ -72,7 +67,6 @@ private:
     unsigned int     i_width, i_height;
     unsigned int     i_tb_width, i_tb_height;
 
-    WNDPROC pf_wndproc;
     VLCWindowsManager _WindowsManager;
 
     int i_last_position;
-- 
1.7.7.1.msysgit.0



From rsatom at gmail.com  Sat Dec 24 02:17:14 2011
From: rsatom at gmail.com (Sergey Radionov)
Date: Sat, 24 Dec 2011 08:17:14 +0700
Subject: [vlc-devel] [PATCH 7/7] npapi win32: unused members were removed
In-Reply-To: <1324689434-3552-2-git-send-email-RSATom@gmail.com>
References: <1324689434-3552-2-git-send-email-RSATom@gmail.com>
Message-ID: <1324689434-3552-7-git-send-email-RSATom@gmail.com>

---
 npapi/vlcplugin_win.h |    7 -------
 1 files changed, 0 insertions(+), 7 deletions(-)

diff --git a/npapi/vlcplugin_win.h b/npapi/vlcplugin_win.h
index 255219f..673997d 100644
--- a/npapi/vlcplugin_win.h
+++ b/npapi/vlcplugin_win.h
@@ -41,8 +41,6 @@ public:
     VlcPluginWin(NPP, NPuint16_t);
     virtual ~VlcPluginWin();
 
-    int                 setSize(unsigned width, unsigned height);
-
     void toggle_fullscreen();
     void set_fullscreen( int );
     int  get_fullscreen();
@@ -64,12 +62,7 @@ protected:
 private:
     void set_player_window(){};
 
-    unsigned int     i_width, i_height;
-    unsigned int     i_tb_width, i_tb_height;
-
     VLCWindowsManager _WindowsManager;
-
-    int i_last_position;
 };
 
 #endif /* __VLCPLUGIN_WIN_H__ */
-- 
1.7.7.1.msysgit.0



From rsatom at gmail.com  Sat Dec 24 02:27:09 2011
From: rsatom at gmail.com (Sergey Radionov)
Date: Sat, 24 Dec 2011 08:27:09 +0700
Subject: [vlc-devel] [PATCH 1/7] npapi: added members for notification
 child classes about creation/destroying of media_player_t
In-Reply-To: <1324689365-2372-1-git-send-email-RSATom@gmail.com>
References: <1324689365-2372-1-git-send-email-RSATom@gmail.com>
Message-ID: <CADOtZwherQNRFodq372+=3LsZxQvm=J5XT63Q2DJ7M-yRp41fA@mail.gmail.com>

all patches from current patchset is available from
git://github.com/RSATom/gtk-npapi-vlc.git
or
https://github.com/RSATom/gtk-npapi-vlc/commits/master


From rsatom at gmail.com  Sat Dec 24 08:31:15 2011
From: rsatom at gmail.com (Sergey Radionov)
Date: Sat, 24 Dec 2011 14:31:15 +0700
Subject: [vlc-devel] [PATCH 1/3] npapi win32: added support of Event
	Listeners
Message-ID: <1324711877-1056-1-git-send-email-RSATom@gmail.com>

---
 npapi/support/npwin.cpp  |    7 +++++++
 npapi/vlcplugin_base.cpp |    2 +-
 2 files changed, 8 insertions(+), 1 deletions(-)

diff --git a/npapi/support/npwin.cpp b/npapi/support/npwin.cpp
index a189cc1..d3fa906 100644
--- a/npapi/support/npwin.cpp
+++ b/npapi/support/npwin.cpp
@@ -212,6 +212,13 @@ void NPN_Version(int* plugin_major, int* plugin_minor, int* netscape_major, int*
     *netscape_minor = LOBYTE(g_pNavigatorFuncs->version);
 }
 
+void NPN_PluginThreadAsyncCall(NPP plugin, void (*func)(void *), void *userData)
+{
+#if (((NP_VERSION_MAJOR << 8) + NP_VERSION_MINOR) >= 20)
+    (g_pNavigatorFuncs->pluginthreadasynccall)(plugin, func, userData);
+#endif
+}
+
 NPError NPN_GetValue(NPP instance, NPNVariable variable, void *result)
 {
     return g_pNavigatorFuncs->getvalue(instance, variable, result);
diff --git a/npapi/vlcplugin_base.cpp b/npapi/vlcplugin_base.cpp
index bd6db07..c100583 100644
--- a/npapi/vlcplugin_base.cpp
+++ b/npapi/vlcplugin_base.cpp
@@ -375,7 +375,7 @@ void VlcPluginBase::eventAsync(void *param)
 void VlcPluginBase::event_callback(const libvlc_event_t* event,
                 NPVariant *npparams, uint32_t npcount)
 {
-#ifdef XP_UNIX
+#if defined(XP_UNIX) || defined(XP_WIN)
     events.callback(event, npparams, npcount);
     NPN_PluginThreadAsyncCall(getBrowser(), eventAsync, this);
 #else
-- 
1.7.7.1.msysgit.0



From rsatom at gmail.com  Sat Dec 24 08:31:16 2011
From: rsatom at gmail.com (Sergey Radionov)
Date: Sat, 24 Dec 2011 14:31:16 +0700
Subject: [vlc-devel] [PATCH 2/3] npapi: fixed event delivery reentrancy
	problem
In-Reply-To: <1324711877-1056-1-git-send-email-RSATom@gmail.com>
References: <1324711877-1056-1-git-send-email-RSATom@gmail.com>
Message-ID: <1324711877-1056-2-git-send-email-RSATom@gmail.com>

---
 npapi/vlcplugin_base.cpp |    5 +++++
 npapi/vlcplugin_base.h   |    3 ++-
 2 files changed, 7 insertions(+), 1 deletions(-)

diff --git a/npapi/vlcplugin_base.cpp b/npapi/vlcplugin_base.cpp
index c100583..d4ed630 100644
--- a/npapi/vlcplugin_base.cpp
+++ b/npapi/vlcplugin_base.cpp
@@ -203,7 +203,11 @@ EventObj::~EventObj()
 
 void EventObj::deliver(NPP browser)
 {
+    if(_already_in_deliver)
+        return;
+
     plugin_lock(&lock);
+    _already_in_deliver = true;
 
     for( ev_l::iterator iter = _elist.begin(); iter != _elist.end(); ++iter )
     {
@@ -237,6 +241,7 @@ void EventObj::deliver(NPP browser)
     }
     _elist.clear();
 
+    _already_in_deliver = false;
     plugin_unlock(&lock);
 }
 
diff --git a/npapi/vlcplugin_base.h b/npapi/vlcplugin_base.h
index c9c8a59..67ffc3d 100644
--- a/npapi/vlcplugin_base.h
+++ b/npapi/vlcplugin_base.h
@@ -153,7 +153,7 @@ private:
     };
     libvlc_event_manager_t *_em; /* libvlc media_player event manager */
 public:
-    EventObj(): _em(NULL)  { /* deferred to init() */ }
+    EventObj(): _em(NULL), _already_in_deliver(false) { /* deferred to init() */ }
     bool init();
     ~EventObj();
 
@@ -172,6 +172,7 @@ private:
     ev_l _elist; /* scheduled events list for delivery to browser */
 
     plugin_lock_t lock;
+    bool _already_in_deliver;
 };
 
 typedef enum vlc_toolbar_clicked_e {
-- 
1.7.7.1.msysgit.0



From rsatom at gmail.com  Sat Dec 24 08:31:17 2011
From: rsatom at gmail.com (Sergey Radionov)
Date: Sat, 24 Dec 2011 14:31:17 +0700
Subject: [vlc-devel] [PATCH 3/3] npapi: allow a Event Listener to subscribe
	to more than one event type
In-Reply-To: <1324711877-1056-1-git-send-email-RSATom@gmail.com>
References: <1324711877-1056-1-git-send-email-RSATom@gmail.com>
Message-ID: <1324711877-1056-3-git-send-email-RSATom@gmail.com>

---
 npapi/vlcplugin_base.cpp |   19 +++++++++----------
 1 files changed, 9 insertions(+), 10 deletions(-)

diff --git a/npapi/vlcplugin_base.cpp b/npapi/vlcplugin_base.cpp
index d4ed630..fbdff1f 100644
--- a/npapi/vlcplugin_base.cpp
+++ b/npapi/vlcplugin_base.cpp
@@ -279,18 +279,17 @@ bool EventObj::insert(const NPString &name, NPObject *listener, bool bubble)
     if( !event )
         return false;
 
-    lr_l::iterator iter;
-    for( iter = _llist.begin(); iter != _llist.end(); ++iter )
-        if( iter->listener() == listener && iter->bubble() == bubble )
-            break;
-
-    if( iter == _llist.end() )
-    {
-        _llist.push_back(Listener(event, listener, bubble));
-        return true;
+    for(lr_l::iterator iter = _llist.begin(); iter != _llist.end(); ++iter ){
+        if(iter->listener() == listener &&
+           event->libvlc_type == iter->event_type() &&
+           iter->bubble() == bubble)
+        {
+            return false;
+        }
     }
 
-    return false;
+    _llist.push_back(Listener(event, listener, bubble));
+    return true;
 }
 
 bool EventObj::remove(const NPString &name, NPObject *listener, bool bubble)
-- 
1.7.7.1.msysgit.0



From rsatom at gmail.com  Sat Dec 24 08:34:24 2011
From: rsatom at gmail.com (Sergey Radionov)
Date: Sat, 24 Dec 2011 14:34:24 +0700
Subject: [vlc-devel] [PATCH 1/3] npapi win32: added support of Event
	Listeners
In-Reply-To: <1324711877-1056-1-git-send-email-RSATom@gmail.com>
References: <1324711877-1056-1-git-send-email-RSATom@gmail.com>
Message-ID: <CADOtZwi8A4dvbONf4JrZ-LYMMwD=Kub98Gc6fJtw0gUSMs1XZw@mail.gmail.com>

all patches from this and previous
(http://mailman.videolan.org/pipermail/vlc-devel/2011-December/084371.html)
patchsets are available
from:git://github.com/RSATom/gtk-npapi-vlc.gitorhttps://github.com/RSATom/gtk-npapi-vlc/commits/master


From kaarlo.raiha at gmail.com  Sat Dec 24 08:52:20 2011
From: kaarlo.raiha at gmail.com (=?ISO-8859-1?B?S2FhcmxvIFLkaWjk?=)
Date: Sat, 24 Dec 2011 09:52:20 +0200
Subject: [vlc-devel] start-time has no effect when transcoding
In-Reply-To: <1324663672.94288.YahooMailNeo@web125701.mail.ne1.yahoo.com>
References: <1324663672.94288.YahooMailNeo@web125701.mail.ne1.yahoo.com>
Message-ID: <CAMKrNtyr=JaTCX2TLwKJ9QQ6JXC4ge-9c0vd0_Ux-CqM3eqwzQ@mail.gmail.com>

2011/12/23 zhu shi song <zhushisongzhu at yahoo.com>

> I use VLC to  transcode one wmv video to mp4 video.  I hope I can clip the
> video using start-time and stop-time.  When reaching stop-time, the
> transcoding ended. But start-time has no effect,  transcoding always
> started from 0.  Who has the experience to cut one video using start-time
> to stop-time transcoding?
>

If the input file has incorrect timecodes or VLC cannot parse them
correctly, the start-time and stop-time can go to wrong spot.


>
> Thanks
>
> _______________________________________________
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
> http://mailman.videolan.org/listinfo/vlc-devel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111224/9a2de367/attachment.html>

From zhushisongzhu at yahoo.com  Sat Dec 24 12:47:55 2011
From: zhushisongzhu at yahoo.com (zhu shi song)
Date: Sat, 24 Dec 2011 03:47:55 -0800 (PST)
Subject: [vlc-devel] start-time has no effect when transcoding
Message-ID: <1324727275.36454.YahooMailNeo@web125706.mail.ne1.yahoo.com>

After several tests, I found that if? start-time worked ok for the video in MP4(h.264+AAC) format.? But start-time didn't work correctly using WMV format video.? I don't know if it's the bug or not.?? 

thanks



From zhushisongzhu at yahoo.com  Sat Dec 24 14:19:46 2011
From: zhushisongzhu at yahoo.com (zhu shi song)
Date: Sat, 24 Dec 2011 05:19:46 -0800 (PST)
Subject: [vlc-devel] the reason why start-time has no effect while
	transcoding
Message-ID: <1324732786.39888.YahooMailNeo@web125702.mail.ne1.yahoo.com>


I can't control progress bar while playing wmv video file. 
VLC-1.1.11,VLC-1.2.0-pre3 all have the same problem.? It's the reason 
why start-time can't work for transcoding.?? But? windows media player 
works fine to control the time progress bar while playing the video. So I think it's the bug of VLC.? How can I do to fix the problem?
Thanks.



From zhushisongzhu at yahoo.com  Sat Dec 24 14:51:55 2011
From: zhushisongzhu at yahoo.com (zhu shi song)
Date: Sat, 24 Dec 2011 05:51:55 -0800 (PST)
Subject: [vlc-devel] WMV video start-time problem
Message-ID: <1324734715.8495.YahooMailNeo@web125705.mail.ne1.yahoo.com>

When partially transcoding using start-time and stop-time, the real beginning time is always earlier 10 seconds than start-time preset.?
Is it? a bug of VLC? 



From kaarlo.raiha at gmail.com  Sat Dec 24 15:08:17 2011
From: kaarlo.raiha at gmail.com (=?ISO-8859-1?B?S2FhcmxvIFLkaWjk?=)
Date: Sat, 24 Dec 2011 16:08:17 +0200
Subject: [vlc-devel] WMV video start-time problem
In-Reply-To: <1324734715.8495.YahooMailNeo@web125705.mail.ne1.yahoo.com>
References: <1324734715.8495.YahooMailNeo@web125705.mail.ne1.yahoo.com>
Message-ID: <CAMKrNtzHGmjYS_SgStDw5+4JMFFxf2=yJPnBWYq2fsusy3gu9g@mail.gmail.com>

2011/12/24 zhu shi song <zhushisongzhu at yahoo.com>

> When partially transcoding using start-time and stop-time, the real
> beginning time is always earlier 10 seconds than start-time preset.
> Is it  a bug of VLC?
>

Please, don't multipost about the same issue. It won't help.

As I said earlier, it is either buggy file with wrong timecodes or bug in
VLC demuxing of those kinds of files.


>
> _______________________________________________
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
> http://mailman.videolan.org/listinfo/vlc-devel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111224/1867fe56/attachment.html>

From edward.c.wang at compdigitec.com  Sat Dec 24 19:59:27 2011
From: edward.c.wang at compdigitec.com (Edward Wang)
Date: Sat, 24 Dec 2011 13:59:27 -0500
Subject: [vlc-devel] [PATCH] Qt4: Add aspect ratio combobox to toolbar editor
Message-ID: <1324753167-26617-1-git-send-email-edward.c.wang@compdigitec.com>

---
 modules/gui/qt4/components/controller.cpp        |    3 ++
 modules/gui/qt4/components/controller.hpp        |    1 +
 modules/gui/qt4/components/controller_widget.cpp |   23 +++++++++++++++++++++
 modules/gui/qt4/components/controller_widget.hpp |   24 ++++++++++++++++++++++
 modules/gui/qt4/dialogs/toolbar.cpp              |    8 +++++++
 5 files changed, 59 insertions(+), 0 deletions(-)

diff --git a/modules/gui/qt4/components/controller.cpp b/modules/gui/qt4/components/controller.cpp
index 3d1bd78..22d3792 100644
--- a/modules/gui/qt4/components/controller.cpp
+++ b/modules/gui/qt4/components/controller.cpp
@@ -465,6 +465,9 @@ QWidget *AbstractController::createWidget( buttonType_e button, int options )
         CONNECT_MAP_SET( play, PLAY_ACTION );
         }
         break;
+    case ASPECT_RATIO_COMBOBOX:
+        widget = new AspectRatioComboBox( p_intf );
+        break;
     default:
         msg_Warn( p_intf, "This should not happen %i", button );
         break;
diff --git a/modules/gui/qt4/components/controller.hpp b/modules/gui/qt4/components/controller.hpp
index 42ce504..901d3df 100644
--- a/modules/gui/qt4/components/controller.hpp
+++ b/modules/gui/qt4/components/controller.hpp
@@ -99,6 +99,7 @@ typedef enum buttonType_e
     TELETEXT_BUTTONS,
     ADVANCED_CONTROLLER,
     PLAYBACK_BUTTONS,
+    ASPECT_RATIO_COMBOBOX,
     SPECIAL_MAX,
 
     WIDGET_SPACER = 0x40,
diff --git a/modules/gui/qt4/components/controller_widget.cpp b/modules/gui/qt4/components/controller_widget.cpp
index acd8b7f..370ef3f 100644
--- a/modules/gui/qt4/components/controller_widget.cpp
+++ b/modules/gui/qt4/components/controller_widget.cpp
@@ -267,3 +267,26 @@ void LoopButton::updateButtonIcons( int value )
     setIcon( ( value == REPEAT_ONE ) ? QIcon( ":/buttons/playlist/repeat_one" )
                                      : QIcon( ":/buttons/playlist/repeat_all" ) );
 }
+
+void AspectRatioComboBox::updateRatios() {
+    /* Clear the list before updating */
+    this->clear();
+    vlc_value_t val_list, text_list;
+    vout_thread_t* p_vout = THEMIM->getVout();
+    /* Disable if there is no vout */
+    if( p_vout == NULL ) {
+        addItem("Aspect Ratio");
+        setDisabled( true );
+        return;
+    }
+    var_Change( p_vout, "aspect-ratio", VLC_VAR_GETLIST, &val_list, &text_list );
+    for( int i = 0; i < val_list.p_list->i_count; i++ )
+        addItem( QString( text_list.p_list->p_values[i].psz_string ), QString( val_list.p_list->p_values[i].psz_string ) );
+    setEnabled( true );
+    var_FreeList( &val_list, &text_list );
+}
+
+void AspectRatioComboBox::updateAspectRatio( int x ) {
+    vout_thread_t* p_vout = THEMIM->getVout();
+    if( p_vout && x >= 0 ) var_SetString( p_vout, "aspect-ratio", qtu( itemData(x).toString() ) );
+}
diff --git a/modules/gui/qt4/components/controller_widget.hpp b/modules/gui/qt4/components/controller_widget.hpp
index 9b5af88..04663f5 100644
--- a/modules/gui/qt4/components/controller_widget.hpp
+++ b/modules/gui/qt4/components/controller_widget.hpp
@@ -29,9 +29,12 @@
 #endif
 
 #include "qt4.hpp"
+#include "input_manager.hpp"
+#include <vlc_vout.h>                       /* vout_thread_t for aspect ratio combobox */
 
 #include <QWidget>
 #include <QToolButton>
+#include <QComboBox>
 
 class QLabel;
 class QFrame;
@@ -67,6 +70,27 @@ private slots:
     void updateButtonIcons( bool, bool );
 };
 
+class AspectRatioComboBox : public QComboBox
+{
+    Q_OBJECT
+    public:
+    AspectRatioComboBox( intf_thread_t* _p_intf ) {
+        p_intf = _p_intf;
+        CONNECT( THEMIM->getIM(), voutChanged( bool ),
+                 this, updateRatios() );
+        CONNECT( this, currentIndexChanged( int ),
+                 this, updateAspectRatio( int ) );
+        this->updateRatios();
+    }
+
+    public slots:
+    void updateRatios();
+    void updateAspectRatio( int );
+
+    private:
+    intf_thread_t* p_intf;
+};
+
 #define VOLUME_MAX 200
 class SoundWidget : public QWidget
 {
diff --git a/modules/gui/qt4/dialogs/toolbar.cpp b/modules/gui/qt4/dialogs/toolbar.cpp
index 93401ac..616a03e 100644
--- a/modules/gui/qt4/dialogs/toolbar.cpp
+++ b/modules/gui/qt4/dialogs/toolbar.cpp
@@ -35,6 +35,10 @@
 #include "util/buttons/BrowseButton.hpp"
 #include "util/buttons/RoundButton.hpp"
 
+#include "qt4.hpp"
+#include "input_manager.hpp"
+#include <vlc_vout.h>                       /* vout_thread_t for aspect ratio combobox */
+
 #include <QScrollArea>
 #include <QGroupBox>
 #include <QLabel>
@@ -435,6 +439,10 @@ WidgetListing::WidgetListing( intf_thread_t *p_intf, QWidget *_parent )
                 layout->setRoundButton( play );
             }
             break;
+        case ASPECT_RATIO_COMBOBOX:
+            widget = new AspectRatioComboBox( p_intf );
+            widgetItem->setText( qtr("Aspect ratio selector") );
+            break;
         default:
             msg_Warn( p_intf, "This should not happen %i", i );
             break;
-- 
1.7.5.4



From rsatom at gmail.com  Sun Dec 25 04:54:34 2011
From: rsatom at gmail.com (Sergey Radionov)
Date: Sun, 25 Dec 2011 10:54:34 +0700
Subject: [vlc-devel] [PATCH 1/1] add bin to PATH to let libvlc.dll be found
	by vlc-cache-gen.exe
Message-ID: <1324785274-696-1-git-send-email-RSATom@gmail.com>

Fix "This application has failed to start because libvlc.dll
was not found. Re-installing the application may fix this problem."
during building of vlc under Windows.
---
 Makefile.am |    1 +
 1 files changed, 1 insertions(+), 0 deletions(-)

diff --git a/Makefile.am b/Makefile.am
index 358ac5f..dae8ad1 100644
--- a/Makefile.am
+++ b/Makefile.am
@@ -406,6 +406,7 @@ vlc$(EXEEXT):
 ###############################################################################
 install-exec-hook:
 	if test "$(build)" = "$(host)"; then \
+		PATH="$(DESTDIR)$(bindir):$$PATH" \
 		LD_LIBRARY_PATH="$(DESTDIR)$(libdir):$$LD_LIBRARY_PATH" \
 		"$(DESTDIR)$(vlclibdir)/vlc-cache-gen$(EXEEXT)" \
 			 "$(DESTDIR)$(vlclibdir)/plugins" ; \
-- 
1.7.7.1.msysgit.0



From jpsaman at videolan.org  Sun Dec 25 14:00:58 2011
From: jpsaman at videolan.org (Jean-Paul Saman)
Date: Sun, 25 Dec 2011 14:00:58 +0100
Subject: [vlc-devel] How to trim/cut clips with VLC
In-Reply-To: <1324664861.86424.YahooMailNeo@web125701.mail.ne1.yahoo.com>
References: <1324664861.86424.YahooMailNeo@web125701.mail.ne1.yahoo.com>
Message-ID: <CAK9US3A+D_G4gp1wSbXUf51dmZTjFreVwqnqRcb5oZbnbWATjQ@mail.gmail.com>

On Fri, Dec 23, 2011 at 7:27 PM, zhu shi song <zhushisongzhu at yahoo.com> wrote:
> I using the following command to transcode one piece of video from start-time to stop-time.
>
> vlc hzvideo.wmv --start-time=10 --stop-time=20 --sout=#transcode{vcodec=h264, venc=x264{preset=veryfast}, vb=0,scale=0,acodec=mp4a,
> ab=128,channels=2,samplerate=44100}:file{dst='11.mp4'}
>
> when reaching stop-time, the transcoding ended.? But start-time had no effect, it always started from the beginning.? What's wrong with it?
> Thanks

Both start-time and stop-time work for playback and have never been
tested with streaming or transcoding.

Kind regards,
Jean-Paul Saman


From kaarlo.raiha at gmail.com  Sun Dec 25 18:53:14 2011
From: kaarlo.raiha at gmail.com (=?ISO-8859-1?B?S2FhcmxvIFLkaWjk?=)
Date: Sun, 25 Dec 2011 19:53:14 +0200
Subject: [vlc-devel] How to trim/cut clips with VLC
In-Reply-To: <CAK9US3A+D_G4gp1wSbXUf51dmZTjFreVwqnqRcb5oZbnbWATjQ@mail.gmail.com>
References: <1324664861.86424.YahooMailNeo@web125701.mail.ne1.yahoo.com>
	<CAK9US3A+D_G4gp1wSbXUf51dmZTjFreVwqnqRcb5oZbnbWATjQ@mail.gmail.com>
Message-ID: <CAMKrNtwEirW48QYog=zwU8_zc5cSD8kKPNyJbH7o1LBi9Q94PA@mail.gmail.com>

2011/12/25 Jean-Paul Saman <jpsaman at videolan.org>

> On Fri, Dec 23, 2011 at 7:27 PM, zhu shi song <zhushisongzhu at yahoo.com>
> wrote:
> > I using the following command to transcode one piece of video from
> start-time to stop-time.
> >
> > vlc hzvideo.wmv --start-time=10 --stop-time=20
> --sout=#transcode{vcodec=h264, venc=x264{preset=veryfast},
> vb=0,scale=0,acodec=mp4a,
> > ab=128,channels=2,samplerate=44100}:file{dst='11.mp4'}
> >
> > when reaching stop-time, the transcoding ended.  But start-time had no
> effect, it always started from the beginning.  What's wrong with it?
> > Thanks
>
> Both start-time and stop-time work for playback and have never been
> tested with streaming or transcoding.
>
>
I think they were tested, since AFAIK wizard (the one used with Mac) does
partial with start-time and stop-time. And they do work for streaming and
transcoding according to the Windows forum users.


> Kind regards,
> Jean-Paul Saman
> _______________________________________________
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
> http://mailman.videolan.org/listinfo/vlc-devel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111225/649348b6/attachment.html>

From typx at dinauz.org  Mon Dec 26 00:39:13 2011
From: typx at dinauz.org (Denis Charmet)
Date: Mon, 26 Dec 2011 00:39:13 +0100
Subject: [vlc-devel] [PATCH 2/2] Fix a crash with corrupted MKV
Message-ID: <1324856353-16722-1-git-send-email-typx@dinauz.org>

If the blocksize is corrupted and has a lace, you may have a buffer overflow. Should fix #5658.
---
 modules/demux/mkv/mkv.cpp |   15 ++++++++++++++-
 1 files changed, 14 insertions(+), 1 deletions(-)

diff --git a/modules/demux/mkv/mkv.cpp b/modules/demux/mkv/mkv.cpp
index a350a43..5f52656 100644
--- a/modules/demux/mkv/mkv.cpp
+++ b/modules/demux/mkv/mkv.cpp
@@ -520,6 +520,14 @@ void BlockDecode( demux_t *p_demux, KaxBlock *block, KaxSimpleBlock *simpleblock
     tk->b_inited = true;
 
 
+    size_t frame_size = 0;
+    size_t block_size = 0;
+
+    if( simpleblock != NULL )
+        block_size = simpleblock->GetSize();
+    else
+        block_size = block->GetSize();
+ 
     for( unsigned int i = 0;
          ( block != NULL && i < block->NumberFrames()) || ( simpleblock != NULL && i < simpleblock->NumberFrames() );
          i++ )
@@ -535,9 +543,14 @@ void BlockDecode( demux_t *p_demux, KaxBlock *block, KaxSimpleBlock *simpleblock
         else
         {
             data = &block->GetBuffer(i);
+            // condition when the DTS is correct (keyframe or B frame == NOT P frame)
         }
-        if( !data->Buffer() || data->Size() > SIZE_MAX )
+        frame_size += data->Size();
+        if( !data->Buffer() || data->Size() > SIZE_MAX || frame_size > block_size  )
+        {
+            msg_Warn( p_demux, "Cannot read frame (too long or no frame)" );
             break;
+        }
 
         if( tk->i_compression_type == MATROSKA_COMPRESSION_HEADER && tk->p_compression_data != NULL )
             p_block = MemToBlock( data->Buffer(), data->Size(), tk->p_compression_data->GetSize() );
-- 
1.7.7.3



From funman at videolan.org  Mon Dec 26 02:03:54 2011
From: funman at videolan.org (=?UTF-8?q?Rafa=C3=ABl=20Carr=C3=A9?=)
Date: Sun, 25 Dec 2011 20:03:54 -0500
Subject: [vlc-devel] [PATCH] bluray: don't crash when file is NULL
Message-ID: <1324861434-24174-1-git-send-email-funman@videolan.org>

e.g. vlc bluray://0
---
 modules/access/bluray.c |    8 +++++---
 1 files changed, 5 insertions(+), 3 deletions(-)

diff --git a/modules/access/bluray.c b/modules/access/bluray.c
index 983eade..7bfc3f0 100644
--- a/modules/access/bluray.c
+++ b/modules/access/bluray.c
@@ -94,7 +94,7 @@ static int blurayOpen( vlc_object_t *object )
 
     char *pos_title;
     int i_title = -1;
-    char bd_path[PATH_MAX];
+    char bd_path[PATH_MAX] = { '\0' };
     const char *error_msg = NULL;
 
     if (strcmp(p_demux->psz_access, "bluray")) {
@@ -117,8 +117,10 @@ static int blurayOpen( vlc_object_t *object )
     TAB_INIT( p_sys->i_title, p_sys->pp_title );
 
     /* store current bd_path */
-    strncpy(bd_path, p_demux->psz_file, sizeof(bd_path));
-    bd_path[PATH_MAX - 1] = '\0';
+    if (p_demux->psz_file) {
+        strncpy(bd_path, p_demux->psz_file, sizeof(bd_path));
+        bd_path[PATH_MAX - 1] = '\0';
+    }
 
     p_sys->bluray = bd_open(bd_path, NULL);
     if (!p_sys->bluray) {
-- 
1.7.7.3


From hugo at videolan.org  Mon Dec 26 15:31:40 2011
From: hugo at videolan.org (=?UTF-8?Q?Hugo_Beauz=C3=A9e=2DLuyssen?=)
Date: Mon, 26 Dec 2011 15:31:40 +0100
Subject: [vlc-devel] FOSDEM 2012
In-Reply-To: <CAK-C75QBGcaC-GNZoWSHq_m0TO_DcDsd4xFeFvjj3G-ZNWe_Ug@mail.gmail.com>
References: <CAK-C75QBGcaC-GNZoWSHq_m0TO_DcDsd4xFeFvjj3G-ZNWe_Ug@mail.gmail.com>
Message-ID: <CAK-C75R+AvXJ5XRcGHsYDJa1F2A5sGTyea72oTKvDtfxyXe-7Q@mail.gmail.com>

On Wed, Dec 14, 2011 at 11:43 AM, Hugo Beauz?e-Luyssen
<hugo at videolan.org> wrote:
> Hi,
>
> As some of you may know, the VideoLAN team will meet at FOSDEM! So if
> you wish to join us, now is the time to speak!
>
> FOSDEM 2012 will take place on February 4th, 5th. There's a beer event
> on February 3rd, for those interested.
>
> As usual, for active community members, VideoLAN will pay for one
> night, and for the travel. ?That means if you wish to attend the beer
> event, you'll be paying the extra hotel night.
>
> We will book rooms for two, so you'll have to share a room with a fellow geek.
> If you wish to attend, you can either mail trez at videolan.org with some
> basic info :
> ? ?- When do you arrive? Friday or Saturday?
> ? ?- Where will you be taking your train/plane from, and do you
> travel back, or to another location.
> ? ?- Do you have a special geek/dev/pony you want to share the room with?
> ? ?- Will you book your ticket by yourself? Or do you want me to book
> it for you? (This can ease up the refund, especially if you leave
> outside the EU)
>
> Or you can just fill this form out:
> https://docs.google.com/spreadsheet/viewform?formkey=dF9VUDFCeVRIaHZGYS1OVlFYVXQyZ0E6MQ&ifq
>
> If you book your ticket yourself, please do it quick! The sooner the cheaper.
> Please answer by the end of December so we can book the hotel.
>
> Regards,
>

Ping,

Please respond by the end of December or I won't be able to book the
hotel for you.

Regards,

-- 
Hugo Beauz?e-Luyssen


From typx at dinauz.org  Mon Dec 26 19:29:31 2011
From: typx at dinauz.org (Denis Charmet)
Date: Mon, 26 Dec 2011 19:29:31 +0100
Subject: [vlc-devel] [PATCH] Fix seeking without cues in mkv
Message-ID: <1324924171-19629-1-git-send-email-typx@dinauz.org>

the demux generates its own cues while reading and when seeking.
This should also avoid crashes when seeking without cues.

Closes #5085 and #5712
---
 modules/demux/mkv/matroska_segment.cpp |   43 ++++++++++++++++---------------
 modules/demux/mkv/mkv.cpp              |   30 +++++++++------------
 2 files changed, 35 insertions(+), 38 deletions(-)

diff --git a/modules/demux/mkv/matroska_segment.cpp b/modules/demux/mkv/matroska_segment.cpp
index 072dd71..6bb18ee 100644
--- a/modules/demux/mkv/matroska_segment.cpp
+++ b/modules/demux/mkv/matroska_segment.cpp
@@ -427,7 +427,7 @@ void matroska_segment_c::IndexAppendCluster( KaxCluster *cluster )
     idx.i_track       = -1;
     idx.i_block_number= -1;
     idx.i_position    = cluster->GetElementPosition();
-    idx.i_time        = -1;
+    idx.i_time        = cluster->GlobalTimecode()/ (mtime_t) 1000;
     idx.b_key         = true;
 
     i_index++;
@@ -668,8 +668,9 @@ bool matroska_segment_c::LoadSeekHeadItem( const EbmlCallbacks & ClassInfos, int
 
 struct spoint
 {
-    spoint(unsigned int tk): i_track(tk),i_date(0), i_seek_pos(0), i_cluster_pos(0),
-                             p_next(NULL){}
+    spoint(unsigned int tk, mtime_t date, int64_t pos, int64_t cpos):
+        i_track(tk),i_date(date), i_seek_pos(pos),
+        i_cluster_pos(cpos), p_next(NULL){}
     unsigned int     i_track;
     mtime_t i_date;
     int64_t i_seek_pos;
@@ -693,7 +694,13 @@ void matroska_segment_c::Seek( mtime_t i_date, mtime_t i_time_offset, int64_t i_
     {
         /* Special case for seeking in files with no cues */
         EbmlElement *el = NULL;
-        es.I_O().setFilePointer( i_start_pos, seek_beginning );
+
+        /* Start from the last known index instead of the beginning eachtime */
+        if( i_index == 0)
+            es.I_O().setFilePointer( i_start_pos, seek_beginning );
+        else
+            es.I_O().setFilePointer( p_indexes[ i_index - 1 ].i_position,
+                                     seek_beginning );
         delete ep;
         ep = new EbmlParser( &es, segment, &sys.demuxer );
         cluster = NULL;
@@ -705,20 +712,16 @@ void matroska_segment_c::Seek( mtime_t i_date, mtime_t i_time_offset, int64_t i_
                 cluster = (KaxCluster *)el;
                 i_cluster_pos = cluster->GetElementPosition();
                 if( i_index == 0 ||
-                        ( i_index > 0 && p_indexes[i_index - 1].i_position < (int64_t)cluster->GetElementPosition() ) )
+                    ( i_index > 0 &&
+                      p_indexes[i_index - 1].i_position < (int64_t)cluster->GetElementPosition() ) )
                 {
+                    ParseCluster();
                     IndexAppendCluster( cluster );
                 }
                 if( es.I_O().getFilePointer() >= (unsigned) i_global_position )
-                {
-                    ParseCluster();
-                    msg_Dbg( &sys.demuxer, "we found a cluster that is in the neighbourhood" );
-                    return;
-                }
+                    break;
             }
         }
-        msg_Err( &sys.demuxer, "This file has no cues, and we were unable to seek to the requested position by parsing." );
-        return;
     }
 
     /* Don't try complex seek if we seek to 0 */
@@ -766,7 +769,7 @@ void matroska_segment_c::Seek( mtime_t i_date, mtime_t i_time_offset, int64_t i_
     {
         if( tracks[i_track]->fmt.i_cat == VIDEO_ES )
         {
-            spoint * seekpoint = new spoint(i_track);
+            spoint * seekpoint = new spoint(i_track, i_seek_time, i_seek_position, i_seek_position);
             if( unlikely( !seekpoint ) )
             {
                 for( spoint * sp = p_first; sp; )
@@ -813,7 +816,6 @@ void matroska_segment_c::Seek( mtime_t i_date, mtime_t i_time_offset, int64_t i_
             i_pts = sys.i_chapter_time + simpleblock->GlobalTimecode() / (mtime_t) 1000;
         else
             i_pts = sys.i_chapter_time + block->GlobalTimecode() / (mtime_t) 1000;
-
         if( i_track < tracks.size() )
         {
             if( tracks[i_track]->fmt.i_cat == VIDEO_ES && b_key_picture )
@@ -1441,13 +1443,6 @@ int matroska_segment_c::BlockGet( KaxBlock * & pp_block, KaxSimpleBlock * & pp_s
                 cluster = (KaxCluster*)el;
                 i_cluster_pos = cluster->GetElementPosition();
 
-                /* add it to the index */
-                if( i_index == 0 ||
-                    ( i_index > 0 && p_indexes[i_index - 1].i_position < (int64_t)cluster->GetElementPosition() ) )
-                {
-                    IndexAppendCluster( cluster );
-                }
-
                 // reset silent tracks
                 for (size_t i=0; i<tracks.size(); i++)
                 {
@@ -1473,6 +1468,12 @@ int matroska_segment_c::BlockGet( KaxBlock * & pp_block, KaxSimpleBlock * & pp_s
 
                 ctc.ReadData( es.I_O(), SCOPE_ALL_DATA );
                 cluster->InitTimecode( uint64( ctc ), i_timescale );
+ 
+                /* add it to the index */
+                if( i_index == 0 ||
+                    ( i_index > 0 &&
+                      p_indexes[i_index - 1].i_position < (int64_t)cluster->GetElementPosition() ) )
+                    IndexAppendCluster( cluster );
             }
             else if( MKV_IS_ID( el, KaxClusterSilentTracks ) )
             {
diff --git a/modules/demux/mkv/mkv.cpp b/modules/demux/mkv/mkv.cpp
index 5f52656..ebbcafa 100644
--- a/modules/demux/mkv/mkv.cpp
+++ b/modules/demux/mkv/mkv.cpp
@@ -410,49 +410,45 @@ static void Seek( demux_t *p_demux, mtime_t i_date, double f_percent, virtual_ch
     msg_Dbg( p_demux, "seek request to %"PRId64" (%f%%)", i_date, f_percent );
     if( i_date < 0 && f_percent < 0 )
     {
-        msg_Warn( p_demux, "cannot seek nowhere !" );
+        msg_Warn( p_demux, "cannot seek nowhere!" );
         return;
     }
     if( f_percent > 1.0 )
     {
-        msg_Warn( p_demux, "cannot seek so far !" );
+        msg_Warn( p_demux, "cannot seek so far!" );
+        return;
+    }
+    if( p_sys->f_duration < 0 )
+    {
+        msg_Warn( p_demux, "cannot seek without duration!");
         return;
     }
 
     /* seek without index or without date */
     if( f_percent >= 0 && (var_InheritBool( p_demux, "mkv-seek-percent" ) || !p_segment->b_cues || i_date < 0 ))
     {
-        if( p_sys->f_duration >= 0 && p_segment->b_cues )
-        {
-            i_date = int64_t( f_percent * p_sys->f_duration * 1000.0 );
-        }
-        else
+        i_date = int64_t( f_percent * p_sys->f_duration * 1000.0 );
+        if( !p_segment->b_cues )
         {
             int64_t i_pos = int64_t( f_percent * stream_Size( p_demux->s ) );
 
-            msg_Dbg( p_demux, "inaccurate way of seeking for pos:%"PRId64, i_pos );
+            msg_Dbg( p_demux, "lengthy way of seeking for pos:%"PRId64, i_pos );
             for( i_index = 0; i_index < p_segment->i_index; i_index++ )
             {
-                if( p_segment->b_cues && p_segment->p_indexes[i_index].i_position < i_pos )
-                    break;
-                if( !p_segment->b_cues && p_segment->p_indexes[i_index].i_position >= i_pos && p_segment->p_indexes[i_index].i_time > 0 )
+                if( p_segment->p_indexes[i_index].i_position >= i_pos &&
+                    p_segment->p_indexes[i_index].i_time > 0 )
                     break;
             }
             if( i_index == p_segment->i_index )
-            {
                 i_index--;
-            }
 
-            i_date = p_segment->p_indexes[i_index].i_time;
-
-            if( !p_segment->b_cues && ( p_segment->p_indexes[i_index].i_position < i_pos || p_segment->p_indexes[i_index].i_position - i_pos > 2000000 ))
+            if( p_segment->p_indexes[i_index].i_position < i_pos )
             {
                 msg_Dbg( p_demux, "no cues, seek request to global pos: %"PRId64, i_pos );
                 i_global_position = i_pos;
             }
         }
     }
-
     p_vsegment->Seek( *p_demux, i_date, i_time_offset, p_chapter, i_global_position );
 }
 
-- 
1.7.7.3



From brendonjustin at gmail.com  Tue Dec 27 03:07:31 2011
From: brendonjustin at gmail.com (Brendon Justin)
Date: Mon, 26 Dec 2011 21:07:31 -0500
Subject: [vlc-devel] [PATCH] Make the Mac DMG prettier.
Message-ID: <1324951651-31383-1-git-send-email-brendonjustin@gmail.com>

---
 extras/package/macosx/dmg_setup.scpt |   25 +++++++++++++++++++++++++
 extras/package/macosx/package.mak    |    5 ++++-
 2 files changed, 29 insertions(+), 1 deletions(-)
 create mode 100644 extras/package/macosx/dmg_setup.scpt

diff --git a/extras/package/macosx/dmg_setup.scpt b/extras/package/macosx/dmg_setup.scpt
new file mode 100644
index 0000000..f50ab36
--- /dev/null
+++ b/extras/package/macosx/dmg_setup.scpt
@@ -0,0 +1,25 @@
+on run argv
+    tell application "Finder"
+        tell disk (item 1 of argv)
+            open
+            set current view of container window to icon view
+            set toolbar visible of container window to false
+            set statusbar visible of container window to false
+            set the bounds of container window to {300, 100, 750, 500}
+            set theViewOptions to the icon view options of container window
+            set arrangement of theViewOptions to not arranged
+            set icon size of theViewOptions to 104
+            # Don't set a background image, for now
+            # set background picture of theViewOptions to file ".background/background.png"
+            set position of item "VLC.app" of container window to {110, 100}
+            set position of item "Applications" of container window to {335, 100}
+            set position of item "Read Me.rtf" of container window to {110, 275}
+            set position of item "Goodies" of container window to {335, 275}
+	    # Force saving changes to the disk by closing and opening the window
+            close
+            open
+            update without registering applications
+            delay 5
+        end tell
+    end tell
+end run
diff --git a/extras/package/macosx/package.mak b/extras/package/macosx/package.mak
index ab4d096..1c8b2f6 100644
--- a/extras/package/macosx/package.mak
+++ b/extras/package/macosx/package.mak
@@ -67,7 +67,10 @@ package-macosx: VLC-release.app
 	cp $(srcdir)/extras/package/macosx/Resources/about_bg.png $(top_builddir)/vlc-$(VERSION)/.background/background.png
 	$(LN_S) -f /Applications $(top_builddir)/vlc-$(VERSION)/
 	rm -f "$(top_builddir)/vlc-$(VERSION)-rw.dmg"
-	hdiutil create -verbose -srcfolder "$(top_builddir)/vlc-$(VERSION)" "$(top_builddir)/vlc-$(VERSION)-rw.dmg" -scrub
+	hdiutil create -verbose -srcfolder "$(top_builddir)/vlc-$(VERSION)" "$(top_builddir)/vlc-$(VERSION)-rw.dmg" -scrub -format UDRW
+	hdiutil attach -readwrite -noverify -noautoopen "vlc-$(VERSION)-rw.dmg"
+	osascript $(srcdir)/extras/package/macosx/dmg_setup.scpt vlc-$(VERSION)
+	hdiutil detach /Volumes/vlc-$(VERSION)
 # Make sure the image is not writable
 # Note: We can't directly create a read only dmg as we do the bless stuff
 	rm -f "$(top_builddir)/vlc-$(VERSION).dmg"
-- 
1.7.5.4



From funman at videolan.org  Tue Dec 27 03:30:23 2011
From: funman at videolan.org (=?UTF-8?q?Rafa=C3=ABl=20Carr=C3=A9?=)
Date: Mon, 26 Dec 2011 21:30:23 -0500
Subject: [vlc-devel] [PATCH 2/5] contrib: lua: fix android build
In-Reply-To: <1324953026-2767-1-git-send-email-funman@videolan.org>
References: <1324953026-2767-1-git-send-email-funman@videolan.org>
Message-ID: <1324953026-2767-3-git-send-email-funman@videolan.org>

android's struct lconv is empty
according to the headers it's not supported at all, and defined only to
make libstdc++ build happy
---
 contrib/src/lua/android.patch |   13 +++++++++++++
 contrib/src/lua/rules.mak     |    3 +++
 2 files changed, 16 insertions(+), 0 deletions(-)
 create mode 100644 contrib/src/lua/android.patch

diff --git a/contrib/src/lua/android.patch b/contrib/src/lua/android.patch
new file mode 100644
index 0000000..02d67c0
--- /dev/null
+++ b/contrib/src/lua/android.patch
@@ -0,0 +1,13 @@
+--- lua/src/llex.c.orig	2011-12-25 23:39:35.254829002 -0500
++++ lua/src/llex.c	2011-12-25 23:39:59.662829001 -0500
+@@ -176,9 +176,8 @@
+ 
+ static void trydecpoint (LexState *ls, SemInfo *seminfo) {
+   /* format error: try to update decimal point separator */
+-  struct lconv *cv = localeconv();
+   char old = ls->decpoint;
+-  ls->decpoint = (cv ? cv->decimal_point[0] : '.');
++  ls->decpoint = '.';
+   buffreplace(ls, old, ls->decpoint);  /* try updated decimal separator */
+   if (!luaO_str2d(luaZ_buffer(ls->buff), &seminfo->r)) {
+     /* format error with correct decimal point: no more options */
diff --git a/contrib/src/lua/rules.mak b/contrib/src/lua/rules.mak
index fc11618..328d96f 100644
--- a/contrib/src/lua/rules.mak
+++ b/contrib/src/lua/rules.mak
@@ -33,6 +33,9 @@ lua: lua-$(LUA_VERSION).tar.gz .sum-lua
 	$(UNPACK)
 	$(APPLY) $(SRC)/lua/lua-noreadline.patch
 	$(APPLY) $(SRC)/lua/luac-32bits.patch
+ifdef HAVE_ANDROID
+	$(APPLY) $(SRC)/lua/android.patch
+endif
 ifdef HAVE_MACOSX
 	(cd $(UNPACK_DIR) && \
 	sed -e 's%gcc%$(CC)%' \
-- 
1.7.7.3


From funman at videolan.org  Tue Dec 27 03:30:22 2011
From: funman at videolan.org (=?UTF-8?q?Rafa=C3=ABl=20Carr=C3=A9?=)
Date: Mon, 26 Dec 2011 21:30:22 -0500
Subject: [vlc-devel] [PATCH 1/5] contrib: gme: fix android build
In-Reply-To: <1324953026-2767-1-git-send-email-funman@videolan.org>
References: <1324953026-2767-1-git-send-email-funman@videolan.org>
Message-ID: <1324953026-2767-2-git-send-email-funman@videolan.org>

gcc doesn't understand __restrict, use __restrict__
might break MSVC++ compatibility
---
 contrib/src/gme/android.patch |   11 +++++++++++
 contrib/src/gme/rules.mak     |    1 +
 2 files changed, 12 insertions(+), 0 deletions(-)
 create mode 100644 contrib/src/gme/android.patch

diff --git a/contrib/src/gme/android.patch b/contrib/src/gme/android.patch
new file mode 100644
index 0000000..f615a7d
--- /dev/null
+++ b/contrib/src/gme/android.patch
@@ -0,0 +1,11 @@
+--- game-music-emu/gme/Blip_Buffer.h.orig	2011-12-25 22:27:15.546377001 -0500
++++ game-music-emu/gme/Blip_Buffer.h	2011-12-25 22:27:37.146377002 -0500
+@@ -268,7 +268,7 @@
+ };
+ 
+ 	#if defined (__GNUC__) || _MSC_VER >= 1100
+-		#define BLIP_RESTRICT __restrict
++		#define BLIP_RESTRICT __restrict__
+ 	#else
+ 		#define BLIP_RESTRICT
+ 	#endif
diff --git a/contrib/src/gme/rules.mak b/contrib/src/gme/rules.mak
index ac5dcc5..ce633b8 100644
--- a/contrib/src/gme/rules.mak
+++ b/contrib/src/gme/rules.mak
@@ -13,6 +13,7 @@ $(TARBALLS)/game-music-emu-$(GME_VERSION).tar.bz2:
 game-music-emu: game-music-emu-$(GME_VERSION).tar.bz2 .sum-gme
 	$(UNPACK)
 	$(APPLY) $(SRC)/gme/gme-static.patch
+	$(APPLY) $(SRC)/gme/android.patch
 	$(MOVE)
 
 .gme: game-music-emu toolchain.cmake
-- 
1.7.7.3


From funman at videolan.org  Tue Dec 27 03:30:24 2011
From: funman at videolan.org (=?UTF-8?q?Rafa=C3=ABl=20Carr=C3=A9?=)
Date: Mon, 26 Dec 2011 21:30:24 -0500
Subject: [vlc-devel] [PATCH 3/5] contrib: gnutls: fix android build
In-Reply-To: <1324953026-2767-1-git-send-email-funman@videolan.org>
References: <1324953026-2767-1-git-send-email-funman@videolan.org>
Message-ID: <1324953026-2767-4-git-send-email-funman@videolan.org>

gl/ builds replacements for standard headers in case the system ones don't work
but it has the reverse effect: provided headers work and built ones don't

also re-run autoreconf so configure knows about the NDK toolchain (arm-linux-androideabi-)
---
 contrib/src/gnutls/no-gl.patch |   20 ++++++++++++++++++++
 contrib/src/gnutls/rules.mak   |    4 ++++
 2 files changed, 24 insertions(+), 0 deletions(-)
 create mode 100644 contrib/src/gnutls/no-gl.patch

diff --git a/contrib/src/gnutls/no-gl.patch b/contrib/src/gnutls/no-gl.patch
new file mode 100644
index 0000000..ac2b92a
--- /dev/null
+++ b/contrib/src/gnutls/no-gl.patch
@@ -0,0 +1,20 @@
+--- gnutls/lib/Makefile.am.orig	2011-12-25 23:15:45.170377002 -0500
++++ gnutls/lib/Makefile.am	2011-12-25 23:15:54.094377001 -0500
+@@ -23,7 +23,7 @@
+ 
+ ACLOCAL_AMFLAGS = -I m4 -I gl/m4
+ 
+-SUBDIRS = gl po includes x509
++SUBDIRS = po includes x509
+ if ENABLE_MINITASN1
+ SUBDIRS += minitasn1
+ endif
+@@ -120,7 +120,7 @@
+ libgnutls_la_LDFLAGS = -no-undefined \
+ 	-version-info $(LT_CURRENT):$(LT_REVISION):$(LT_AGE)
+ 
+-libgnutls_la_LIBADD = gl/liblgnu.la x509/libgnutls_x509.la \
++libgnutls_la_LIBADD = x509/libgnutls_x509.la \
+ 	$(LTLIBZ)  $(LTLIBINTL) $(LIBSOCKET) $(LTLIBDL) \
+ 	$(LTLIBPTHREAD)
+ 
diff --git a/contrib/src/gnutls/rules.mak b/contrib/src/gnutls/rules.mak
index d2db2a0..d2480a0 100644
--- a/contrib/src/gnutls/rules.mak
+++ b/contrib/src/gnutls/rules.mak
@@ -18,6 +18,9 @@ gnutls: gnutls-$(GNUTLS_VERSION).tar.bz2 .sum-gnutls
 ifdef HAVE_WIN32
 	$(APPLY) $(SRC)/gnutls/gnutls-win32.patch
 endif
+ifdef HAVE_ANDROID
+	$(APPLY) $(SRC)/gnutls/no-gl.patch
+endif
 	$(APPLY) $(SRC)/gnutls/gnutls-no-egd.patch
 	$(MOVE)
 
@@ -50,6 +53,7 @@ DEPS_gnutls = nettle $(DEPS_nettle)
 endif
 
 .gnutls: gnutls
+	$(RECONF)
 	cd $< && $(HOSTVARS) ./configure $(GNUTLS_CONF)
 	cd $</lib && $(MAKE) install
 	touch $@
-- 
1.7.7.3


From funman at videolan.org  Tue Dec 27 03:30:25 2011
From: funman at videolan.org (=?UTF-8?q?Rafa=C3=ABl=20Carr=C3=A9?=)
Date: Mon, 26 Dec 2011 21:30:25 -0500
Subject: [vlc-devel] [PATCH 4/5] contrib: run autoreconf -ivf for some
	packages
In-Reply-To: <1324953026-2767-1-git-send-email-funman@videolan.org>
References: <1324953026-2767-1-git-send-email-funman@videolan.org>
Message-ID: <1324953026-2767-5-git-send-email-funman@videolan.org>

fix detection of android's NDK toolchain (arm-linux-androideabi-)
---
 contrib/src/gmp/rules.mak      |    1 +
 contrib/src/jpeg/rules.mak     |    1 +
 contrib/src/libmpeg2/rules.mak |    1 +
 contrib/src/orc/rules.mak      |    2 +-
 contrib/src/tiff/rules.mak     |    1 +
 5 files changed, 5 insertions(+), 1 deletions(-)

diff --git a/contrib/src/gmp/rules.mak b/contrib/src/gmp/rules.mak
index 89ad5fc..27e7e1b 100644
--- a/contrib/src/gmp/rules.mak
+++ b/contrib/src/gmp/rules.mak
@@ -16,6 +16,7 @@ gmp: gmp-$(GMP_VERSION).tar.bz2 .sum-gmp
 	$(MOVE)
 
 .gmp: gmp
+	$(RECONF)
 	cd $< && $(HOSTVARS) ./configure $(HOSTCONF)
 	cd $< && $(MAKE) install
 	touch $@
diff --git a/contrib/src/jpeg/rules.mak b/contrib/src/jpeg/rules.mak
index a5d2d5d..7cf3211 100644
--- a/contrib/src/jpeg/rules.mak
+++ b/contrib/src/jpeg/rules.mak
@@ -14,6 +14,7 @@ jpeg: jpegsrc.v$(JPEG_VERSION).tar.gz .sum-jpeg
 	$(MOVE)
 
 .jpeg: jpeg
+	$(RECONF)
 	cd $< && $(HOSTVARS) ./configure $(HOSTCONF)
 	cd $< && $(MAKE) install
 	cd $< && if test -e $(PREFIX)/lib/libjpeg.a; then $(RANLIB) $(PREFIX)/lib/libjpeg.a; fi
diff --git a/contrib/src/libmpeg2/rules.mak b/contrib/src/libmpeg2/rules.mak
index 6fb9ec0..4a8e207 100644
--- a/contrib/src/libmpeg2/rules.mak
+++ b/contrib/src/libmpeg2/rules.mak
@@ -20,6 +20,7 @@ libmpeg2: libmpeg2-$(LIBMPEG2_VERSION).tar.gz .sum-libmpeg2
 	$(MOVE)
 
 .libmpeg2: libmpeg2
+	$(RECONF)
 	cd $< && $(HOSTVARS) ./configure $(HOSTCONF) --without-x --disable-sdl
 	cd $</libmpeg2 && make && make install
 	cd $</include && make && make install
diff --git a/contrib/src/orc/rules.mak b/contrib/src/orc/rules.mak
index c326600..8c385d3 100644
--- a/contrib/src/orc/rules.mak
+++ b/contrib/src/orc/rules.mak
@@ -19,7 +19,7 @@ orc: orc-$(ORC_VERSION).tar.gz .sum-orc
 	$(MOVE)
 
 .orc: orc
-	#$(RECONF)
+	$(RECONF)
 	cd $< && $(HOSTVARS) ./configure $(HOSTCONF)
 	cd $< && $(MAKE) install
 	touch $@
diff --git a/contrib/src/tiff/rules.mak b/contrib/src/tiff/rules.mak
index aba48c1..aa411ba 100644
--- a/contrib/src/tiff/rules.mak
+++ b/contrib/src/tiff/rules.mak
@@ -13,6 +13,7 @@ tiff: tiff-$(TIFF_VERSION).tar.gz .sum-tiff
 	$(MOVE)
 
 .tiff: tiff
+	$(RECONF)
 	cd $< && $(HOSTVARS) ./configure $(HOSTCONF) \
 		--disable-jpeg \
 		--disable-zlib \
-- 
1.7.7.3


From funman at videolan.org  Tue Dec 27 03:30:26 2011
From: funman at videolan.org (=?UTF-8?q?Rafa=C3=ABl=20Carr=C3=A9?=)
Date: Mon, 26 Dec 2011 21:30:26 -0500
Subject: [vlc-devel] [PATCH 5/5] contrib: nettle: fix android build
In-Reply-To: <1324953026-2767-1-git-send-email-funman@videolan.org>
References: <1324953026-2767-1-git-send-email-funman@videolan.org>
Message-ID: <1324953026-2767-6-git-send-email-funman@videolan.org>

run autoreconf -ivf
also run libtoolize -ivf so config.sub is updated properly
(this package doesn't use automake)
---
 contrib/src/nettle/rules.mak |    1 +
 1 files changed, 1 insertions(+), 0 deletions(-)

diff --git a/contrib/src/nettle/rules.mak b/contrib/src/nettle/rules.mak
index d02ae14..f057769 100644
--- a/contrib/src/nettle/rules.mak
+++ b/contrib/src/nettle/rules.mak
@@ -17,6 +17,7 @@ nettle: nettle-$(NETTLE_VERSION).tar.gz .sum-nettle
 DEPS_nettle = gmp $(DEPS_gmp)
 
 .nettle: nettle
+	$(RECONF) && libtoolize -ivf
 	cd $< && $(HOSTVARS) ./configure $(HOSTCONF)
 	cd $< && $(MAKE) install
 	touch $@
-- 
1.7.7.3


From funman at videolan.org  Tue Dec 27 03:30:21 2011
From: funman at videolan.org (=?UTF-8?q?Rafa=C3=ABl=20Carr=C3=A9?=)
Date: Mon, 26 Dec 2011 21:30:21 -0500
Subject: [vlc-devel] android contribs
Message-ID: <1324953026-2767-1-git-send-email-funman@videolan.org>

This patch series fix build of some packages on android

Packages I disabled in my build:

sdl SDL_image
    sdl needs autoreconf refresh, which fails with recent autoheader (AC_DEFINE misused)
    we don't need sdl anyway.
    Not sure why we need SDL_image anyway

fontconfig ass freetyp2 fribidi zvbi kate
    Not needed as long as we don't support subtitles

caca
    doesn't make much sense and would probably not run

gettext
    Should be fixed, same problem than gnutls: GNU replacement functions/headers fail to work correctly; the originals probably work out of the box.

mpcdec
    relies on features absent from google's toolchain
    musepack/mpcenc/mpcenc.h:47:23: error: langinfo.h: No such file or directory
    musepack/mpcenc/mpcenc.h:54:27: error: fpu_control.h: No such file or directory


From rafael.carre at gmail.com  Tue Dec 27 03:39:43 2011
From: rafael.carre at gmail.com (=?ISO-8859-1?Q?Rafa=EBl_Carr=E9?=)
Date: Mon, 26 Dec 2011 21:39:43 -0500
Subject: [vlc-devel] [PATCH] Make the Mac DMG prettier.
In-Reply-To: <1324951651-31383-1-git-send-email-brendonjustin@gmail.com>
References: <1324951651-31383-1-git-send-email-brendonjustin@gmail.com>
Message-ID: <4EF92FEF.70009@gmail.com>

Hello,

Le 2011-12-26 21:07, Brendon Justin a ?crit :
> ---
>   extras/package/macosx/dmg_setup.scpt |   25 +++++++++++++++++++++++++
>   extras/package/macosx/package.mak    |    5 ++++-
>   2 files changed, 29 insertions(+), 1 deletions(-)
>   create mode 100644 extras/package/macosx/dmg_setup.scpt
>
> diff --git a/extras/package/macosx/dmg_setup.scpt b/extras/package/macosx/dmg_setup.scpt
> new file mode 100644
> index 0000000..f50ab36
> --- /dev/null
> +++ b/extras/package/macosx/dmg_setup.scpt
> @@ -0,0 +1,25 @@
> +on run argv
> +    tell application "Finder"
> +        tell disk (item 1 of argv)
> +            open
> +            set current view of container window to icon view
> +            set toolbar visible of container window to false
> +            set statusbar visible of container window to false
> +            set the bounds of container window to {300, 100, 750, 500}
> +            set theViewOptions to the icon view options of container window
> +            set arrangement of theViewOptions to not arranged
> +            set icon size of theViewOptions to 104
> +            # Don't set a background image, for now
> +            # set background picture of theViewOptions to file ".background/background.png"
> +            set position of item "VLC.app" of container window to {110, 100}
> +            set position of item "Applications" of container window to {335, 100}
> +            set position of item "Read Me.rtf" of container window to {110, 275}
> +            set position of item "Goodies" of container window to {335, 275}
> +	    # Force saving changes to the disk by closing and opening the window
> +            close
> +            open
> +            update without registering applications
> +            delay 5
> +        end tell
> +    end tell
> +end run
> diff --git a/extras/package/macosx/package.mak b/extras/package/macosx/package.mak
> index ab4d096..1c8b2f6 100644
> --- a/extras/package/macosx/package.mak
> +++ b/extras/package/macosx/package.mak
> @@ -67,7 +67,10 @@ package-macosx: VLC-release.app
>   	cp $(srcdir)/extras/package/macosx/Resources/about_bg.png $(top_builddir)/vlc-$(VERSION)/.background/background.png
>   	$(LN_S) -f /Applications $(top_builddir)/vlc-$(VERSION)/
>   	rm -f "$(top_builddir)/vlc-$(VERSION)-rw.dmg"
> -	hdiutil create -verbose -srcfolder "$(top_builddir)/vlc-$(VERSION)" "$(top_builddir)/vlc-$(VERSION)-rw.dmg" -scrub
> +	hdiutil create -verbose -srcfolder "$(top_builddir)/vlc-$(VERSION)" "$(top_builddir)/vlc-$(VERSION)-rw.dmg" -scrub -format UDRW
> +	hdiutil attach -readwrite -noverify -noautoopen "vlc-$(VERSION)-rw.dmg"

I think it'd be better to mount the dmg to a mount point in the current 
directory instead of /Volumes

> +	osascript $(srcdir)/extras/package/macosx/dmg_setup.scpt vlc-$(VERSION)

$(srcdir) should be in quotes "" in case it has spaces in it.

> +	hdiutil detach /Volumes/vlc-$(VERSION)
>   # Make sure the image is not writable
>   # Note: We can't directly create a read only dmg as we do the bless stuff
>   	rm -f "$(top_builddir)/vlc-$(VERSION).dmg"



From fkuehne.videolan at googlemail.com  Tue Dec 27 04:07:58 2011
From: fkuehne.videolan at googlemail.com (=?iso-8859-1?Q?Felix_Paul_K=FChne?=)
Date: Tue, 27 Dec 2011 04:07:58 +0100
Subject: [vlc-devel] [PATCH] Make the Mac DMG prettier.
In-Reply-To: <4EF92FEF.70009@gmail.com>
References: <1324951651-31383-1-git-send-email-brendonjustin@gmail.com>
	<4EF92FEF.70009@gmail.com>
Message-ID: <156282EB-429F-4071-B51F-3CCBAF5DDC5E@gmail.com>

Hello,

Am 27.12.2011 um 03:39 schrieb Rafa?l Carr?:

> Le 2011-12-26 21:07, Brendon Justin a ?crit :
>> 
[cut]
>>  	rm -f "$(top_builddir)/vlc-$(VERSION)-rw.dmg"
>> -	hdiutil create -verbose -srcfolder "$(top_builddir)/vlc-$(VERSION)" "$(top_builddir)/vlc-$(VERSION)-rw.dmg" -scrub
>> +	hdiutil create -verbose -srcfolder "$(top_builddir)/vlc-$(VERSION)" "$(top_builddir)/vlc-$(VERSION)-rw.dmg" -scrub -format UDRW
>> +	hdiutil attach -readwrite -noverify -noautoopen "vlc-$(VERSION)-rw.dmg"
> 
> I think it'd be better to mount the dmg to a mount point in the current directory instead of /Volumes

I tend to disagree here for the sake of simplicity. The mounted dmg is only present for a few seconds while this script is running, so there is a high chance of not conflicting with anything running on the dev's box.

Additionally, the needed option "-mountroot" (for the attach command) only understands path up to a total length of 90  characters, a limit which was only recently moved to 1024 characters?


There is another problem with Brendon's script which he mentioned in the related trac ticket: it requires the accessibility settings to be enabled on the box creating the dmg to let the AppleScript do the stuff he wants it to do.
At present, this script doesn't check for the enabled accessibility stuff and will utterly fail on Macs with the default configuration.

Except for that, it's a nice contribution :-)

Best regards,

Felix



From funman at videolan.org  Tue Dec 27 05:04:14 2011
From: funman at videolan.org (=?windows-1252?Q?Rafa=EBl_Carr=E9?=)
Date: Mon, 26 Dec 2011 23:04:14 -0500
Subject: [vlc-devel] [PATCH] Make the Mac DMG prettier.
In-Reply-To: <156282EB-429F-4071-B51F-3CCBAF5DDC5E@gmail.com>
References: <1324951651-31383-1-git-send-email-brendonjustin@gmail.com>
	<4EF92FEF.70009@gmail.com>
	<156282EB-429F-4071-B51F-3CCBAF5DDC5E@gmail.com>
Message-ID: <4EF943BE.8000907@videolan.org>

Le 11-12-26 22:07, Felix Paul K?hne a ?crit :
> Hello,
>
> Am 27.12.2011 um 03:39 schrieb Rafa?l Carr?:
>
>> Le 2011-12-26 21:07, Brendon Justin a ?crit :
>>>
> [cut]
>>>   	rm -f "$(top_builddir)/vlc-$(VERSION)-rw.dmg"
>>> -	hdiutil create -verbose -srcfolder "$(top_builddir)/vlc-$(VERSION)" "$(top_builddir)/vlc-$(VERSION)-rw.dmg" -scrub
>>> +	hdiutil create -verbose -srcfolder "$(top_builddir)/vlc-$(VERSION)" "$(top_builddir)/vlc-$(VERSION)-rw.dmg" -scrub -format UDRW
>>> +	hdiutil attach -readwrite -noverify -noautoopen "vlc-$(VERSION)-rw.dmg"
>>
>> I think it'd be better to mount the dmg to a mount point in the current directory instead of /Volumes
>
> I tend to disagree here for the sake of simplicity.

> The mounted dmg is only present for a few seconds while this script is running, so there is a high chance of not conflicting with anything running on the dev's box.

Using a path relative to our build directory, there is infinitely less 
chances of a conflict.

Also no knowledge of MacOSX internals is required since the mount point 
is explicited on the command line.

> Additionally, the needed option "-mountroot" (for the attach command) only understands path up to a total length of 90  characters, a limit which was only recently moved to 1024 characters?

Relative paths work: "a" is 1 character, it fits the lowest limit 
easily; but we can use something more explicit like "./mount" or whatever.

Did I convince you ?

> There is another problem with Brendon's script which he mentioned in the related trac ticket: it requires the accessibility settings to be enabled on the box creating the dmg to let the AppleScript do the stuff he wants it to do.
> At present, this script doesn't check for the enabled accessibility stuff and will utterly fail on Macs with the default configuration.

Will it fail with bad side effects?

> Except for that, it's a nice contribution :-)
>
> Best regards,
>
> Felix


From fkuehne.videolan at googlemail.com  Tue Dec 27 10:16:22 2011
From: fkuehne.videolan at googlemail.com (=?iso-8859-1?Q?Felix_Paul_K=FChne?=)
Date: Tue, 27 Dec 2011 10:16:22 +0100
Subject: [vlc-devel] [PATCH] Make the Mac DMG prettier.
In-Reply-To: <4EF943BE.8000907@videolan.org>
References: <1324951651-31383-1-git-send-email-brendonjustin@gmail.com>
	<4EF92FEF.70009@gmail.com>
	<156282EB-429F-4071-B51F-3CCBAF5DDC5E@gmail.com>
	<4EF943BE.8000907@videolan.org>
Message-ID: <22D3FD39-BC12-4E68-9595-F780F375FC01@gmail.com>


Am 27.12.2011 um 05:04 schrieb Rafa?l Carr?:

>> The mounted dmg is only present for a few seconds while this script is running, so there is a high chance of not conflicting with anything running on the dev's box.
> 
> Using a path relative to our build directory, there is infinitely less chances of a conflict.
True, but I think this would solve a non-existing problem. Since 2004, I never heard a single person complaining about this...

> Also no knowledge of MacOSX internals is required since the mount point is explicited on the command line.
So, you guess that people try to compile who don't know /Volumes? Well, could be possible, but chances aren't that high I guess..

> 
>> Additionally, the needed option "-mountroot" (for the attach command) only understands path up to a total length of 90  characters, a limit which was only recently moved to 1024 characters?
> 
> Relative paths work: "a" is 1 character, it fits the lowest limit easily; but we can use something more explicit like "./mount" or whatever.
> 
> Did I convince you ?
Not really, since you need to pass an absolute path ;-)

More seriously, it's because of this:
The make script passes the dmg volume's name as an argument to osascript blabla. Within the AppleScript, 3rd line, there is "tell disk" which resolves to "tell disk "VLC-1.2.x"". I doubt that AS can resolve in a custom location as a volume correctly, but feel free to prove me wrong. Note that Brendon uses the Finder's AS suite here, which behaves the same way and can do the exact same things a user can do, just faster.
Let's push this some more: on the Mac, there's always exactly one volume with the exact same name. Let's say you open 2 VLC dmgs of the same version, both called "VLC-1.2.x". In the Finder, you'll see 2 volumes with the same name. Within /Volumes, you'll see one of them with a " 1" prefix. If you want to manipulate the you mounted later on, it'll be the one with the " 1" suffix. You can't guess this from any public API. You need to know that.
So, whatever you do with the custom mount point, the Finder will not be your friend here, but it's the only guy in this game who can do what you want: make the design more pretty..


>> There is another problem with Brendon's script which he mentioned in the related trac ticket: it requires the accessibility settings to be enabled on the box creating the dmg to let the AppleScript do the stuff he wants it to do.
>> At present, this script doesn't check for the enabled accessibility stuff and will utterly fail on Macs with the default configuration.
> 
> Will it fail with bad side effects?
Sorry, I withdraw that statement. I'll check another time on a clean account, but to my surprise, it works. Must have been the sleep deprivation last night...


Best regards,

Felix



From f.grotkowski at gmail.com  Tue Dec 27 11:31:54 2011
From: f.grotkowski at gmail.com (Filip Grotkowski)
Date: Tue, 27 Dec 2011 11:31:54 +0100
Subject: [vlc-devel] Stereoscopic 3D support -- how TV recognize 3D format
Message-ID: <CACMC7H7uC7GXJTUhM6Wu63AZ+tnRZQ0wvM2W1f+LGuGeEgRpYQ@mail.gmail.com>

Hello,

I've got a question connected with a great Stereoscopic/3D support in
VLC Media Player
described in http://wiki.videolan.org/SoC_2011/Stereoscopic_Video
I'm interested is there implemented negotiation with TV about the type
of 3D frames we are
sending (Side-by-side, Top-bottom)? I know I can switch this using
remote control, but I'd
like to do this automatically.
Also, is there any other video player that supports this feature?
If not, what actions should I take to tell 3DTV that I'm sending --
for example -- Stereoscopics
3D Side-by-Side frames?

Thank you for your help.

Kind regards,
Filip Grotkowski


From kaarlo.raiha at gmail.com  Tue Dec 27 12:28:45 2011
From: kaarlo.raiha at gmail.com (=?ISO-8859-1?B?S2FhcmxvIFLkaWjk?=)
Date: Tue, 27 Dec 2011 13:28:45 +0200
Subject: [vlc-devel] Video XP bug on VLC 1.2
In-Reply-To: <201112082151.16000.remi@remlab.net>
References: <20111208012751.GA23860@videolan.org>
	<20111208192314.GA10151@elivagar.org>
	<20111208194639.GA29098@videolan.org>
	<201112082151.16000.remi@remlab.net>
Message-ID: <CAMKrNtw0SiO265b2M+5+t2xm0VEHmEvHxFzDEkfkaJf6dGBSaA@mail.gmail.com>

2011/12/8 R?mi Denis-Courmont <remi at remlab.net>

> Le jeudi 8 d?cembre 2011 21:46:39 Jean-Baptiste Kempf, vous avez ?crit :
> > On Thu, Dec 08, 2011 at 08:23:14PM +0100, Laurent Aimar wrote :
> > >  Btw, I haven't updated ffmpeg in some time, but on my PC on linux, I
> > >  also
> > >
> > > experience deadlock (on slightly corrupted streams like H264 TS coming
> > > from DVB-T) with ffmpeg-mt.
> > >
> > >  Am I the only one?
> >
> > I haven't seen that, so far. Samples?
> >
> > Laurent, do you think I should add an option to deactivate ffmpeg-mt in
> > Simple Preferences?
>
> If it's so bad, it should be an option to ENABLE it.
>

It seems that the issue is fixable now with patch
https://github.com/jeeb/mpc-hc/commit/c0e19aea94ccf38f2c7fe5ef80cc333b4d958cf8


>
> --
> R?mi Denis-Courmont
> http://www.remlab.net/
> http://fi.linkedin.com/in/remidenis
> _______________________________________________
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
> http://mailman.videolan.org/listinfo/vlc-devel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111227/19430298/attachment.html>

From jpsaman at videolan.org  Tue Dec 27 13:40:51 2011
From: jpsaman at videolan.org (Jean-Paul Saman)
Date: Tue, 27 Dec 2011 13:40:51 +0100
Subject: [vlc-devel] FOSDEM 2012
In-Reply-To: <CAK-C75R+AvXJ5XRcGHsYDJa1F2A5sGTyea72oTKvDtfxyXe-7Q@mail.gmail.com>
References: <CAK-C75QBGcaC-GNZoWSHq_m0TO_DcDsd4xFeFvjj3G-ZNWe_Ug@mail.gmail.com>
	<CAK-C75R+AvXJ5XRcGHsYDJa1F2A5sGTyea72oTKvDtfxyXe-7Q@mail.gmail.com>
Message-ID: <CAK9US3Ac0uLfkS1+rFJO=NUExbuEGzsjv=4pyj5jZkyXmL49Xg@mail.gmail.com>

Hugo,

On Mon, Dec 26, 2011 at 3:31 PM, Hugo Beauz?e-Luyssen <hugo at videolan.org> wrote:
> On Wed, Dec 14, 2011 at 11:43 AM, Hugo Beauz?e-Luyssen
> <hugo at videolan.org> wrote:
>
> Ping,
>
> Please respond by the end of December or I won't be able to book the
> hotel for you.
>
> Regards,
>
> --
> Hugo Beauz?e-Luyssen

You might book for me friday night and saturday night.

The friday night is not certain yet and I'll explain why:

In that period end-januari beginning of februari I have my 'brown
belt' exam for karate. I have already indicated that the weekend of 3
and 4 februari is spoken for, but a date for the exam has not been set
yet. So there is still a chance it will collide with FOSDEM, in that
case I'll be able to attend on Saturday night and sunday but not
sooner.

Sorry for being a PITA planning wise.

Kind regards,

Jean-Paul Saman


From rafael.carre at gmail.com  Tue Dec 27 16:39:58 2011
From: rafael.carre at gmail.com (=?ISO-8859-1?Q?Rafa=EBl_Carr=E9?=)
Date: Tue, 27 Dec 2011 10:39:58 -0500
Subject: [vlc-devel] Video XP bug on VLC 1.2
In-Reply-To: <CAMKrNtw0SiO265b2M+5+t2xm0VEHmEvHxFzDEkfkaJf6dGBSaA@mail.gmail.com>
References: <20111208012751.GA23860@videolan.org>
	<20111208192314.GA10151@elivagar.org>
	<20111208194639.GA29098@videolan.org>
	<201112082151.16000.remi@remlab.net>
	<CAMKrNtw0SiO265b2M+5+t2xm0VEHmEvHxFzDEkfkaJf6dGBSaA@mail.gmail.com>
Message-ID: <4EF9E6CE.4060504@gmail.com>

Le 2011-12-27 06:28, Kaarlo R?ih? a ?crit :

>      > Laurent, do you think I should add an option to deactivate
>     ffmpeg-mt in
>      > Simple Preferences?
>
>     If it's so bad, it should be an option to ENABLE it.
>
>
> It seems that the issue is fixable now with patch
> https://github.com/jeeb/mpc-hc/commit/c0e19aea94ccf38f2c7fe5ef80cc333b4d958cf8

The issue has been fixed already upstream by Sergey:

http://git.videolan.org/?p=ffmpeg.git;a=history;f=libavcodec/w32pthreads.h;h=70b84cf2e17e6482a000ce24a586a5aa028b93d8;hb=HEAD

I guess mpc-hc just updated their (source) copy of FFmpeg, our binary 
used by the buildbots should be updated already.


From rafael.carre at gmail.com  Tue Dec 27 17:00:00 2011
From: rafael.carre at gmail.com (=?windows-1252?Q?Rafa=EBl_Carr=E9?=)
Date: Tue, 27 Dec 2011 11:00:00 -0500
Subject: [vlc-devel] [PATCH] Make the Mac DMG prettier.
In-Reply-To: <22D3FD39-BC12-4E68-9595-F780F375FC01@gmail.com>
References: <1324951651-31383-1-git-send-email-brendonjustin@gmail.com>
	<4EF92FEF.70009@gmail.com>
	<156282EB-429F-4071-B51F-3CCBAF5DDC5E@gmail.com>
	<4EF943BE.8000907@videolan.org>
	<22D3FD39-BC12-4E68-9595-F780F375FC01@gmail.com>
Message-ID: <4EF9EB80.90500@gmail.com>

Le 2011-12-27 04:16, Felix Paul K?hne a ?crit :
>
> Am 27.12.2011 um 05:04 schrieb Rafa?l Carr?:
>
>>> The mounted dmg is only present for a few seconds while this script is running, so there is a high chance of not conflicting with anything running on the dev's box.
>>
>> Using a path relative to our build directory, there is infinitely less chances of a conflict.
> True, but I think this would solve a non-existing problem. Since 2004, I never heard a single person complaining about this...
>
>> Also no knowledge of MacOSX internals is required since the mount point is explicited on the command line.
> So, you guess that people try to compile who don't know /Volumes? Well, could be possible, but chances aren't that high I guess..

It happened to me, I built VLC without knowing that mounts went in /Volumes

mount command needs a mountpoint since the early days of unix:
http://www.unix.com/man-page/FreeBSD/8/mount/

>>> Additionally, the needed option "-mountroot" (for the attach command) only understands path up to a total length of 90  characters, a limit which was only recently moved to 1024 characters?
>>
>> Relative paths work: "a" is 1 character, it fits the lowest limit easily; but we can use something more explicit like "./mount" or whatever.
>>
>> Did I convince you ?
> Not really, since you need to pass an absolute path ;-)

I tested that relative paths worked before sending this mail, what 
doesn't work for you?

> More seriously, it's because of this:
> The make script passes the dmg volume's name as an argument to osascript blabla. Within the AppleScript, 3rd line, there is "tell disk" which resolves to "tell disk "VLC-1.2.x"". I doubt that AS can resolve in a custom location as a volume correctly, but feel free to prove me wrong. Note that Brendon uses the Finder's AS suite here, which behaves the same way and can do the exact same things a user can do, just faster.
> Let's push this some more: on the Mac, there's always exactly one volume with the exact same name. Let's say you open 2 VLC dmgs of the same version, both called "VLC-1.2.x". In the Finder, you'll see 2 volumes with the same name. Within /Volumes, you'll see one of them with a " 1" prefix.

I'll check

>If you want to manipulate the you mounted later on, it'll be the one 
with the " 1" suffix. You can't guess this from any public API. You need 
to know that.

That's why specifying the mount point feels better.

> So, whatever you do with the custom mount point, the Finder will not be your friend here, but it's the only guy in this game who can do what you want: make the design more pretty..



>>> There is another problem with Brendon's script which he mentioned in the related trac ticket: it requires the accessibility settings to be enabled on the box creating the dmg to let the AppleScript do the stuff he wants it to do.
>>> At present, this script doesn't check for the enabled accessibility stuff and will utterly fail on Macs with the default configuration.
>>
>> Will it fail with bad side effects?
> Sorry, I withdraw that statement. I'll check another time on a clean account, but to my surprise, it works. Must have been the sleep deprivation last night...

Ok, let's see


From remi at remlab.net  Tue Dec 27 17:58:04 2011
From: remi at remlab.net (=?iso-8859-1?q?R=E9mi?= Denis-Courmont)
Date: Tue, 27 Dec 2011 17:58:04 +0100
Subject: [vlc-devel] [PATCH 2/5] contrib: lua: fix android build
In-Reply-To: <1324953026-2767-3-git-send-email-funman@videolan.org>
References: <1324953026-2767-1-git-send-email-funman@videolan.org>
	<1324953026-2767-3-git-send-email-funman@videolan.org>
Message-ID: <201112271758.04520.remi@remlab.net>

   Hello,

On Tuesday 27 December 2011, Rafa?l Carr? wrote:
> android's struct lconv is empty
> according to the headers it's not supported at all, and defined only to
> make libstdc++ build happy

This is going to be a problem on all platforms, not just Android. localeconv() 
is not thread-safe.

-- 
R?mi Denis-Courmont
http://www.remlab.info


From remi at remlab.net  Tue Dec 27 17:58:58 2011
From: remi at remlab.net (=?iso-8859-1?q?R=E9mi?= Denis-Courmont)
Date: Tue, 27 Dec 2011 17:58:58 +0100
Subject: [vlc-devel] [PATCH 4/5] contrib: run autoreconf -ivf for some
	packages
In-Reply-To: <1324953026-2767-5-git-send-email-funman@videolan.org>
References: <1324953026-2767-1-git-send-email-funman@videolan.org>
	<1324953026-2767-5-git-send-email-funman@videolan.org>
Message-ID: <201112271758.58636.remi@remlab.net>

   Hello,

Updating config.guess and config.sub would be orders of magnitude faster and 
safer (no weird autofoo versions conflicts).

On Tuesday 27 December 2011, Rafa?l Carr? wrote:
> fix detection of android's NDK toolchain (arm-linux-androideabi-)
> ---
>  contrib/src/gmp/rules.mak      |    1 +
>  contrib/src/jpeg/rules.mak     |    1 +
>  contrib/src/libmpeg2/rules.mak |    1 +
>  contrib/src/orc/rules.mak      |    2 +-
>  contrib/src/tiff/rules.mak     |    1 +
>  5 files changed, 5 insertions(+), 1 deletions(-)
> 
> diff --git a/contrib/src/gmp/rules.mak b/contrib/src/gmp/rules.mak
> index 89ad5fc..27e7e1b 100644
> --- a/contrib/src/gmp/rules.mak
> +++ b/contrib/src/gmp/rules.mak
> @@ -16,6 +16,7 @@ gmp: gmp-$(GMP_VERSION).tar.bz2 .sum-gmp
>  	$(MOVE)
> 
>  .gmp: gmp
> +	$(RECONF)
>  	cd $< && $(HOSTVARS) ./configure $(HOSTCONF)
>  	cd $< && $(MAKE) install
>  	touch $@
> diff --git a/contrib/src/jpeg/rules.mak b/contrib/src/jpeg/rules.mak
> index a5d2d5d..7cf3211 100644
> --- a/contrib/src/jpeg/rules.mak
> +++ b/contrib/src/jpeg/rules.mak
> @@ -14,6 +14,7 @@ jpeg: jpegsrc.v$(JPEG_VERSION).tar.gz .sum-jpeg
>  	$(MOVE)
> 
>  .jpeg: jpeg
> +	$(RECONF)
>  	cd $< && $(HOSTVARS) ./configure $(HOSTCONF)
>  	cd $< && $(MAKE) install
>  	cd $< && if test -e $(PREFIX)/lib/libjpeg.a; then $(RANLIB)
> $(PREFIX)/lib/libjpeg.a; fi diff --git a/contrib/src/libmpeg2/rules.mak
> b/contrib/src/libmpeg2/rules.mak index 6fb9ec0..4a8e207 100644
> --- a/contrib/src/libmpeg2/rules.mak
> +++ b/contrib/src/libmpeg2/rules.mak
> @@ -20,6 +20,7 @@ libmpeg2: libmpeg2-$(LIBMPEG2_VERSION).tar.gz
> .sum-libmpeg2 $(MOVE)
> 
>  .libmpeg2: libmpeg2
> +	$(RECONF)
>  	cd $< && $(HOSTVARS) ./configure $(HOSTCONF) --without-x --disable-sdl
>  	cd $</libmpeg2 && make && make install
>  	cd $</include && make && make install
> diff --git a/contrib/src/orc/rules.mak b/contrib/src/orc/rules.mak
> index c326600..8c385d3 100644
> --- a/contrib/src/orc/rules.mak
> +++ b/contrib/src/orc/rules.mak
> @@ -19,7 +19,7 @@ orc: orc-$(ORC_VERSION).tar.gz .sum-orc
>  	$(MOVE)
> 
>  .orc: orc
> -	#$(RECONF)
> +	$(RECONF)
>  	cd $< && $(HOSTVARS) ./configure $(HOSTCONF)
>  	cd $< && $(MAKE) install
>  	touch $@
> diff --git a/contrib/src/tiff/rules.mak b/contrib/src/tiff/rules.mak
> index aba48c1..aa411ba 100644
> --- a/contrib/src/tiff/rules.mak
> +++ b/contrib/src/tiff/rules.mak
> @@ -13,6 +13,7 @@ tiff: tiff-$(TIFF_VERSION).tar.gz .sum-tiff
>  	$(MOVE)
> 
>  .tiff: tiff
> +	$(RECONF)
>  	cd $< && $(HOSTVARS) ./configure $(HOSTCONF) \
>  		--disable-jpeg \
>  		--disable-zlib \
-- 
R?mi Denis-Courmont
http://www.remlab.info


From rafael.carre at gmail.com  Tue Dec 27 18:00:09 2011
From: rafael.carre at gmail.com (=?ISO-8859-1?Q?Rafa=EBl_Carr=E9?=)
Date: Tue, 27 Dec 2011 12:00:09 -0500
Subject: [vlc-devel] [PATCH 4/5] contrib: run autoreconf -ivf for some
 packages
In-Reply-To: <201112271758.58636.remi@remlab.net>
References: <1324953026-2767-1-git-send-email-funman@videolan.org>
	<1324953026-2767-5-git-send-email-funman@videolan.org>
	<201112271758.58636.remi@remlab.net>
Message-ID: <4EF9F999.3040904@gmail.com>

Le 2011-12-27 11:58, R?mi Denis-Courmont a ?crit :
>    Hello,
> 
> Updating config.guess and config.sub would be orders of magnitude faster and 
> safer (no weird autofoo versions conflicts).

Do you suggest we ship them in contrib/src and copy them manually after
extracting?


From remi at remlab.net  Tue Dec 27 18:07:22 2011
From: remi at remlab.net (=?iso-8859-1?q?R=E9mi?= Denis-Courmont)
Date: Tue, 27 Dec 2011 18:07:22 +0100
Subject: [vlc-devel] [PATCH 4/5] contrib: run autoreconf -ivf for some
	packages
In-Reply-To: <4EF9F999.3040904@gmail.com>
References: <1324953026-2767-1-git-send-email-funman@videolan.org>
	<201112271758.58636.remi@remlab.net> <4EF9F999.3040904@gmail.com>
Message-ID: <201112271807.22469.remi@remlab.net>

   Hello,

On Tuesday 27 December 2011, Rafa?l Carr? wrote:
> Le 2011-12-27 11:58, R?mi Denis-Courmont a ?crit :
> >    Hello,
> > 
> > Updating config.guess and config.sub would be orders of magnitude faster
> > and safer (no weird autofoo versions conflicts).
> 
> Do you suggest we ship them in contrib/src and copy them manually after
> extracting?

This is obviously not Android-specific and Debian does something like that 
IIRC. I have not tried myself, so I wouldn't know the dirty little details.

-- 
R?mi Denis-Courmont
http://www.remlab.info


From rafael.carre at gmail.com  Tue Dec 27 18:08:50 2011
From: rafael.carre at gmail.com (=?ISO-8859-1?Q?Rafa=EBl_Carr=E9?=)
Date: Tue, 27 Dec 2011 12:08:50 -0500
Subject: [vlc-devel] [PATCH 4/5] contrib: run autoreconf -ivf for some
 packages
In-Reply-To: <201112271807.22469.remi@remlab.net>
References: <1324953026-2767-1-git-send-email-funman@videolan.org>
	<201112271758.58636.remi@remlab.net> <4EF9F999.3040904@gmail.com>
	<201112271807.22469.remi@remlab.net>
Message-ID: <4EF9FBA2.4000304@gmail.com>

Le 2011-12-27 12:07, R?mi Denis-Courmont a ?crit :
>    Hello,
> 
> On Tuesday 27 December 2011, Rafa?l Carr? wrote:
>> Le 2011-12-27 11:58, R?mi Denis-Courmont a ?crit :
>>>    Hello,
>>>
>>> Updating config.guess and config.sub would be orders of magnitude faster
>>> and safer (no weird autofoo versions conflicts).
>>
>> Do you suggest we ship them in contrib/src and copy them manually after
>> extracting?
> 
> This is obviously not Android-specific and Debian does something like that 
> IIRC. I have not tried myself, so I wouldn't know the dirty little details.

Yeah I added something like that for win64 too, IIRC

xtophe do you have details on how/why Debian does that ?


From funman at videolan.org  Tue Dec 27 18:53:46 2011
From: funman at videolan.org (=?UTF-8?q?Rafa=C3=ABl=20Carr=C3=A9?=)
Date: Tue, 27 Dec 2011 12:53:46 -0500
Subject: [vlc-devel] [PATCH] udev discovery: fix bluray
Message-ID: <1325008426-13513-1-git-send-email-funman@videolan.org>

- the scheme is wrong (bd instead of bluray)
- libbluray needs to access the mount point, not the devnode
we parse /proc/mounts to find where this devnode is mounted

note: the 2 discs I have have no spaces in their label (replaced by _)
I anm not sure if it's legal for discs to have spaces, but for some
other device it appears in /proc/mounts as \040 (octal)
---
 modules/services_discovery/udev.c |   52 +++++++++++++++++++++++++++++++++++-
 1 files changed, 50 insertions(+), 2 deletions(-)

diff --git a/modules/services_discovery/udev.c b/modules/services_discovery/udev.c
index 385a52b..cbaec54 100644
--- a/modules/services_discovery/udev.c
+++ b/modules/services_discovery/udev.c
@@ -29,11 +29,13 @@
 #include <vlc_services_discovery.h>
 #include <vlc_plugin.h>
 #include <vlc_url.h>
+#include <vlc_fs.h>
 #ifdef HAVE_SEARCH_H
 # include <search.h>
 #endif
 #include <poll.h>
 #include <errno.h>
+#include <stdio.h>
 
 static int OpenV4L (vlc_object_t *);
 #ifdef HAVE_ALSA
@@ -540,6 +542,41 @@ int OpenALSA (vlc_object_t *obj)
 
 
 /*** Discs support ***/
+static char *get_mount_point (const char *devnode)
+{
+    char *mnt = NULL;
+    char *line = NULL;
+    size_t len = 0;
+    size_t devlen = strlen(devnode);
+    FILE *f = vlc_fopen("/proc/mounts", "r");
+
+    if (!f)
+        return NULL;
+
+    while (getline(&line, &len, f) != -1) {
+        if (strncmp(line, devnode, devlen))
+            continue;
+
+        /* if device is not followed immediately by a space it's not the one */
+        /* e.g. when looking for "/dev/sda1", "/dev/sda XXX" won't match */
+        char *space = strchr(line, ' ');
+        if (space != &line[devlen])
+            continue;
+
+        space = strchr(space + 1, ' '); /* space points just after the mount dir */
+        if (!space) /* corrupted line ? */
+            break;
+
+        *space = '\0';
+        /* FIXME : spaces (at least) are octal escaped */
+        mnt = strdup(&line[devlen+1]);
+    }
+
+    free(line);
+    fclose(f);
+    return mnt;
+}
+
 static char *disc_get_mrl (struct udev_device *dev)
 {
     const char *val;
@@ -563,7 +600,7 @@ static char *disc_get_mrl (struct udev_device *dev)
 
     val = udev_device_get_property_value (dev, "ID_CDROM_MEDIA_BD");
     if (val && atoi (val))
-        scheme = "bd";
+        scheme = "bluray";
 #ifdef LOL
     val = udev_device_get_property_value (dev, "ID_CDROM_MEDIA_HDDVD");
     if (val && atoi (val))
@@ -576,7 +613,18 @@ static char *disc_get_mrl (struct udev_device *dev)
         return NULL;
 
     val = udev_device_get_devnode (dev);
-    return make_URI (val, scheme);
+    char *uri = NULL;
+    if (!strcmp(scheme, "bluray")) {
+        /* bluray needs the mount point, not the device node */
+        char *mount = get_mount_point(val);
+        if (mount) {
+           uri = make_URI (mount, scheme);
+           free(mount);
+        }
+    } else {
+        uri = make_URI (val, scheme);
+    }
+    return uri;
 }
 
 static char *disc_get_name (struct udev_device *dev)
-- 
1.7.7.3


From remi at remlab.net  Tue Dec 27 19:00:03 2011
From: remi at remlab.net (=?iso-8859-1?q?R=E9mi?= Denis-Courmont)
Date: Tue, 27 Dec 2011 19:00:03 +0100
Subject: [vlc-devel] [PATCH] udev discovery: fix bluray
In-Reply-To: <1325008426-13513-1-git-send-email-funman@videolan.org>
References: <1325008426-13513-1-git-send-email-funman@videolan.org>
Message-ID: <201112271900.03378.remi@remlab.net>

   Hello,

This is never going to work, since udev will not emit an event when the node 
gets mounted.

Fix libbluray.

On Tuesday 27 December 2011, Rafa?l Carr? wrote:
> - the scheme is wrong (bd instead of bluray)
> - libbluray needs to access the mount point, not the devnode
> we parse /proc/mounts to find where this devnode is mounted
> 
> note: the 2 discs I have have no spaces in their label (replaced by _)
> I anm not sure if it's legal for discs to have spaces, but for some
> other device it appears in /proc/mounts as \040 (octal)
> ---
>  modules/services_discovery/udev.c |   52
> +++++++++++++++++++++++++++++++++++- 1 files changed, 50 insertions(+), 2
> deletions(-)
> 
> diff --git a/modules/services_discovery/udev.c
> b/modules/services_discovery/udev.c index 385a52b..cbaec54 100644
> --- a/modules/services_discovery/udev.c
> +++ b/modules/services_discovery/udev.c
> @@ -29,11 +29,13 @@
>  #include <vlc_services_discovery.h>
>  #include <vlc_plugin.h>
>  #include <vlc_url.h>
> +#include <vlc_fs.h>
>  #ifdef HAVE_SEARCH_H
>  # include <search.h>
>  #endif
>  #include <poll.h>
>  #include <errno.h>
> +#include <stdio.h>
> 
>  static int OpenV4L (vlc_object_t *);
>  #ifdef HAVE_ALSA
> @@ -540,6 +542,41 @@ int OpenALSA (vlc_object_t *obj)
> 
> 
>  /*** Discs support ***/
> +static char *get_mount_point (const char *devnode)
> +{
> +    char *mnt = NULL;
> +    char *line = NULL;
> +    size_t len = 0;
> +    size_t devlen = strlen(devnode);
> +    FILE *f = vlc_fopen("/proc/mounts", "r");
> +
> +    if (!f)
> +        return NULL;
> +
> +    while (getline(&line, &len, f) != -1) {
> +        if (strncmp(line, devnode, devlen))
> +            continue;
> +
> +        /* if device is not followed immediately by a space it's not the
> one */ +        /* e.g. when looking for "/dev/sda1", "/dev/sda XXX" won't
> match */ +        char *space = strchr(line, ' ');
> +        if (space != &line[devlen])
> +            continue;
> +
> +        space = strchr(space + 1, ' '); /* space points just after the
> mount dir */ +        if (!space) /* corrupted line ? */
> +            break;
> +
> +        *space = '\0';
> +        /* FIXME : spaces (at least) are octal escaped */
> +        mnt = strdup(&line[devlen+1]);
> +    }
> +
> +    free(line);
> +    fclose(f);
> +    return mnt;
> +}
> +
>  static char *disc_get_mrl (struct udev_device *dev)
>  {
>      const char *val;
> @@ -563,7 +600,7 @@ static char *disc_get_mrl (struct udev_device *dev)
> 
>      val = udev_device_get_property_value (dev, "ID_CDROM_MEDIA_BD");
>      if (val && atoi (val))
> -        scheme = "bd";
> +        scheme = "bluray";
>  #ifdef LOL
>      val = udev_device_get_property_value (dev, "ID_CDROM_MEDIA_HDDVD");
>      if (val && atoi (val))
> @@ -576,7 +613,18 @@ static char *disc_get_mrl (struct udev_device *dev)
>          return NULL;
> 
>      val = udev_device_get_devnode (dev);
> -    return make_URI (val, scheme);
> +    char *uri = NULL;
> +    if (!strcmp(scheme, "bluray")) {
> +        /* bluray needs the mount point, not the device node */
> +        char *mount = get_mount_point(val);
> +        if (mount) {
> +           uri = make_URI (mount, scheme);
> +           free(mount);
> +        }
> +    } else {
> +        uri = make_URI (val, scheme);
> +    }
> +    return uri;
>  }
> 
>  static char *disc_get_name (struct udev_device *dev)
-- 
R?mi Denis-Courmont
http://www.remlab.info


From rafael.carre at gmail.com  Tue Dec 27 19:04:59 2011
From: rafael.carre at gmail.com (=?ISO-8859-1?Q?Rafa=EBl_Carr=E9?=)
Date: Tue, 27 Dec 2011 13:04:59 -0500
Subject: [vlc-devel] [PATCH] udev discovery: fix bluray
In-Reply-To: <201112271900.03378.remi@remlab.net>
References: <1325008426-13513-1-git-send-email-funman@videolan.org>
	<201112271900.03378.remi@remlab.net>
Message-ID: <4EFA08CB.6080707@gmail.com>

Le 2011-12-27 13:00, R?mi Denis-Courmont a ?crit :
>    Hello,
> 
> This is never going to work

Wrong.

It works if VLC is started while the disc is mounted.

> since udev will not emit an event when the node
> gets mounted.

Yeah. At least it doesn't show an unplayable item in the GUI.

> Fix libbluray.

j-b ?

libbluray mentions "device_path" in its code / API; although it's
explicitely the mount point and not the device node.


From remi at remlab.net  Tue Dec 27 19:12:04 2011
From: remi at remlab.net (=?iso-8859-1?q?R=E9mi?= Denis-Courmont)
Date: Tue, 27 Dec 2011 19:12:04 +0100
Subject: [vlc-devel] [PATCH] udev discovery: fix bluray
In-Reply-To: <4EFA08CB.6080707@gmail.com>
References: <1325008426-13513-1-git-send-email-funman@videolan.org>
	<201112271900.03378.remi@remlab.net> <4EFA08CB.6080707@gmail.com>
Message-ID: <201112271912.04458.remi@remlab.net>

   Hello,

On Tuesday 27 December 2011, Rafa?l Carr? wrote:
> Le 2011-12-27 13:00, R?mi Denis-Courmont a ?crit :
> >    Hello,
> > 
> > This is never going to work
> 
> Wrong.
> 
> It works if VLC is started while the disc is mounted.

This is a *UDEV* plugin for a reason.

-- 
R?mi Denis-Courmont
http://www.remlab.info


From brendonjustin at gmail.com  Tue Dec 27 19:49:49 2011
From: brendonjustin at gmail.com (Brendon Justin)
Date: Tue, 27 Dec 2011 13:49:49 -0500
Subject: [vlc-devel] [PATCH] Make the Mac DMG prettier.
In-Reply-To: <mailman.263364.1325005217.5480.vlc-devel@videolan.org>
References: <mailman.263364.1325005217.5480.vlc-devel@videolan.org>
Message-ID: <9BC88A3E-D2B3-4D8A-B4D6-8A09A067DB5E@gmail.com>

Hi,

On Dec 27, 2011, at 12:00 PM, vlc-devel-request at videolan.org wrote:

>>>> There is another problem with Brendon's script which he mentioned in the related trac ticket: it requires the accessibility settings to be enabled on the box creating the dmg to let the AppleScript do the stuff he wants it to do.
>>>> At present, this script doesn't check for the enabled accessibility stuff and will utterly fail on Macs with the default configuration.
>>> 
>>> Will it fail with bad side effects?
>> Sorry, I withdraw that statement. I'll check another time on a clean account, but to my surprise, it works. Must have been the sleep deprivation last night...
> 
> Ok, let's see

I gave it another try, and it does work without assistive devices enabled.  I only have Lion machines so I can't test on anything earlier.

Thanks,
Brendon Justin
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111227/9791aa7a/attachment.html>

From cheng.sun at ymail.com  Tue Dec 27 19:53:26 2011
From: cheng.sun at ymail.com (Cheng Sun)
Date: Tue, 27 Dec 2011 18:53:26 +0000
Subject: [vlc-devel] [PATCH] Fix visibility of libvlc_printerr (missing
	LIBVLC_API)
Message-ID: <4EFA1426.9090101@ymail.com>

This allows the web plugin to work nicely with Opera 11.60.

-- Cheng
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-Fix-visibility-of-libvlc_printerr-missing-LIBVLC_API.patch
Type: text/x-patch
Size: 1046 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111227/5b89cb88/attachment.bin>

From fkuehne.videolan at googlemail.com  Tue Dec 27 20:07:06 2011
From: fkuehne.videolan at googlemail.com (=?iso-8859-1?Q?Felix_Paul_K=FChne?=)
Date: Tue, 27 Dec 2011 20:07:06 +0100
Subject: [vlc-devel] [PATCH] Make the Mac DMG prettier.
In-Reply-To: <9BC88A3E-D2B3-4D8A-B4D6-8A09A067DB5E@gmail.com>
References: <mailman.263364.1325005217.5480.vlc-devel@videolan.org>
	<9BC88A3E-D2B3-4D8A-B4D6-8A09A067DB5E@gmail.com>
Message-ID: <7534EFAC-701A-4AD8-A05E-A0F3796544F7@gmail.com>


Am 27.12.2011 um 19:49 schrieb Brendon Justin:

> Hi,
> 
> On Dec 27, 2011, at 12:00 PM, vlc-devel-request at videolan.org wrote:
> 
>>>>> There is another problem with Brendon's script which he mentioned in the related trac ticket: it requires the accessibility settings to be enabled on the box creating the dmg to let the AppleScript do the stuff he wants it to do.
>>>>> At present, this script doesn't check for the enabled accessibility stuff and will utterly fail on Macs with the default configuration.
>>>> 
>>>> Will it fail with bad side effects?
>>> Sorry, I withdraw that statement. I'll check another time on a clean account, but to my surprise, it works. Must have been the sleep deprivation last night...
>> 
>> Ok, let's see
> 
> I gave it another try, and it does work without assistive devices enabled.  I only have Lion machines so I can't test on anything earlier.

Same for me, but I'll test with my virtual machines when being home again (in the next few days). I don't see why it shouldn't work on earlier releases though.

Thanks for your contribution!

Felix

From funman at videolan.org  Tue Dec 27 20:10:33 2011
From: funman at videolan.org (=?UTF-8?q?Rafa=C3=ABl=20Carr=C3=A9?=)
Date: Tue, 27 Dec 2011 14:10:33 -0500
Subject: [vlc-devel] [PATCH] Make the Mac DMG prettier.
In-Reply-To: <4EF9EB80.90500@gmail.com>
References: <4EF9EB80.90500@gmail.com>
Message-ID: <1325013033-41255-1-git-send-email-funman@videolan.org>

Modified-by: Rafa?l Carr? <funman at videolan.org>
    Specify mount point explicitely
    Quote $(srcdir)

Signed-off-by: Rafa?l Carr? <funman at videolan.org>
---

The argument to the script is the volume name, not the mount point
Tested on 10.7

 extras/package/macosx/dmg_setup.scpt |   25 +++++++++++++++++++++++++
 extras/package/macosx/package.mak    |    6 +++++-
 2 files changed, 30 insertions(+), 1 deletions(-)
 create mode 100644 extras/package/macosx/dmg_setup.scpt

diff --git a/extras/package/macosx/dmg_setup.scpt b/extras/package/macosx/dmg_setup.scpt
new file mode 100644
index 0000000..f50ab36
--- /dev/null
+++ b/extras/package/macosx/dmg_setup.scpt
@@ -0,0 +1,25 @@
+on run argv
+    tell application "Finder"
+        tell disk (item 1 of argv)
+            open
+            set current view of container window to icon view
+            set toolbar visible of container window to false
+            set statusbar visible of container window to false
+            set the bounds of container window to {300, 100, 750, 500}
+            set theViewOptions to the icon view options of container window
+            set arrangement of theViewOptions to not arranged
+            set icon size of theViewOptions to 104
+            # Don't set a background image, for now
+            # set background picture of theViewOptions to file ".background/background.png"
+            set position of item "VLC.app" of container window to {110, 100}
+            set position of item "Applications" of container window to {335, 100}
+            set position of item "Read Me.rtf" of container window to {110, 275}
+            set position of item "Goodies" of container window to {335, 275}
+	    # Force saving changes to the disk by closing and opening the window
+            close
+            open
+            update without registering applications
+            delay 5
+        end tell
+    end tell
+end run
diff --git a/extras/package/macosx/package.mak b/extras/package/macosx/package.mak
index ab4d096..d861dc4 100644
--- a/extras/package/macosx/package.mak
+++ b/extras/package/macosx/package.mak
@@ -67,7 +67,11 @@ package-macosx: VLC-release.app
 	cp $(srcdir)/extras/package/macosx/Resources/about_bg.png $(top_builddir)/vlc-$(VERSION)/.background/background.png
 	$(LN_S) -f /Applications $(top_builddir)/vlc-$(VERSION)/
 	rm -f "$(top_builddir)/vlc-$(VERSION)-rw.dmg"
-	hdiutil create -verbose -srcfolder "$(top_builddir)/vlc-$(VERSION)" "$(top_builddir)/vlc-$(VERSION)-rw.dmg" -scrub
+	hdiutil create -verbose -srcfolder "$(top_builddir)/vlc-$(VERSION)" "$(top_builddir)/vlc-$(VERSION)-rw.dmg" -scrub -format UDRW
+	mkdir -p ./mount
+	hdiutil attach -readwrite -noverify -noautoopen -mountRoot ./mount "vlc-$(VERSION)-rw.dmg"
+	osascript "$(srcdir)"/extras/package/macosx/dmg_setup.scpt "vlc-$(VERSION)"
+	hdiutil detach ./mount/"vlc-$(VERSION)"
 # Make sure the image is not writable
 # Note: We can't directly create a read only dmg as we do the bless stuff
 	rm -f "$(top_builddir)/vlc-$(VERSION).dmg"
-- 
1.7.4.4


From ssbssa at yahoo.de  Tue Dec 27 20:18:36 2011
From: ssbssa at yahoo.de (Hannes Domani)
Date: Tue, 27 Dec 2011 19:18:36 +0000 (GMT)
Subject: [vlc-devel] [PATCH 1/2] alphabetical order in vlc_keys
Message-ID: <1325013516.51877.YahooMailNeo@web24819.mail.ird.yahoo.com>

Hello

As I was trying out the nighties I noticed that the Global Hotkeys aren't working any more.
So here the first patch to fix this.

Regards
Domani Hannes
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-alphabetical-order-in-vlc_keys.patch
Type: application/octet-stream
Size: 936 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111227/c24c74a3/attachment.obj>

From ssbssa at yahoo.de  Tue Dec 27 20:20:15 2011
From: ssbssa at yahoo.de (Hannes Domani)
Date: Tue, 27 Dec 2011 19:20:15 +0000 (GMT)
Subject: [vlc-devel] [PATCH 2/2] Win32 global hotkeys: vlc_GetActionId()
	needs "key-" prefix
Message-ID: <1325013615.29068.YahooMailNeo@web24806.mail.ird.yahoo.com>

And here is the last.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0002-Win32-global-hotkeys-vlc_GetActionId-needs-key.patch
Type: application/octet-stream
Size: 1286 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111227/ce1d589c/attachment.obj>

From rafael.carre at gmail.com  Tue Dec 27 20:46:16 2011
From: rafael.carre at gmail.com (=?ISO-8859-1?Q?Rafa=EBl_Carr=E9?=)
Date: Tue, 27 Dec 2011 14:46:16 -0500
Subject: [vlc-devel] [PATCH] udev discovery: fix bluray
In-Reply-To: <201112271900.03378.remi@remlab.net>
References: <1325008426-13513-1-git-send-email-funman@videolan.org>
	<201112271900.03378.remi@remlab.net>
Message-ID: <4EFA2088.6010907@gmail.com>

Le 2011-12-27 13:00, R?mi Denis-Courmont a ?crit :
>    Hello,

> Fix libbluray.

Done:
http://mailman.videolan.org/pipermail/libbluray-devel/2011-December/000550.html

>> @@ -563,7 +600,7 @@ static char *disc_get_mrl (struct udev_device *dev)
>>
>>      val = udev_device_get_property_value (dev, "ID_CDROM_MEDIA_BD");
>>      if (val && atoi (val))
>> -        scheme = "bd";
>> +        scheme = "bluray";
>>  #ifdef LOL
>>      val = udev_device_get_property_value (dev, "ID_CDROM_MEDIA_HDDVD");
>>      if (val && atoi (val))

I committed this part


From remi at remlab.net  Tue Dec 27 21:04:08 2011
From: remi at remlab.net (=?iso-8859-1?q?R=E9mi?= Denis-Courmont)
Date: Tue, 27 Dec 2011 21:04:08 +0100
Subject: [vlc-devel] [PATCH 1/2] alphabetical order in vlc_keys
In-Reply-To: <1325013516.51877.YahooMailNeo@web24819.mail.ird.yahoo.com>
References: <1325013516.51877.YahooMailNeo@web24819.mail.ird.yahoo.com>
Message-ID: <201112272104.08699.remi@remlab.net>

   Hello,

Both merged, thanks.

On Tuesday 27 December 2011, Hannes Domani wrote:
> Hello
> 
> As I was trying out the nighties I noticed that the Global Hotkeys aren't
> working any more. So here the first patch to fix this.
> 
> Regards
> Domani Hannes
-- 
R?mi Denis-Courmont
http://www.remlab.info


From remi at remlab.net  Tue Dec 27 21:04:21 2011
From: remi at remlab.net (=?iso-8859-1?q?R=E9mi?= Denis-Courmont)
Date: Tue, 27 Dec 2011 21:04:21 +0100
Subject: [vlc-devel] [PATCH] Fix visibility of libvlc_printerr (missing
	LIBVLC_API)
In-Reply-To: <4EFA1426.9090101@ymail.com>
References: <4EFA1426.9090101@ymail.com>
Message-ID: <201112272104.21309.remi@remlab.net>

   Hello,

Merged, thanks.

On Tuesday 27 December 2011, Cheng Sun wrote:
> This allows the web plugin to work nicely with Opera 11.60.
> 
> -- Cheng
-- 
R?mi Denis-Courmont
http://www.remlab.info


From adanecito at yahoo.com  Tue Dec 27 21:08:18 2011
From: adanecito at yahoo.com (Tony Anecito)
Date: Tue, 27 Dec 2011 12:08:18 -0800 (PST)
Subject: [vlc-devel] [PATCH] udev discovery: fix bluray
In-Reply-To: <4EFA2088.6010907@gmail.com>
Message-ID: <1325016498.81915.YahooMailClassic@web113613.mail.gq1.yahoo.com>

Thanks guys I have been watching these posts waiting for bluray to be supported.
?
So which nightlybuild wil this be in? 1.x, 1.2.x or 1.3.x
?
Thanks,
-Tony

--- On Tue, 12/27/11, Rafa?l Carr? <rafael.carre at gmail.com> wrote:


From: Rafa?l Carr? <rafael.carre at gmail.com>
Subject: Re: [vlc-devel] [PATCH] udev discovery: fix bluray
To: "Mailing list for VLC media player developers" <vlc-devel at videolan.org>
Date: Tuesday, December 27, 2011, 12:46 PM


Le 2011-12-27 13:00, R?mi Denis-Courmont a ?crit :
>? ? Hello,

> Fix libbluray.

Done:
http://mailman.videolan.org/pipermail/libbluray-devel/2011-December/000550.html

>> @@ -563,7 +600,7 @@ static char *disc_get_mrl (struct udev_device *dev)
>>
>>? ? ? val = udev_device_get_property_value (dev, "ID_CDROM_MEDIA_BD");
>>? ? ? if (val && atoi (val))
>> -? ? ? ? scheme = "bd";
>> +? ? ? ? scheme = "bluray";
>>? #ifdef LOL
>>? ? ? val = udev_device_get_property_value (dev, "ID_CDROM_MEDIA_HDDVD");
>>? ? ? if (val && atoi (val))

I committed this part
_______________________________________________
vlc-devel mailing list
To unsubscribe or modify your subscription options:
http://mailman.videolan.org/listinfo/vlc-devel
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111227/9c4f2c3c/attachment.html>

From benoit.tuduri at gmail.com  Tue Dec 27 22:11:05 2011
From: benoit.tuduri at gmail.com (=?ISO-8859-1?Q?beno=EEt_tuduri?=)
Date: Tue, 27 Dec 2011 22:11:05 +0100
Subject: [vlc-devel] service discovery scripts in macosx build
Message-ID: <CALKXa+-B5Nrh9P_4JG1f1JixEutU70tO0WBLUwPz=veAxFhR5A@mail.gmail.com>

Hello,

I?m a bit surprise, in my macosx I can?t found the lua scripts which
available (like freebox.lua in service discovery) in vlc git
repository in a build version. These scripts are voluntary missing ?

Sincerly,

Tuduri Beno?t


From typx at dinauz.org  Tue Dec 27 23:33:30 2011
From: typx at dinauz.org (Denis Charmet)
Date: Tue, 27 Dec 2011 23:33:30 +0100
Subject: [vlc-devel] Race conditions in picture_pool with ffmpeg-mt
Message-ID: <20111227223330.GA12283@dinauz.org>

Hi,


I've seen with the latest releases a lot of assertion failure in picture
pool (in the Release function line 266).

I think that ffmpeg-mt get the same picture in two different threads
causing visual artefacts and then the "assertion failed" when the later
one releases it.
I tested with --ffmpeg-thread 1 and the problem disappears.

Regards,

-- 
TypX
Le mauvais esprit est un art de vivre


From jb at videolan.org  Wed Dec 28 00:03:30 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Wed, 28 Dec 2011 00:03:30 +0100
Subject: [vlc-devel] service discovery scripts in macosx build
In-Reply-To: <CALKXa+-B5Nrh9P_4JG1f1JixEutU70tO0WBLUwPz=veAxFhR5A@mail.gmail.com>
References: <CALKXa+-B5Nrh9P_4JG1f1JixEutU70tO0WBLUwPz=veAxFhR5A@mail.gmail.com>
Message-ID: <20111227230330.GA24995@videolan.org>

On Tue, Dec 27, 2011 at 10:11:05PM +0100, beno?t tuduri wrote :
> I?m a bit surprise, in my macosx I can?t found the lua scripts which
> available (like freebox.lua in service discovery) in vlc git
> repository in a build version. These scripts are voluntary missing ?

No.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Wed Dec 28 00:58:58 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Wed, 28 Dec 2011 00:58:58 +0100
Subject: [vlc-devel] [PATCH 2/2] Fix a crash with corrupted MKV
In-Reply-To: <1324856353-16722-1-git-send-email-typx@dinauz.org>
References: <1324856353-16722-1-git-send-email-typx@dinauz.org>
Message-ID: <20111227235858.GA27470@videolan.org>

On Mon, Dec 26, 2011 at 12:39:13AM +0100, Denis Charmet wrote :
> If the blocksize is corrupted and has a lace, you may have a buffer overflow. Should fix #5658.

Applied.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From ssbssa at yahoo.de  Wed Dec 28 01:05:19 2011
From: ssbssa at yahoo.de (Hannes Domani)
Date: Wed, 28 Dec 2011 00:05:19 +0000 (GMT)
Subject: [vlc-devel] [PATCH 1/2] alphabetical order in vlc_keys
In-Reply-To: <201112272104.08699.remi@remlab.net>
References: <1325013516.51877.YahooMailNeo@web24819.mail.ird.yahoo.com>
	<201112272104.08699.remi@remlab.net>
Message-ID: <1325030719.56053.YahooMailNeo@web24805.mail.ird.yahoo.com>

Hello

> Both merged, thanks.
I think they should be applied to 1.2 as well.

Regards

Domani Hannes



From jb at videolan.org  Wed Dec 28 01:21:31 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Wed, 28 Dec 2011 01:21:31 +0100
Subject: [vlc-devel] [PATCH] bluray: don't crash when file is NULL
In-Reply-To: <1324861434-24174-1-git-send-email-funman@videolan.org>
References: <1324861434-24174-1-git-send-email-funman@videolan.org>
Message-ID: <20111228002131.GA32049@videolan.org>

On Sun, Dec 25, 2011 at 08:03:54PM -0500, Rafa?l Carr? wrote :
> e.g. vlc bluray://0

Applied.


-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Wed Dec 28 01:21:39 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Wed, 28 Dec 2011 01:21:39 +0100
Subject: [vlc-devel] [PATCH] Fix seeking without cues in mkv
In-Reply-To: <1324924171-19629-1-git-send-email-typx@dinauz.org>
References: <1324924171-19629-1-git-send-email-typx@dinauz.org>
Message-ID: <20111228002139.GB32049@videolan.org>

On Mon, Dec 26, 2011 at 07:29:31PM +0100, Denis Charmet wrote :
> the demux generates its own cues while reading and when seeking.
> This should also avoid crashes when seeking without cues.
> 
> Closes #5085 and #5712

Applied.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Wed Dec 28 01:30:09 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Wed, 28 Dec 2011 01:30:09 +0100
Subject: [vlc-devel] [PATCH] Qt4: Add aspect ratio combobox to toolbar
 editor
In-Reply-To: <1324753167-26617-1-git-send-email-edward.c.wang@compdigitec.com>
References: <1324753167-26617-1-git-send-email-edward.c.wang@compdigitec.com>
Message-ID: <20111228003009.GA1105@videolan.org>

On Sat, Dec 24, 2011 at 01:59:27PM -0500, Edward Wang wrote :
> ---

Applied.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Wed Dec 28 01:41:56 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Wed, 28 Dec 2011 01:41:56 +0100
Subject: [vlc-devel] [PATCH 1/2] alphabetical order in vlc_keys
In-Reply-To: <1325030719.56053.YahooMailNeo@web24805.mail.ird.yahoo.com>
References: <1325013516.51877.YahooMailNeo@web24819.mail.ird.yahoo.com>
	<201112272104.08699.remi@remlab.net>
	<1325030719.56053.YahooMailNeo@web24805.mail.ird.yahoo.com>
Message-ID: <20111228004156.GA3999@videolan.org>

On Wed, Dec 28, 2011 at 12:05:19AM +0000, Hannes Domani wrote :
> > Both merged, thanks.
> I think they should be applied to 1.2 as well.

Done.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Wed Dec 28 01:42:12 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Wed, 28 Dec 2011 01:42:12 +0100
Subject: [vlc-devel] [PATCH 2/2] Win32 global hotkeys: vlc_GetActionId()
 needs "key-" prefix
In-Reply-To: <1325013615.29068.YahooMailNeo@web24806.mail.ird.yahoo.com>
References: <1325013615.29068.YahooMailNeo@web24806.mail.ird.yahoo.com>
Message-ID: <20111228004212.GB3999@videolan.org>

On Tue, Dec 27, 2011 at 07:20:15PM +0000, Hannes Domani wrote :
> And here is the last.

Applied both in master and 1.2


-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From edward.c.wang at compdigitec.com  Wed Dec 28 03:30:52 2011
From: edward.c.wang at compdigitec.com (Edward Wang)
Date: Tue, 27 Dec 2011 21:30:52 -0500
Subject: [vlc-devel] [PATCH] Don't keep VLC on top when paused
Message-ID: <1325039452-22490-1-git-send-email-edward.c.wang@compdigitec.com>

Have the "always on top" preference only take effect while a movie is playing. When not playing VLC will act like any other window.
Fixes #4689 if committed.
Thanks,
	Edward Wang

---
 src/video_output/video_output.c |    5 +++++
 1 files changed, 5 insertions(+), 0 deletions(-)

diff --git a/src/video_output/video_output.c b/src/video_output/video_output.c
index a8e4e24..322c9f8 100644
--- a/src/video_output/video_output.c
+++ b/src/video_output/video_output.c
@@ -1193,6 +1193,11 @@ static void ThreadChangePause(vout_thread_t *vout, bool is_paused, mtime_t date)
 {
     assert(!vout->p->pause.is_on || !is_paused);
 
+    /* Only stay on top if not paused */
+    bool is_on_top = var_GetBool( vout, "video-on-top" );
+    vout_SetWindowState(vout->p->display.vd, (!is_paused && is_on_top) ? VOUT_WINDOW_STATE_ABOVE
+                                                       : VOUT_WINDOW_STATE_NORMAL );
+
     if (vout->p->pause.is_on) {
         const mtime_t duration = date - vout->p->pause.date;
 
-- 
1.7.5.4



From jb at videolan.org  Wed Dec 28 03:43:31 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Wed, 28 Dec 2011 03:43:31 +0100
Subject: [vlc-devel] [vlc-commits] Tag 1.2.0-pre3 : VLC media player -
 'Twoflower' - 1.2.0-pre3
In-Reply-To: <20111228022657.31958144A6B@albiero.videolan.org>
References: <20111228022657.31958144A6B@albiero.videolan.org>
Message-ID: <20111228024331.GA27834@videolan.org>

On Wed, Dec 28, 2011 at 03:26:57AM +0100, git at videolan.org wrote :
> VLC media player - 'Twoflower' - 1.2.0-pre3

ftp://ftp.videolan.org/pub/videolan/testing/vlc-1.2.0-pre3/vlc-1.2.0-pre3.tar.xz

Win32 binaries will follow

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From brendonjustin at gmail.com  Wed Dec 28 05:42:39 2011
From: brendonjustin at gmail.com (Brendon Justin)
Date: Tue, 27 Dec 2011 23:42:39 -0500
Subject: [vlc-devel] VLC Development at School
Message-ID: <0D977D15-C207-403F-8AEE-8775DDEF4A1F@gmail.com>

Hi everyone,

I am a student at Rensselaer Polytechnic Institute (RPI) and we have an awesome program for students called RCOS: http://rcos.rpi.edu/ .  Students work on open source software, like they would work on a research project.  They do pay us like undergraduate research students, i.e. some but not a whole lot.  I thought I would contribute to VLC next semester in RCOS, instead of doing my own project.  I want to make sure that you guys know and have no problem with it before I do the RCOS paperwork.  What do you think?

I'm in between semesters until January 24th, so work I do before then may not indicate what I can do during school :-).  While classes are going on, I will have 3-8 hours in a typical week to contribute.

I'm on the IRC channel pretty frequently, as bjustin.

Thanks,
Brendon Justin

p.s. The project that I did last two semesters was a tracker for the shuttles at school, mostly the iPhone app: http://itunes.apple.com/us/app/rpi-shuttles/id429089081?mt=8 but also a bit for the web server it pulls data from.

From rafael.carre at gmail.com  Wed Dec 28 06:00:06 2011
From: rafael.carre at gmail.com (=?windows-1252?Q?Rafa=EBl_Carr=E9?=)
Date: Wed, 28 Dec 2011 00:00:06 -0500
Subject: [vlc-devel] service discovery scripts in macosx build
In-Reply-To: <CALKXa+-B5Nrh9P_4JG1f1JixEutU70tO0WBLUwPz=veAxFhR5A@mail.gmail.com>
References: <CALKXa+-B5Nrh9P_4JG1f1JixEutU70tO0WBLUwPz=veAxFhR5A@mail.gmail.com>
Message-ID: <4EFAA256.9090403@gmail.com>

Le 2011-12-27 16:11, beno?t tuduri a ?crit :
> Hello,
> 
> I?m a bit surprise, in my macosx I can?t found the lua scripts which
> available (like freebox.lua in service discovery) in vlc git
> repository in a build version. These scripts are voluntary missing ?

VLC version?




From kaarlo.raiha at gmail.com  Wed Dec 28 07:29:51 2011
From: kaarlo.raiha at gmail.com (=?ISO-8859-1?B?S2FhcmxvIFLkaWjk?=)
Date: Wed, 28 Dec 2011 08:29:51 +0200
Subject: [vlc-devel] [PATCH] udev discovery: fix bluray
In-Reply-To: <1325016498.81915.YahooMailClassic@web113613.mail.gq1.yahoo.com>
References: <4EFA2088.6010907@gmail.com>
	<1325016498.81915.YahooMailClassic@web113613.mail.gq1.yahoo.com>
Message-ID: <CAMKrNtxmC5w9ehOLfbccrnjNHdFBzddQ0YM5pLJYwT6cP9s4kA@mail.gmail.com>

2011/12/27 Tony Anecito <adanecito at yahoo.com>

> Thanks guys I have been watching these posts waiting for bluray to be
> supported.
>
> So which nightlybuild wil this be in? 1.x, 1.2.x or 1.3.x
>

1.2.x has libbluray support.


>
> Thanks,
> -Tony
>
> --- On *Tue, 12/27/11, Rafa?l Carr? <rafael.carre at gmail.com>* wrote:
>
>
> From: Rafa?l Carr? <rafael.carre at gmail.com>
> Subject: Re: [vlc-devel] [PATCH] udev discovery: fix bluray
> To: "Mailing list for VLC media player developers" <vlc-devel at videolan.org
> >
> Date: Tuesday, December 27, 2011, 12:46 PM
>
>
> Le 2011-12-27 13:00, R?mi Denis-Courmont a ?crit :
> >    Hello,
>
> > Fix libbluray.
>
> Done:
>
> http://mailman.videolan.org/pipermail/libbluray-devel/2011-December/000550.html
>
> >> @@ -563,7 +600,7 @@ static char *disc_get_mrl (struct udev_device *dev)
> >>
> >>      val = udev_device_get_property_value (dev, "ID_CDROM_MEDIA_BD");
> >>      if (val && atoi (val))
> >> -        scheme = "bd";
> >> +        scheme = "bluray";
> >>  #ifdef LOL
> >>      val = udev_device_get_property_value (dev, "ID_CDROM_MEDIA_HDDVD");
> >>      if (val && atoi (val))
>
> I committed this part
> _______________________________________________
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
> http://mailman.videolan.org/listinfo/vlc-devel
>
>
> _______________________________________________
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
> http://mailman.videolan.org/listinfo/vlc-devel
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111228/8c9cd9bf/attachment.html>

From jb at videolan.org  Wed Dec 28 10:40:45 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Wed, 28 Dec 2011 10:40:45 +0100
Subject: [vlc-devel] VLC Development at School
In-Reply-To: <0D977D15-C207-403F-8AEE-8775DDEF4A1F@gmail.com>
References: <0D977D15-C207-403F-8AEE-8775DDEF4A1F@gmail.com>
Message-ID: <20111228094045.GA6016@videolan.org>

Hello,

On Tue, Dec 27, 2011 at 11:42:39PM -0500, Brendon Justin wrote :
> I want to make sure that you guys know and have no problem with it before I do the RCOS paperwork.  What do you think?

Great idea!

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From beauze.h at gmail.com  Wed Dec 28 11:04:34 2011
From: beauze.h at gmail.com (=?UTF-8?Q?Hugo_Beauz=C3=A9e=2DLuyssen?=)
Date: Wed, 28 Dec 2011 11:04:34 +0100
Subject: [vlc-devel] Race conditions in picture_pool with ffmpeg-mt
In-Reply-To: <20111227223330.GA12283@dinauz.org>
References: <20111227223330.GA12283@dinauz.org>
Message-ID: <CAK-C75QY7oAr3uoQ4BzMUsVUmPpC8bw_wuRxb1drvZMuo4GaXw@mail.gmail.com>

On Tue, Dec 27, 2011 at 11:33 PM, Denis Charmet <typx at dinauz.org> wrote:
> Hi,
>
>
> I've seen with the latest releases a lot of assertion failure in picture
> pool (in the Release function line 266).
>
> I think that ffmpeg-mt get the same picture in two different threads
> causing visual artefacts and then the "assertion failed" when the later
> one releases it.
> I tested with --ffmpeg-thread 1 and the problem disappears.
>
> Regards,
>

Hello, I encountered the same problem, worked with --ffmpeg-threads 4
though (instead of the default ffmpeg-threads 8)

Regards,

-- 
Hugo Beauz?e-Luyssen


From fyhuel at viotech.net  Wed Dec 28 11:14:18 2011
From: fyhuel at viotech.net (=?UTF-8?q?Fr=C3=A9d=C3=A9ric=20Yhuel?=)
Date: Wed, 28 Dec 2011 11:14:18 +0100
Subject: [vlc-devel] [PATCH] stream_filter/decomp.c: fix Read function
Message-ID: <1325067258-2886-1-git-send-email-fyhuel@viotech.net>

---
 modules/stream_filter/decomp.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/modules/stream_filter/decomp.c b/modules/stream_filter/decomp.c
index 7aef4f9..b86f846 100644
--- a/modules/stream_filter/decomp.c
+++ b/modules/stream_filter/decomp.c
@@ -186,7 +186,7 @@ static int Read (stream_t *stream, void *buf, unsigned int buflen)
         p_sys->offset += length;
 
         if (buflen > 0)
-            length += Read (stream, ((char *)buf) + length, buflen - length);
+            length += Read (stream, ((char *)buf) + length, buflen);
         return length;
     }
     assert ((buf != NULL) || (buflen == 0));
-- 
1.7.5.4



From fyhuel at viotech.net  Wed Dec 28 11:41:37 2011
From: fyhuel at viotech.net (Frederic YHUEL)
Date: Wed, 28 Dec 2011 11:41:37 +0100
Subject: [vlc-devel] [PATCH] stream_filter/decomp.c: fix Read function
In-Reply-To: <1325067258-2886-1-git-send-email-fyhuel@viotech.net>
References: <1325067258-2886-1-git-send-email-fyhuel@viotech.net>
Message-ID: <CAKHNsC2_ipEk7qT0aV9bOPADpDPD+rwqVYB2+QHcdFgggDrG4w@mail.gmail.com>

2011/12/28 Fr?d?ric Yhuel <fyhuel at viotech.net>:
> ---
> ?modules/stream_filter/decomp.c | ? ?2 +-
> ?1 files changed, 1 insertions(+), 1 deletions(-)
>
> diff --git a/modules/stream_filter/decomp.c b/modules/stream_filter/decomp.c
> index 7aef4f9..b86f846 100644
> --- a/modules/stream_filter/decomp.c
> +++ b/modules/stream_filter/decomp.c
> @@ -186,7 +186,7 @@ static int Read (stream_t *stream, void *buf, unsigned int buflen)
> ? ? ? ? p_sys->offset += length;
>
> ? ? ? ? if (buflen > 0)
> - ? ? ? ? ? ?length += Read (stream, ((char *)buf) + length, buflen - length);
> + ? ? ? ? ? ?length += Read (stream, ((char *)buf) + length, buflen);
> ? ? ? ? return length;
> ? ? }
> ? ? assert ((buf != NULL) || (buflen == 0));
> --
> 1.7.5.4
>

PS: This patch is not tested, because I've no idea about how to test
it... but it seems trivial enough :-)

Actually I was reading this code to try and understand what *exactly*
the Peek function is supposed to do in a stream_filter module.

So now it's clear, but seriously, this is not trivial. For example,
you could be mislead (as I did) by the Peek function of decomp.c if
you don't read also the Read function of the same file.

I'll try and write a wiki page for stream_filter modules if you agree.

Best Regards,

-- 
Fr?d?ric


From fkuehne.videolan at googlemail.com  Wed Dec 28 12:27:36 2011
From: fkuehne.videolan at googlemail.com (=?iso-8859-1?Q?Felix_Paul_K=FChne?=)
Date: Wed, 28 Dec 2011 12:27:36 +0100
Subject: [vlc-devel] [vlc-commits] Tag 1.2.0-pre3 : VLC media player -
	'Twoflower' - 1.2.0-pre3
In-Reply-To: <20111228024331.GA27834@videolan.org>
References: <20111228022657.31958144A6B@albiero.videolan.org>
	<20111228024331.GA27834@videolan.org>
Message-ID: <812DF43F-134E-43A6-B330-2C94830E44B8@gmail.com>


Am 28.12.2011 um 03:43 schrieb Jean-Baptiste Kempf:

> On Wed, Dec 28, 2011 at 03:26:57AM +0100, git at videolan.org wrote :
>> VLC media player - 'Twoflower' - 1.2.0-pre3
> 
> ftp://ftp.videolan.org/pub/videolan/testing/vlc-1.2.0-pre3/vlc-1.2.0-pre3.tar.xz
> 
> Win32 binaries will follow

Mac OS X binaries for Intel (separate 32 and 64 bit packages) here:

http://download.videolan.org/pub/testing/vlc-1.2.0-pre3/macosx/

A binary for PowerPC-based Macs might follow later on.


Best regards,

Felix

--
VideoLAN
Felix Paul K?hne
Developer of VLC's Mac OS X port
www.videolan.org



From remi at remlab.net  Wed Dec 28 14:39:29 2011
From: remi at remlab.net (=?iso-8859-1?q?R=E9mi?= Denis-Courmont)
Date: Wed, 28 Dec 2011 14:39:29 +0100
Subject: [vlc-devel] [PATCH] stream_filter/decomp.c: fix Read function
In-Reply-To: <CAKHNsC2_ipEk7qT0aV9bOPADpDPD+rwqVYB2+QHcdFgggDrG4w@mail.gmail.com>
References: <1325067258-2886-1-git-send-email-fyhuel@viotech.net>
	<CAKHNsC2_ipEk7qT0aV9bOPADpDPD+rwqVYB2+QHcdFgggDrG4w@mail.gmail.com>
Message-ID: <201112281439.29348.remi@remlab.net>

   Hello,

On Wednesday 28 December 2011, Frederic YHUEL wrote:
> 2011/12/28 Fr?d?ric Yhuel <fyhuel at viotech.net>:
> > ---
> >  modules/stream_filter/decomp.c |    2 +-
> >  1 files changed, 1 insertions(+), 1 deletions(-)
> > 
> > diff --git a/modules/stream_filter/decomp.c
> > b/modules/stream_filter/decomp.c index 7aef4f9..b86f846 100644
> > --- a/modules/stream_filter/decomp.c
> > +++ b/modules/stream_filter/decomp.c
> > @@ -186,7 +186,7 @@ static int Read (stream_t *stream, void *buf,
> > unsigned int buflen) p_sys->offset += length;
> > 
> >         if (buflen > 0)
> > -            length += Read (stream, ((char *)buf) + length, buflen -
> > length); +            length += Read (stream, ((char *)buf) + length,
> > buflen); return length;
> >     }
> >     assert ((buf != NULL) || (buflen == 0));
> > --
> > 1.7.5.4
> 
> PS: This patch is not tested, because I've no idea about how to test
> it... but it seems trivial enough :-)

I am not applying untested patches because (other) people have abused this far 
too much far too recently.

> Actually I was reading this code to try and understand what *exactly*
> the Peek function is supposed to do in a stream_filter module.
> 
> So now it's clear, but seriously, this is not trivial. For example,
> you could be mislead (as I did) by the Peek function of decomp.c if
> you don't read also the Read function of the same file.

I don't think anybody disagrees. Nobody has bothered to simplify the interface 
though. And it gets harder and harder as more plugins get added.

-- 
R?mi Denis-Courmont
http://www.remlab.info


From rafael.carre at gmail.com  Wed Dec 28 15:02:33 2011
From: rafael.carre at gmail.com (=?ISO-8859-1?Q?Rafa=EBl_Carr=E9?=)
Date: Wed, 28 Dec 2011 09:02:33 -0500
Subject: [vlc-devel] [vlc-commits] Tag 1.2.0-pre3 : VLC media player -
 'Twoflower' - 1.2.0-pre3
In-Reply-To: <812DF43F-134E-43A6-B330-2C94830E44B8@gmail.com>
References: <20111228022657.31958144A6B@albiero.videolan.org>
	<20111228024331.GA27834@videolan.org>
	<812DF43F-134E-43A6-B330-2C94830E44B8@gmail.com>
Message-ID: <4EFB2179.7000106@gmail.com>

Le 2011-12-28 06:27, Felix Paul K?hne a ?crit :
> 
> Am 28.12.2011 um 03:43 schrieb Jean-Baptiste Kempf:
> 
>> On Wed, Dec 28, 2011 at 03:26:57AM +0100, git at videolan.org wrote :
>>> VLC media player - 'Twoflower' - 1.2.0-pre3
>>
>> ftp://ftp.videolan.org/pub/videolan/testing/vlc-1.2.0-pre3/vlc-1.2.0-pre3.tar.xz
>>
>> Win32 binaries will follow
> 
> Mac OS X binaries for Intel (separate 32 and 64 bit packages) here:
> 
> http://download.videolan.org/pub/testing/vlc-1.2.0-pre3/macosx/
> 
> A binary for PowerPC-based Macs might follow later on.

Can you test PPC binaries?


From rafael.carre at gmail.com  Wed Dec 28 15:19:50 2011
From: rafael.carre at gmail.com (=?ISO-8859-1?Q?Rafa=EBl_Carr=E9?=)
Date: Wed, 28 Dec 2011 09:19:50 -0500
Subject: [vlc-devel] [PATCH 1/1] add bin to PATH to let libvlc.dll be
 found by vlc-cache-gen.exe
In-Reply-To: <1324785274-696-1-git-send-email-RSATom@gmail.com>
References: <1324785274-696-1-git-send-email-RSATom@gmail.com>
Message-ID: <4EFB2586.6070507@gmail.com>

Le 2011-12-24 22:54, Sergey Radionov a ?crit :
> Fix "This application has failed to start because libvlc.dll
> was not found. Re-installing the application may fix this problem."
> during building of vlc under Windows.
> ---
>  Makefile.am |    1 +
>  1 files changed, 1 insertions(+), 0 deletions(-)
> 
> diff --git a/Makefile.am b/Makefile.am
> index 358ac5f..dae8ad1 100644
> --- a/Makefile.am
> +++ b/Makefile.am
> @@ -406,6 +406,7 @@ vlc$(EXEEXT):
>  ###############################################################################
>  install-exec-hook:
>  	if test "$(build)" = "$(host)"; then \
> +		PATH="$(DESTDIR)$(bindir):$$PATH" \
>  		LD_LIBRARY_PATH="$(DESTDIR)$(libdir):$$LD_LIBRARY_PATH" \
>  		"$(DESTDIR)$(vlclibdir)/vlc-cache-gen$(EXEEXT)" \
>  			 "$(DESTDIR)$(vlclibdir)/plugins" ; \

LGTM


From fyhuel at viotech.net  Wed Dec 28 15:57:24 2011
From: fyhuel at viotech.net (Frederic YHUEL)
Date: Wed, 28 Dec 2011 15:57:24 +0100
Subject: [vlc-devel] [PATCH] stream_filter/decomp.c: fix Read function
In-Reply-To: <201112281439.29348.remi@remlab.net>
References: <1325067258-2886-1-git-send-email-fyhuel@viotech.net>
	<CAKHNsC2_ipEk7qT0aV9bOPADpDPD+rwqVYB2+QHcdFgggDrG4w@mail.gmail.com>
	<201112281439.29348.remi@remlab.net>
Message-ID: <CAKHNsC2X4+DYQSYUYTz_19n9m2-SZpaQtmMb16fK=9SbuWVENQ@mail.gmail.com>

>>
>> PS: This patch is not tested, because I've no idea about how to test
>> it... but it seems trivial enough :-)
>
> I am not applying untested patches because (other) people have abused this far
> too much far too recently.
>

Ok, I understand. I'll find a way to test it.

>> Actually I was reading this code to try and understand what *exactly*
>> the Peek function is supposed to do in a stream_filter module.
>>
>> So now it's clear, but seriously, this is not trivial. For example,
>> you could be mislead (as I did) by the Peek function of decomp.c if
>> you don't read also the Read function of the same file.
>
> I don't think anybody disagrees. Nobody has bothered to simplify the interface
> though. And it gets harder and harder as more plugins get added.
>

Ok. At least it's not just me, it's reassuring :-)

-- 
Fr?d?ric


From edward.c.wang at compdigitec.com  Wed Dec 28 16:59:40 2011
From: edward.c.wang at compdigitec.com (Edward Wang)
Date: Wed, 28 Dec 2011 10:59:40 -0500
Subject: [vlc-devel] [PATCH] Qt4: Make playlist window drag&drop to current
	selected PLSelItem
Message-ID: <1325087980-30045-1-git-send-email-edward.c.wang@compdigitec.com>

Make playlist window drag and drop to current selection (either media library or playlist). This should make it somewhat more user friendly because it drops into where the user expects it to.
Fixes #4873
Thanks,
	Edward Wang

---
 modules/gui/qt4/components/playlist/playlist.cpp |   12 +++++++++---
 modules/gui/qt4/components/playlist/playlist.hpp |    2 ++
 modules/gui/qt4/main_interface.cpp               |   15 ++++++++++++---
 modules/gui/qt4/main_interface.hpp               |    3 ++-
 4 files changed, 25 insertions(+), 7 deletions(-)

diff --git a/modules/gui/qt4/components/playlist/playlist.cpp b/modules/gui/qt4/components/playlist/playlist.cpp
index 661f376..469fc69 100644
--- a/modules/gui/qt4/components/playlist/playlist.cpp
+++ b/modules/gui/qt4/components/playlist/playlist.cpp
@@ -63,8 +63,8 @@ PlaylistWidget::PlaylistWidget( intf_thread_t *_p_i, QWidget *_par )
     leftSplitter = new QSplitter( Qt::Vertical, this );
 
     /* Source Selector */
-    PLSelector *selector = new PLSelector( this, p_intf );
-    leftSplitter->addWidget( selector);
+    selector = new PLSelector( this, p_intf );
+    leftSplitter->addWidget( selector );
 
     /* Create a Container for the Art Label
        in order to have a beautiful resizing for the selector above it */
@@ -211,8 +211,14 @@ PlaylistWidget::~PlaylistWidget()
 
 void PlaylistWidget::dropEvent( QDropEvent *event )
 {
+    bool is_pl;
+    int x = selector->currentItem()->data( 0, SPECIAL_ROLE ).toInt();
+    if( x == IS_PL ) is_pl = true;
+    else if( x == IS_ML ) is_pl = false;
+    else return;
+
     if( p_intf->p_sys->p_mi )
-        p_intf->p_sys->p_mi->dropEventPlay( event, false );
+        p_intf->p_sys->p_mi->dropEventPlay( event, false, is_pl );
 }
 void PlaylistWidget::dragEnterEvent( QDragEnterEvent *event )
 {
diff --git a/modules/gui/qt4/components/playlist/playlist.hpp b/modules/gui/qt4/components/playlist/playlist.hpp
index 7172f95..9709a29 100644
--- a/modules/gui/qt4/components/playlist/playlist.hpp
+++ b/modules/gui/qt4/components/playlist/playlist.hpp
@@ -33,6 +33,7 @@
 #include "qt4.hpp"
 
 //#include <vlc_playlist.h>
+#include "selector.hpp"
 
 #include <QSplitter>
 
@@ -61,6 +62,7 @@ private:
     QSplitter            *leftSplitter;
     QSplitter            *split;
     StandardPLPanel      *mainView;
+    PLSelector           *selector;
 
     QAction *viewActions[ 4 /* StandardPLPanel::VIEW_COUNT*/ ];
 
diff --git a/modules/gui/qt4/main_interface.cpp b/modules/gui/qt4/main_interface.cpp
index 0e88231..19c65a8 100644
--- a/modules/gui/qt4/main_interface.cpp
+++ b/modules/gui/qt4/main_interface.cpp
@@ -1192,7 +1192,16 @@ void MainInterface::dropEvent(QDropEvent *event)
     dropEventPlay( event, true );
 }
 
-void MainInterface::dropEventPlay( QDropEvent *event, bool b_play )
+/**
+ * dropEventPlay
+ *
+ * Event called if something is dropped onto a VLC window
+ * \param event the event in question
+ * \param b_play whether to play the file immediately
+ * \param b_playlist true to add to playlist, false to add to media library
+ * \return nothing
+ */
+void MainInterface::dropEventPlay( QDropEvent *event, bool b_play, bool b_playlist )
 {
     if( event->possibleActions() & ( Qt::CopyAction | Qt::MoveAction ) )
        event->setDropAction( Qt::CopyAction );
@@ -1221,7 +1230,7 @@ void MainInterface::dropEventPlay( QDropEvent *event, bool b_play )
             QString mrl = toURI( url.toEncoded().constData() );
             playlist_Add( THEPL, qtu(mrl), NULL,
                           PLAYLIST_APPEND | (first ? PLAYLIST_GO: PLAYLIST_PREPARSE),
-                          PLAYLIST_END, true, pl_Unlocked );
+                          PLAYLIST_END, b_playlist, pl_Unlocked );
             first = false;
             RecentsMRL::getInstance( p_intf )->addRecent( mrl );
         }
@@ -1236,7 +1245,7 @@ void MainInterface::dropEventPlay( QDropEvent *event, bool b_play )
         QString mrl = toURI( mimeData->text() );
         playlist_Add( THEPL, qtu(mrl), NULL,
                       PLAYLIST_APPEND | (first ? PLAYLIST_GO: PLAYLIST_PREPARSE),
-                      PLAYLIST_END, true, pl_Unlocked );
+                      PLAYLIST_END, b_playlist, pl_Unlocked );
     }
     event->accept();
 }
diff --git a/modules/gui/qt4/main_interface.hpp b/modules/gui/qt4/main_interface.hpp
index 87e04d5..142d092 100644
--- a/modules/gui/qt4/main_interface.hpp
+++ b/modules/gui/qt4/main_interface.hpp
@@ -88,7 +88,8 @@ public:
     bool isInterfaceFullScreen() { return b_interfaceFullScreen; }
 
 protected:
-    void dropEventPlay( QDropEvent *, bool);
+    void dropEventPlay( QDropEvent* event, bool b_play ) { dropEventPlay(event, b_play, true); }
+    void dropEventPlay( QDropEvent *, bool, bool );
 #ifdef WIN32
     virtual bool winEvent( MSG *, long * );
 #endif
-- 
1.7.5.4



From fkuehne.videolan at googlemail.com  Wed Dec 28 17:01:18 2011
From: fkuehne.videolan at googlemail.com (=?utf-8?Q?Felix_Paul_K=C3=BChne?=)
Date: Wed, 28 Dec 2011 17:01:18 +0100
Subject: [vlc-devel] [vlc-commits] Tag 1.2.0-pre3 : VLC media player -
	'Twoflower' - 1.2.0-pre3
In-Reply-To: <4EFB2179.7000106@gmail.com>
References: <20111228022657.31958144A6B@albiero.videolan.org>
	<20111228024331.GA27834@videolan.org>
	<812DF43F-134E-43A6-B330-2C94830E44B8@gmail.com>
	<4EFB2179.7000106@gmail.com>
Message-ID: <B75B6FB7-906F-4ED9-841F-00EF320DFEB4@gmail.com>

On 28.12.2011, at 15:02, Rafa?l Carr? <rafael.carre at gmail.com> wrote:

> Le 2011-12-28 06:27, Felix Paul K?hne a ?crit :
>> 
>> Am 28.12.2011 um 03:43 schrieb Jean-Baptiste Kempf:
>> 
>>> On Wed, Dec 28, 2011 at 03:26:57AM +0100, git at videolan.org wrote :
>>>> VLC media player - 'Twoflower' - 1.2.0-pre3
>>> 
>>> ftp://ftp.videolan.org/pub/videolan/testing/vlc-1.2.0-pre3/vlc-1.2.0-pre3.tar.xz
>>> 
>>> Win32 binaries will follow
>> 
>> Mac OS X binaries for Intel (separate 32 and 64 bit packages) here:
>> 
>> http://download.videolan.org/pub/testing/vlc-1.2.0-pre3/macosx/
>> 
>> A binary for PowerPC-based Macs might follow later on.
> 
> Can you test PPC binaries?
Yes, as soon as I'm home from my Christmas holidays, since I still got an iMac G5 :-)


Best regards,
Felix

From funman at videolan.org  Wed Dec 28 17:04:23 2011
From: funman at videolan.org (=?UTF-8?q?Rafa=C3=ABl=20Carr=C3=A9?=)
Date: Wed, 28 Dec 2011 11:04:23 -0500
Subject: [vlc-devel] [PATCH 1/2] contrib: helper to copy recent
	config.{guess, sub} from automake
Message-ID: <1325088264-30929-1-git-send-email-funman@videolan.org>

---
 contrib/src/main.mak |    8 ++++++++
 1 files changed, 8 insertions(+), 0 deletions(-)

diff --git a/contrib/src/main.mak b/contrib/src/main.mak
index fdd0892..3ab0014 100644
--- a/contrib/src/main.mak
+++ b/contrib/src/main.mak
@@ -259,6 +259,14 @@ UNPACK_DIR = $(basename $(basename $(notdir $<)))
 APPLY = (cd $(UNPACK_DIR) && patch -p1) <
 MOVE = mv $(UNPACK_DIR) $@ && touch $@
 
+AUTOMAKE_DATA_DIRS=$(abspath $(dir $(shell which automake))/../share/automake*)
+UPDATE_AUTOCONFIG = for dir in $(AUTOMAKE_DATA_DIRS); do \
+		if test -f "$${dir}/config.sub" -a -f "$${dir}/config.guess"; then \
+			cp "$${dir}/config.sub" "$${dir}/config.guess" $@; \
+			break; \
+		fi; \
+	done
+
 RECONF = mkdir -p -- $(PREFIX)/share/aclocal && \
 	cd $< && autoreconf -fiv $(ACLOCAL_AMFLAGS)
 CMAKE = cmake . -DCMAKE_TOOLCHAIN_FILE=$(abspath toolchain.cmake) \
-- 
1.7.7.3


From funman at videolan.org  Wed Dec 28 17:04:24 2011
From: funman at videolan.org (=?UTF-8?q?Rafa=C3=ABl=20Carr=C3=A9?=)
Date: Wed, 28 Dec 2011 11:04:24 -0500
Subject: [vlc-devel] [PATCH 2/2] contrib: gmp: update config.{guess, sub},
	needed for android
In-Reply-To: <1325088264-30929-1-git-send-email-funman@videolan.org>
References: <1325088264-30929-1-git-send-email-funman@videolan.org>
Message-ID: <1325088264-30929-2-git-send-email-funman@videolan.org>

---
 contrib/src/gmp/rules.mak |    1 +
 1 files changed, 1 insertions(+), 0 deletions(-)

diff --git a/contrib/src/gmp/rules.mak b/contrib/src/gmp/rules.mak
index 89ad5fc..d3b14c3 100644
--- a/contrib/src/gmp/rules.mak
+++ b/contrib/src/gmp/rules.mak
@@ -14,6 +14,7 @@ $(TARBALLS)/gmp-$(GMP_VERSION).tar.bz2:
 gmp: gmp-$(GMP_VERSION).tar.bz2 .sum-gmp
 	$(UNPACK)
 	$(MOVE)
+	$(UPDATE_AUTOCONFIG)
 
 .gmp: gmp
 	cd $< && $(HOSTVARS) ./configure $(HOSTCONF)
-- 
1.7.7.3


From remi at remlab.net  Wed Dec 28 19:00:31 2011
From: remi at remlab.net (=?iso-8859-1?q?R=E9mi?= Denis-Courmont)
Date: Wed, 28 Dec 2011 19:00:31 +0100
Subject: [vlc-devel] [PATCH 1/2] contrib: helper to copy recent
	config.{guess, sub} from automake
In-Reply-To: <1325088264-30929-1-git-send-email-funman@videolan.org>
References: <1325088264-30929-1-git-send-email-funman@videolan.org>
Message-ID: <201112281900.31847.remi@remlab.net>

   Hello,

On Wednesday 28 December 2011, Rafa?l Carr? wrote:
> ---
>  contrib/src/main.mak |    8 ++++++++
>  1 files changed, 8 insertions(+), 0 deletions(-)
> 
> diff --git a/contrib/src/main.mak b/contrib/src/main.mak
> index fdd0892..3ab0014 100644
> --- a/contrib/src/main.mak
> +++ b/contrib/src/main.mak
> @@ -259,6 +259,14 @@ UNPACK_DIR = $(basename $(basename $(notdir $<)))
>  APPLY = (cd $(UNPACK_DIR) && patch -p1) <
>  MOVE = mv $(UNPACK_DIR) $@ && touch $@
> 
> +AUTOMAKE_DATA_DIRS=$(abspath $(dir $(shell which
> automake))/../share/automake*) +UPDATE_AUTOCONFIG = for dir in
> $(AUTOMAKE_DATA_DIRS); do \
> +		if test -f "$${dir}/config.sub" -a -f "$${dir}/config.guess"; then \
> +			cp "$${dir}/config.sub" "$${dir}/config.guess" $@; \
> +			break; \
> +		fi; \
> +	done
> +

Is 'which' available on non-Debian systems?

-- 
R?mi Denis-Courmont
http://www.remlab.info


From remi at remlab.net  Wed Dec 28 19:01:08 2011
From: remi at remlab.net (=?iso-8859-1?q?R=E9mi?= Denis-Courmont)
Date: Wed, 28 Dec 2011 19:01:08 +0100
Subject: [vlc-devel] [PATCH 2/2] contrib: gmp: update config.{guess, sub},
	needed for android
In-Reply-To: <1325088264-30929-2-git-send-email-funman@videolan.org>
References: <1325088264-30929-1-git-send-email-funman@videolan.org>
	<1325088264-30929-2-git-send-email-funman@videolan.org>
Message-ID: <201112281901.08387.remi@remlab.net>

   Hello,

On Wednesday 28 December 2011, Rafa?l Carr? wrote:
> ---
>  contrib/src/gmp/rules.mak |    1 +
>  1 files changed, 1 insertions(+), 0 deletions(-)
> 
> diff --git a/contrib/src/gmp/rules.mak b/contrib/src/gmp/rules.mak
> index 89ad5fc..d3b14c3 100644
> --- a/contrib/src/gmp/rules.mak
> +++ b/contrib/src/gmp/rules.mak
> @@ -14,6 +14,7 @@ $(TARBALLS)/gmp-$(GMP_VERSION).tar.bz2:
>  gmp: gmp-$(GMP_VERSION).tar.bz2 .sum-gmp
>  	$(UNPACK)
>  	$(MOVE)
> +	$(UPDATE_AUTOCONFIG)

Eh. Wrong. This is going to break dependencies.

Either you do it before move or in the dot target.

>  .gmp: gmp
>  	cd $< && $(HOSTVARS) ./configure $(HOSTCONF)
-- 
R?mi Denis-Courmont
http://www.remlab.info


From rafael.carre at gmail.com  Wed Dec 28 19:03:26 2011
From: rafael.carre at gmail.com (=?ISO-8859-1?Q?Rafa=EBl_Carr=E9?=)
Date: Wed, 28 Dec 2011 13:03:26 -0500
Subject: [vlc-devel] [PATCH 2/2] contrib: gmp: update config.{guess, sub},
 needed for android
In-Reply-To: <201112281901.08387.remi@remlab.net>
References: <1325088264-30929-1-git-send-email-funman@videolan.org>
	<1325088264-30929-2-git-send-email-funman@videolan.org>
	<201112281901.08387.remi@remlab.net>
Message-ID: <4EFB59EE.8050606@gmail.com>

Le 2011-12-28 13:01, R?mi Denis-Courmont a ?crit :
>    Hello,
> 
> On Wednesday 28 December 2011, Rafa?l Carr? wrote:
>> ---
>>  contrib/src/gmp/rules.mak |    1 +
>>  1 files changed, 1 insertions(+), 0 deletions(-)
>>
>> diff --git a/contrib/src/gmp/rules.mak b/contrib/src/gmp/rules.mak
>> index 89ad5fc..d3b14c3 100644
>> --- a/contrib/src/gmp/rules.mak
>> +++ b/contrib/src/gmp/rules.mak
>> @@ -14,6 +14,7 @@ $(TARBALLS)/gmp-$(GMP_VERSION).tar.bz2:
>>  gmp: gmp-$(GMP_VERSION).tar.bz2 .sum-gmp
>>  	$(UNPACK)
>>  	$(MOVE)
>> +	$(UPDATE_AUTOCONFIG)
> 
> Eh. Wrong. This is going to break dependencies.
> 
> Either you do it before move or in the dot target.

ok

> 
>>  .gmp: gmp
>>  	cd $< && $(HOSTVARS) ./configure $(HOSTCONF)



From rafael.carre at gmail.com  Wed Dec 28 19:16:55 2011
From: rafael.carre at gmail.com (=?ISO-8859-1?Q?Rafa=EBl_Carr=E9?=)
Date: Wed, 28 Dec 2011 13:16:55 -0500
Subject: [vlc-devel] [PATCH 1/2] contrib: helper to copy recent
 config.{guess, sub} from automake
In-Reply-To: <201112281900.31847.remi@remlab.net>
References: <1325088264-30929-1-git-send-email-funman@videolan.org>
	<201112281900.31847.remi@remlab.net>
Message-ID: <4EFB5D17.6080706@gmail.com>

Le 2011-12-28 13:00, R?mi Denis-Courmont a ?crit :
>    Hello,
> 
> On Wednesday 28 December 2011, Rafa?l Carr? wrote:
>> ---
>>  contrib/src/main.mak |    8 ++++++++
>>  1 files changed, 8 insertions(+), 0 deletions(-)
>>
>> diff --git a/contrib/src/main.mak b/contrib/src/main.mak
>> index fdd0892..3ab0014 100644
>> --- a/contrib/src/main.mak
>> +++ b/contrib/src/main.mak
>> @@ -259,6 +259,14 @@ UNPACK_DIR = $(basename $(basename $(notdir $<)))
>>  APPLY = (cd $(UNPACK_DIR) && patch -p1) <
>>  MOVE = mv $(UNPACK_DIR) $@ && touch $@
>>
>> +AUTOMAKE_DATA_DIRS=$(abspath $(dir $(shell which
>> automake))/../share/automake*) +UPDATE_AUTOCONFIG = for dir in
>> $(AUTOMAKE_DATA_DIRS); do \
>> +		if test -f "$${dir}/config.sub" -a -f "$${dir}/config.guess"; then \
>> +			cp "$${dir}/config.sub" "$${dir}/config.guess" $@; \
>> +			break; \
>> +		fi; \
>> +	done
>> +
> 
> Is 'which' available on non-Debian systems?

It is on osx and msys

opensuse has it in util-linux, fedora in which

looks pretty standard


From adanecito at yahoo.com  Wed Dec 28 19:21:14 2011
From: adanecito at yahoo.com (Tony Anecito)
Date: Wed, 28 Dec 2011 10:21:14 -0800 (PST)
Subject: [vlc-devel] [PATCH] udev discovery: fix bluray
In-Reply-To: <CAMKrNtxmC5w9ehOLfbccrnjNHdFBzddQ0YM5pLJYwT6cP9s4kA@mail.gmail.com>
Message-ID: <1325096474.10709.YahooMailClassic@web113611.mail.gq1.yahoo.com>

Thanks Kaarlo!
-Tony

--- On Tue, 12/27/11, Kaarlo R?ih? <kaarlo.raiha at gmail.com> wrote:


From: Kaarlo R?ih? <kaarlo.raiha at gmail.com>
Subject: Re: [vlc-devel] [PATCH] udev discovery: fix bluray
To: "Mailing list for VLC media player developers" <vlc-devel at videolan.org>
Date: Tuesday, December 27, 2011, 11:29 PM


2011/12/27 Tony Anecito <adanecito at yahoo.com>







Thanks guys I have been watching these posts waiting for bluray to be supported.
?
So which nightlybuild wil this be in? 1.x, 1.2.x or 1.3.x


1.2.x has?libbluray support.
?





?
Thanks,
-Tony

--- On Tue, 12/27/11, Rafa?l Carr? <rafael.carre at gmail.com> wrote:


From: Rafa?l Carr? <rafael.carre at gmail.com>
Subject: Re: [vlc-devel] [PATCH] udev discovery: fix bluray
To: "Mailing list for VLC media player developers" <vlc-devel at videolan.org>
Date: Tuesday, December 27, 2011, 12:46 PM





Le 2011-12-27 13:00, R?mi Denis-Courmont a ?crit :
>? ? Hello,

> Fix libbluray.

Done:
http://mailman.videolan.org/pipermail/libbluray-devel/2011-December/000550.html

>> @@ -563,7 +600,7 @@ static char *disc_get_mrl (struct udev_device *dev)
>>
>>? ? ? val = udev_device_get_property_value (dev, "ID_CDROM_MEDIA_BD");
>>? ? ? if (val && atoi (val))
>> -? ? ? ? scheme = "bd";
>> +? ? ? ? scheme = "bluray";
>>? #ifdef LOL
>>? ? ? val = udev_device_get_property_value (dev, "ID_CDROM_MEDIA_HDDVD");
>>? ? ? if (val && atoi (val))

I committed this part
_______________________________________________
vlc-devel mailing list
To unsubscribe or modify your subscription options:
http://mailman.videolan.org/listinfo/vlc-devel

_______________________________________________
vlc-devel mailing list
To unsubscribe or modify your subscription options:
http://mailman.videolan.org/listinfo/vlc-devel



-----Inline Attachment Follows-----


_______________________________________________
vlc-devel mailing list
To unsubscribe or modify your subscription options:
http://mailman.videolan.org/listinfo/vlc-devel
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111228/4fad849b/attachment.html>

From benoit.tuduri at gmail.com  Wed Dec 28 19:23:21 2011
From: benoit.tuduri at gmail.com (=?ISO-8859-1?Q?beno=EEt_tuduri?=)
Date: Wed, 28 Dec 2011 19:23:21 +0100
Subject: [vlc-devel] service discovery scripts in macosx build
In-Reply-To: <4EFAA256.9090403@gmail.com>
References: <CALKXa+-B5Nrh9P_4JG1f1JixEutU70tO0WBLUwPz=veAxFhR5A@mail.gmail.com>
	<4EFAA256.9090403@gmail.com>
Message-ID: <CALKXa+8G176mwEMgd10fbUv7aVwOLUdXkVqL3H3iEbJDPHerHQ@mail.gmail.com>

Hello, thanks you for your help,

My VLC version is :

version 1.1.12 The Luggage (Intel 64 bit)
compil? par fpk avec llvm-gcc 4.2.1 (Based on Apple Inc.build 5658)
(LLVM build 2335.6)

My OS :
Mac OS X 10.6.8 (10K549)



2011/12/28 Rafa?l Carr? <rafael.carre at gmail.com>:
> Le 2011-12-27 16:11, beno?t tuduri a ?crit :
>> Hello,
>>
>> I?m a bit surprise, in my macosx I can?t found the lua scripts which
>> available (like freebox.lua in service discovery) in vlc git
>> repository in a build version. These scripts are voluntary missing ?
>
> VLC version?
>
>
> _______________________________________________
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
> http://mailman.videolan.org/listinfo/vlc-devel


From rafael.carre at gmail.com  Wed Dec 28 19:24:13 2011
From: rafael.carre at gmail.com (=?windows-1252?Q?Rafa=EBl_Carr=E9?=)
Date: Wed, 28 Dec 2011 13:24:13 -0500
Subject: [vlc-devel] service discovery scripts in macosx build
In-Reply-To: <CALKXa+8G176mwEMgd10fbUv7aVwOLUdXkVqL3H3iEbJDPHerHQ@mail.gmail.com>
References: <CALKXa+-B5Nrh9P_4JG1f1JixEutU70tO0WBLUwPz=veAxFhR5A@mail.gmail.com>
	<4EFAA256.9090403@gmail.com>
	<CALKXa+8G176mwEMgd10fbUv7aVwOLUdXkVqL3H3iEbJDPHerHQ@mail.gmail.com>
Message-ID: <4EFB5ECD.5020403@gmail.com>

Le 2011-12-28 13:23, beno?t tuduri a ?crit :
> Hello, thanks you for your help,
> 
> My VLC version is :
> 
> version 1.1.12 The Luggage (Intel 64 bit)
> compil? par fpk avec llvm-gcc 4.2.1 (Based on Apple Inc.build 5658)
> (LLVM build 2335.6)
> 
> My OS :
> Mac OS X 10.6.8 (10K549)

Please test 1.2 from http://nightlies.videolan.org/ , it should work


From benoit.tuduri at gmail.com  Wed Dec 28 19:25:56 2011
From: benoit.tuduri at gmail.com (=?ISO-8859-1?Q?beno=EEt_tuduri?=)
Date: Wed, 28 Dec 2011 19:25:56 +0100
Subject: [vlc-devel] service discovery scripts in macosx build
In-Reply-To: <4EFB5ECD.5020403@gmail.com>
References: <CALKXa+-B5Nrh9P_4JG1f1JixEutU70tO0WBLUwPz=veAxFhR5A@mail.gmail.com>
	<4EFAA256.9090403@gmail.com>
	<CALKXa+8G176mwEMgd10fbUv7aVwOLUdXkVqL3H3iEbJDPHerHQ@mail.gmail.com>
	<4EFB5ECD.5020403@gmail.com>
Message-ID: <CALKXa+_5b563Gij6202urP3VSPNoVUCfwxnM3e2ySghHYiwzgg@mail.gmail.com>

Ok, :)


2011/12/28 Rafa?l Carr? <rafael.carre at gmail.com>:
> Le 2011-12-28 13:23, beno?t tuduri a ?crit :
>> Hello, thanks you for your help,
>>
>> My VLC version is :
>>
>> version 1.1.12 The Luggage (Intel 64 bit)
>> compil? par fpk avec llvm-gcc 4.2.1 (Based on Apple Inc.build 5658)
>> (LLVM build 2335.6)
>>
>> My OS :
>> Mac OS X 10.6.8 (10K549)
>
> Please test 1.2 from http://nightlies.videolan.org/ , it should work
> _______________________________________________
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
> http://mailman.videolan.org/listinfo/vlc-devel


From benoit.tuduri at gmail.com  Wed Dec 28 19:28:51 2011
From: benoit.tuduri at gmail.com (=?ISO-8859-1?Q?beno=EEt_tuduri?=)
Date: Wed, 28 Dec 2011 19:28:51 +0100
Subject: [vlc-devel] service discovery scripts in macosx build
In-Reply-To: <CALKXa+_5b563Gij6202urP3VSPNoVUCfwxnM3e2ySghHYiwzgg@mail.gmail.com>
References: <CALKXa+-B5Nrh9P_4JG1f1JixEutU70tO0WBLUwPz=veAxFhR5A@mail.gmail.com>
	<4EFAA256.9090403@gmail.com>
	<CALKXa+8G176mwEMgd10fbUv7aVwOLUdXkVqL3H3iEbJDPHerHQ@mail.gmail.com>
	<4EFB5ECD.5020403@gmail.com>
	<CALKXa+_5b563Gij6202urP3VSPNoVUCfwxnM3e2ySghHYiwzgg@mail.gmail.com>
Message-ID: <CALKXa+-BSRQk6TAOO-2=NEHrxL+6f47Oh+KWxGe8ZwpamnuY+g@mail.gmail.com>

Yes, it?s works :)

Thanks you :)

Beno?t

Le 28 d?cembre 2011 19:25, beno?t tuduri <benoit.tuduri at gmail.com> a ?crit :
> Ok, :)
>
>
> 2011/12/28 Rafa?l Carr? <rafael.carre at gmail.com>:
>> Le 2011-12-28 13:23, beno?t tuduri a ?crit :
>>> Hello, thanks you for your help,
>>>
>>> My VLC version is :
>>>
>>> version 1.1.12 The Luggage (Intel 64 bit)
>>> compil? par fpk avec llvm-gcc 4.2.1 (Based on Apple Inc.build 5658)
>>> (LLVM build 2335.6)
>>>
>>> My OS :
>>> Mac OS X 10.6.8 (10K549)
>>
>> Please test 1.2 from http://nightlies.videolan.org/ , it should work
>> _______________________________________________
>> vlc-devel mailing list
>> To unsubscribe or modify your subscription options:
>> http://mailman.videolan.org/listinfo/vlc-devel


From cheng.sun at ymail.com  Wed Dec 28 19:54:06 2011
From: cheng.sun at ymail.com (Cheng Sun)
Date: Wed, 28 Dec 2011 18:54:06 +0000
Subject: [vlc-devel] [PATCH] Audio channel remapping filter
In-Reply-To: <201112232056.57830.remi@remlab.net>
References: <4EF1BC96.6080800@ymail.com> <201112211947.29144.remi@remlab.net>
	<4EF2483D.2070808@ymail.com> <201112232056.57830.remi@remlab.net>
Message-ID: <4EFB65CE.6030109@ymail.com>

On 23/12/11 19:56, R?mi Denis-Courmont wrote:

> However, you should move GetRemapFun() earlier in probing.

Unfortunately not possible in the current implementation (it depends on
b_multiple).

All other issues addressed. (+ fixed mistake in commit message. Of
course you can't upmix with a remapper.)

Cheers,
Cheng
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-Audio-channel-remapping-filter.patch
Type: text/x-patch
Size: 16482 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111228/bd997aec/attachment.bin>

From jeanmichel.lambert7 at gmail.com  Wed Dec 28 23:41:54 2011
From: jeanmichel.lambert7 at gmail.com (JM Lambert)
Date: Wed, 28 Dec 2011 23:41:54 +0100
Subject: [vlc-devel] [PATCH] Change to npapi for giving access to media meta
	thru new input.meta(i) method.
Message-ID: <1325112114-6218-1-git-send-email-jean-michel.lambert7@wanadoo.fr>

---
 npapi/control/npolibvlc.cpp |   36 ++++++++++++++++++++++++++++++++----
 1 files changed, 32 insertions(+), 4 deletions(-)

diff --git a/npapi/control/npolibvlc.cpp b/npapi/control/npolibvlc.cpp
index a8a94f6..8ad8fb2 100644
--- a/npapi/control/npolibvlc.cpp
+++ b/npapi/control/npolibvlc.cpp
@@ -31,6 +31,7 @@
 
 #include "position.h"
 
+
 /*
 ** Local helper macros and function
 */
@@ -44,6 +45,8 @@
 #define ERROR_EVENT_NOT_FOUND "ERROR: One or more events could not be found."
 #define ERROR_API_VERSION "ERROR: NPAPI version not high enough. (Gecko >= 1.9 needed)"
 
+
+
 // Make a copy of an NPVariant.
 NPVariant copyNPVariant(const NPVariant& original)
 {
@@ -588,13 +591,13 @@ LibvlcInputNPObject::setProperty(int index, const NPVariant &value)
 const NPUTF8 * const LibvlcInputNPObject::methodNames[] =
 {
     /* no methods */
-    "none",
+    "meta",
 };
 COUNTNAMES(LibvlcInputNPObject,methodCount,methodNames);
 
 enum LibvlcInputNPObjectMethodIds
 {
-    ID_none,
+    ID_input_meta,
 };
 
 RuntimeNPObject::InvokeResult
@@ -604,10 +607,34 @@ LibvlcInputNPObject::invoke(int index, const NPVariant *args,
     /* is plugin still running */
     if( isPluginRunning() )
     {
+        VlcPlugin* p_plugin = getPrivate<VlcPlugin>();
+        libvlc_media_player_t *p_md = p_plugin->getMD();
+	libvlc_media_t * p_media = libvlc_media_player_get_media(p_md);
+        if( !p_md )
+            RETURN_ON_ERROR;
         switch( index )
         {
-            case ID_none:
-                return INVOKERESULT_NO_SUCH_METHOD;
+            case ID_input_meta:
+		if( argCount == 1)
+                {
+                    char *info;
+                    int i_metaID,  i;
+
+                    /* get subtitles description */
+		     /* check if a number is given by the user
+                     * and get the subtitle number */
+                    if( isNumberValue(args[0]) )
+		    {
+                        i_metaID = numberValue(args[0]);
+			if (i_metaID<0 || i_metaID>libvlc_meta_TrackID)
+                           return INVOKERESULT_INVALID_VALUE;
+
+                        info = libvlc_media_get_meta(p_media,(libvlc_meta_t) i_metaID);
+                    } 
+		    else
+                        return INVOKERESULT_INVALID_VALUE;
+		    return invokeResultString(info, result);
+		}
             default:
                 ;
         }
@@ -1935,3 +1962,4 @@ LibvlcDeinterlaceNPObject::invoke(int index, const NPVariant *args,
     }
     return INVOKERESULT_NO_ERROR;
 }
+
-- 
1.7.5.4



From edward.c.wang at compdigitec.com  Thu Dec 29 02:29:02 2011
From: edward.c.wang at compdigitec.com (Edward Wang)
Date: Wed, 28 Dec 2011 20:29:02 -0500
Subject: [vlc-devel] [PATCH] Qt4: Add dialog to set specific playback speed
Message-ID: <1325122142-23755-1-git-send-email-edward.c.wang@compdigitec.com>

This patch adds a dialog to be able to set a specific playback speed (e.g. 0.97x or 1.01x) instead of fumbling around with the small slider to get a desired value.
Fixes bug #5734.
Thanks,
	Edward Wang

---
 modules/gui/qt4/Modules.am                       |    7 +-
 modules/gui/qt4/components/interface_widgets.cpp |    6 +-
 modules/gui/qt4/components/interface_widgets.hpp |    9 ++-
 modules/gui/qt4/dialogs/setspeed.cpp             |  110 ++++++++++++++++++++++
 modules/gui/qt4/dialogs/setspeed.hpp             |   50 ++++++++++
 modules/gui/qt4/dialogs_provider.cpp             |    6 +
 modules/gui/qt4/dialogs_provider.hpp             |    1 +
 modules/gui/qt4/main_interface.cpp               |    1 +
 8 files changed, 186 insertions(+), 4 deletions(-)
 create mode 100644 modules/gui/qt4/dialogs/setspeed.cpp
 create mode 100644 modules/gui/qt4/dialogs/setspeed.hpp

diff --git a/modules/gui/qt4/Modules.am b/modules/gui/qt4/Modules.am
index 22c2277..56510f5 100644
--- a/modules/gui/qt4/Modules.am
+++ b/modules/gui/qt4/Modules.am
@@ -38,6 +38,7 @@ nodist_SOURCES_qt4 = \
 		dialogs/convert.moc.cpp \
 		dialogs/help.moc.cpp \
 		dialogs/gototime.moc.cpp \
+		dialogs/setspeed.moc.cpp \
 		dialogs/toolbar.moc.cpp \
 		dialogs/open.moc.cpp \
 		dialogs/openurl.moc.cpp \
@@ -245,8 +246,8 @@ endif
 	$(AM_V_at)sed -e 's/Q_(\"_(\\\"\(.*\)\\\")"/Q_("\1"/' $@.tmp >$@
 	$(AM_V_at)rm -f $@.tmp

-.mm.lo:
-	$(top_builddir)/libtool --verbose --mode=compile $(CXX) $(objcxxflags) -DQ_WS_MAC -I$(top_srcdir)/include -I$(top_builddir)/include -I$(top_builddir) $(CXXFLAGS_qt4) -c $< -o $@
+.mm.lo:
+	$(top_builddir)/libtool --verbose --mode=compile $(CXX) $(objcxxflags) -DQ_WS_MAC -I$(top_srcdir)/include -I$(top_builddir)/include -I$(top_builddir) $(CXXFLAGS_qt4) -c $< -o $@

 SOURCES_qt4 = 	qt4.cpp \
 		menus.cpp \
@@ -273,6 +274,7 @@ SOURCES_qt4 = 	qt4.cpp \
 		dialogs/convert.cpp \
 		dialogs/help.cpp \
 		dialogs/gototime.cpp \
+		dialogs/setspeed.cpp \
 		dialogs/toolbar.cpp \
 		dialogs/open.cpp \
 		dialogs/openurl.cpp \
@@ -351,6 +353,7 @@ noinst_HEADERS = \
 	dialogs/convert.hpp \
 	dialogs/help.hpp \
 	dialogs/gototime.hpp \
+	dialogs/setspeed.hpp \
 	dialogs/toolbar.hpp \
 	dialogs/open.hpp \
 	dialogs/openurl.hpp \
diff --git a/modules/gui/qt4/components/interface_widgets.cpp b/modules/gui/qt4/components/interface_widgets.cpp
index e384092..eae1890 100644
--- a/modules/gui/qt4/components/interface_widgets.cpp
+++ b/modules/gui/qt4/components/interface_widgets.cpp
@@ -333,7 +333,11 @@ void VisualSelector::next()
 SpeedLabel::SpeedLabel( intf_thread_t *_p_intf, QWidget *parent )
            : QLabel( parent ), p_intf( _p_intf )
 {
-    tooltipStringPattern = qtr( "Current playback speed: %1\nClick to adjust" );
+    tooltipStringPattern = qtr( "Current playback speed: %1" ) +
+                           QString("\n- ") +
+                           qtr( "Click to adjust" ) +
+                           QString("\n- ") +
+                           qtr( "Right click to set a specific playback speed" );

     /* Create the Speed Control Widget */
     speedControl = new SpeedControlWidget( p_intf, this );
diff --git a/modules/gui/qt4/components/interface_widgets.hpp b/modules/gui/qt4/components/interface_widgets.hpp
index a2ce18f..75b7ec8 100644
--- a/modules/gui/qt4/components/interface_widgets.hpp
+++ b/modules/gui/qt4/components/interface_widgets.hpp
@@ -164,12 +164,19 @@ class SpeedLabel : public QLabel
 public:
     SpeedLabel( intf_thread_t *, QWidget * );
     virtual ~SpeedLabel();
+signals:
+    void speedLabelRightClicked();

 protected:
     virtual void mousePressEvent ( QMouseEvent * event )
     {
-        showSpeedMenu( event->pos() );
+        if( event->button() == Qt::RightButton ) {
+            emit speedLabelRightClicked();
+        } else {
+            showSpeedMenu( event->pos() );
+        }
     }
+
 private slots:
     void showSpeedMenu( QPoint );
     void setRate( float );
diff --git a/modules/gui/qt4/dialogs/setspeed.cpp b/modules/gui/qt4/dialogs/setspeed.cpp
new file mode 100644
index 0000000..eccceb4
--- /dev/null
+++ b/modules/gui/qt4/dialogs/setspeed.cpp
@@ -0,0 +1,110 @@
+/*****************************************************************************
+ * setspeed.cpp : Set speed dialog
+ ****************************************************************************
+ * Copyright (C) 2011 the VideoLAN team
+ * $Id$
+ *
+ * Authors: Edward Wang <edward.c.wang (at) compdigitec.com>
+ *          Jean-Baptiste Kempf <jb (at) videolan.org>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston MA 02110-1301, USA.
+ *****************************************************************************/
+#ifdef HAVE_CONFIG_H
+# include "config.h"
+#endif
+
+#include "dialogs/setspeed.hpp"
+
+#include "input_manager.hpp"
+
+#include <QTabWidget>
+#include <QLabel>
+#include <QDoubleSpinBox>
+#include <QGroupBox>
+#include <QDialogButtonBox>
+#include <QPushButton>
+
+SetSpeedDialog::SetSpeedDialog( intf_thread_t* _p_intf)
+               : QVLCDialog( (QWidget*)_p_intf->p_sys->p_mi, _p_intf )
+{
+    setWindowFlags( Qt::Tool );
+    setWindowTitle( qtr( "Set playback speed" ) );
+    setWindowRole( "vlc-set-time" );
+
+    QGridLayout* mainLayout = new QGridLayout( this );
+    mainLayout->setSizeConstraint( QLayout::SetFixedSize );
+
+    QPushButton* setButton = new QPushButton( qtr( "&Set" ) );
+    QPushButton* cancelButton = new QPushButton( qtr( "&Cancel" ) );
+    QDialogButtonBox* buttonBox = new QDialogButtonBox;
+
+    setButton->setDefault( true );
+    buttonBox->addButton( setButton, QDialogButtonBox::AcceptRole );
+    buttonBox->addButton( cancelButton, QDialogButtonBox::RejectRole );
+
+    QLabel* speedIntro = new QLabel( qtr( "Set playback speed" ) + ":" );
+    speedIntro->setWordWrap( true );
+    speedIntro->setAlignment( Qt::AlignCenter );
+
+    valueBox = new QDoubleSpinBox();
+    valueBox->setDecimals( 2 );
+    valueBox->setMaximum( 32 );
+    valueBox->setMinimum( 0.03F );
+    valueBox->setSingleStep( 0.10F );
+    valueBox->setAlignment( Qt::AlignRight );
+    valueBox->setSizePolicy( QSizePolicy::Expanding, QSizePolicy::Minimum );
+
+    QPushButton* resetButton = new QPushButton( QIcon(":/update"), "" );
+    resetButton->setToolTip( qtr("Reset") );
+
+    mainLayout->addWidget( speedIntro, 0, 0, 1, 1 );
+    mainLayout->addWidget( valueBox, 0, 1, 1, 1 );
+    mainLayout->addWidget( resetButton, 0, 2, 1, 1 );
+
+    mainLayout->addWidget( buttonBox, 1, 0, 1, 3 );
+
+    BUTTONACT( setButton, close() );
+    BUTTONACT( cancelButton, cancel() );
+    BUTTONACT( resetButton, reset() );
+}
+
+SetSpeedDialog::~SetSpeedDialog()
+{
+}
+
+void SetSpeedDialog::toggleVisible()
+{
+    reset();
+    valueBox->setValue( var_InheritFloat( THEPL, "rate" ) );
+    QVLCDialog::toggleVisible();
+}
+
+void SetSpeedDialog::cancel()
+{
+    reset();
+    toggleVisible();
+}
+
+void SetSpeedDialog::close()
+{
+    if ( THEMIM->getIM()->hasInput() )
+        var_SetFloat( THEPL, "rate", valueBox->value() );
+    toggleVisible();
+}
+
+void SetSpeedDialog::reset()
+{
+    valueBox->setValue( 1.00F );
+}
diff --git a/modules/gui/qt4/dialogs/setspeed.hpp b/modules/gui/qt4/dialogs/setspeed.hpp
new file mode 100644
index 0000000..58e43d1
--- /dev/null
+++ b/modules/gui/qt4/dialogs/setspeed.hpp
@@ -0,0 +1,50 @@
+/*****************************************************************************
+ * setspeed.hpp : Set speed dialog
+ ****************************************************************************
+ * Copyright (C) 2011 the VideoLAN team
+ * $Id$
+ *
+ * Authors: Edward Wang <edward.c.wang (at) compdigitec.com>
+ *          Jean-Baptiste Kempf <jb (at) videolan.org>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston MA 02110-1301, USA.
+ *****************************************************************************/
+
+#ifndef QVLC_SETSPEED_DIALOG_H_
+#define QVLC_SETSPEED_DIALOG_H_ 1
+
+#include "util/qvlcframe.hpp"
+#include "util/singleton.hpp"
+
+#include <QDoubleSpinBox>
+
+class SetSpeedDialog : public QVLCDialog, public Singleton<SetSpeedDialog>
+{
+    Q_OBJECT
+private:
+    SetSpeedDialog( intf_thread_t* );
+    virtual ~SetSpeedDialog();
+    QDoubleSpinBox* valueBox;
+private slots:
+    void close();
+    void cancel();
+    void reset();
+
+    friend class    Singleton<SetSpeedDialog>;
+public:
+    void toggleVisible();
+};
+
+#endif
diff --git a/modules/gui/qt4/dialogs_provider.cpp b/modules/gui/qt4/dialogs_provider.cpp
index a18b4b6..b8f5cf4 100644
--- a/modules/gui/qt4/dialogs_provider.cpp
+++ b/modules/gui/qt4/dialogs_provider.cpp
@@ -50,6 +50,7 @@
 #include "dialogs/openurl.hpp"
 #include "dialogs/help.hpp"
 #include "dialogs/gototime.hpp"
+#include "dialogs/setspeed.hpp"
 #include "dialogs/podcast_configuration.hpp"
 #include "dialogs/toolbar.hpp"
 #include "dialogs/plugins.hpp"
@@ -220,6 +221,11 @@ void DialogsProvider::gotoTimeDialog()
     GotoTimeDialog::getInstance( p_intf )->toggleVisible();
 }

+void DialogsProvider::setSpeedDialog()
+{
+    SetSpeedDialog::getInstance( p_intf )->toggleVisible();
+}
+
 #ifdef ENABLE_VLM
 void DialogsProvider::vlmDialog()
 {
diff --git a/modules/gui/qt4/dialogs_provider.hpp b/modules/gui/qt4/dialogs_provider.hpp
index 7429057..d859647 100644
--- a/modules/gui/qt4/dialogs_provider.hpp
+++ b/modules/gui/qt4/dialogs_provider.hpp
@@ -139,6 +139,7 @@ public slots:
 #endif
     void aboutDialog();
     void gotoTimeDialog();
+    void setSpeedDialog();
     void podcastConfigureDialog();
     void toolbarDialog();
     void pluginDialog();
diff --git a/modules/gui/qt4/main_interface.cpp b/modules/gui/qt4/main_interface.cpp
index 19c65a8..abf790a 100644
--- a/modules/gui/qt4/main_interface.cpp
+++ b/modules/gui/qt4/main_interface.cpp
@@ -463,6 +463,7 @@ inline void MainInterface::createStatusBar()
     nameLabel->setTextInteractionFlags( Qt::TextSelectableByMouse
                                       | Qt::TextSelectableByKeyboard );
     SpeedLabel *speedLabel = new SpeedLabel( p_intf, this );
+    CONNECT( speedLabel, speedLabelRightClicked(), THEDP, setSpeedDialog() );

     /* Styling those labels */
     timeLabel->setFrameStyle( QFrame::Sunken | QFrame::Panel );
--
1.7.5.4



From jb at videolan.org  Thu Dec 29 03:08:01 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Thu, 29 Dec 2011 03:08:01 +0100
Subject: [vlc-devel] [PATCH] Avcodec: limit the number of threads
	automatically selected to 12
Message-ID: <1325124481-30031-1-git-send-email-jb@videolan.org>

It seems that over 16 doesn't behave well. Reduce it further for safety
---
 modules/codec/avcodec/video.c |    4 ++++
 1 files changed, 4 insertions(+), 0 deletions(-)

diff --git a/modules/codec/avcodec/video.c b/modules/codec/avcodec/video.c
index 95e64b5..dada16c 100644
--- a/modules/codec/avcodec/video.c
+++ b/modules/codec/avcodec/video.c
@@ -344,7 +344,11 @@ int InitVideoDec( decoder_t *p_dec, AVCodecContext *p_context,
 #ifdef HAVE_AVCODEC_MT
     int i_thread_count = var_InheritInteger( p_dec, "ffmpeg-threads" );
     if( i_thread_count <= 0 )
+    {
         i_thread_count = vlc_GetCPUCount();
+        if( i_thread_count >= 12 )
+            i_thread_count = 12;
+    }
     msg_Dbg( p_dec, "allowing %d thread(s) for decoding", i_thread_count );
     p_sys->p_context->thread_count = i_thread_count;
 #endif
-- 
1.7.8.rc3



From zhushisongzhu at yahoo.com  Thu Dec 29 04:36:03 2011
From: zhushisongzhu at yahoo.com (zhu shi song)
Date: Wed, 28 Dec 2011 19:36:03 -0800 (PST)
Subject: [vlc-devel] Athlon 64 CPU problem
Message-ID: <1325129763.63208.YahooMailNeo@web125705.mail.ne1.yahoo.com>

???? Following the instructions on WEB http://wiki.videolan.org/Win32CompileMSYSNew#Precompiled_contribs, I have compiled VLC 1.1.5 successfully.? I used contrib-20100608-win32-bin-gcc-4.4.4-runtime-3.17-only.tar.bz2.? 

???? I use it on Intel CPU computer , it works well.? But I test it on AMD Athlon 64 CPU computer, it can't play MP4(h.264+AAC) smoothly.? The time progress bar is moving, but video is frozen.
??? The CPU details:?? AMD Athlon 64x2 Dual Core Processor 4400+.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111228/8ab32a59/attachment.html>

From funman at videolan.org  Thu Dec 29 05:57:57 2011
From: funman at videolan.org (=?UTF-8?q?Rafa=C3=ABl=20Carr=C3=A9?=)
Date: Wed, 28 Dec 2011 23:57:57 -0500
Subject: [vlc-devel] [PATCH] motion control: fix for recent kernel
Message-ID: <1325134677-8443-1-git-send-email-funman@videolan.org>

---
What to do : support older kernels too?
Check other drivers if they moved to /platform too?

Btw the --motion-use-rotate doesn't work anymore.
These days it's forbidden to find another module by name (video filter "rotate")

 modules/control/motion.c |    6 +++---
 1 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/modules/control/motion.c b/modules/control/motion.c
index b234f5c..cee8c96 100644
--- a/modules/control/motion.c
+++ b/modules/control/motion.c
@@ -129,11 +129,11 @@ int Open ( vlc_object_t *p_this )
         /* Apple Motion Sensor support */
         p_intf->p_sys->sensor = AMS_SENSOR;
     }
-    else if( access( "/sys/devices/applesmc.768/position", R_OK ) == 0 )
+    else if( access( "/sys/devices/platform/applesmc.768/position", R_OK ) == 0 )
     {
         /* Apple SMC (newer macbooks) */
         /* Should be factorised with HDAPS */
-        f = fopen( "/sys/devices/applesmc.768/calibrate", "r" );
+        f = fopen( "/sys/devices/platform/applesmc.768/calibrate", "r" );
         if( f )
         {
             p_intf->p_sys->i_calibrate = fscanf( f, "(%d,%d)", &i_x, &i_y ) == 2 ? i_x: 0;
@@ -312,7 +312,7 @@ static int GetOrientation( intf_thread_t *p_intf )
             return - i_x * 30; /* FIXME: arbitrary */
 
     case APPLESMC_SENSOR:
-        f = fopen( "/sys/devices/applesmc.768/position", "r" );
+        f = fopen( "/sys/devices/platform/applesmc.768/position", "r" );
         if( !f )
         {
             return 0;
-- 
1.7.7.3


From rafael.carre at gmail.com  Thu Dec 29 05:59:49 2011
From: rafael.carre at gmail.com (=?ISO-8859-1?Q?Rafa=EBl_Carr=E9?=)
Date: Wed, 28 Dec 2011 23:59:49 -0500
Subject: [vlc-devel] [PATCH] Avcodec: limit the number of threads
 automatically selected to 12
In-Reply-To: <1325124481-30031-1-git-send-email-jb@videolan.org>
References: <1325124481-30031-1-git-send-email-jb@videolan.org>
Message-ID: <4EFBF3C5.1060407@gmail.com>

Le 2011-12-28 21:08, Jean-Baptiste Kempf a ?crit :
> It seems that over 16 doesn't behave well. Reduce it further for safety

What is the problem exactly ?

> ---
>  modules/codec/avcodec/video.c |    4 ++++
>  1 files changed, 4 insertions(+), 0 deletions(-)
> 
> diff --git a/modules/codec/avcodec/video.c b/modules/codec/avcodec/video.c
> index 95e64b5..dada16c 100644
> --- a/modules/codec/avcodec/video.c
> +++ b/modules/codec/avcodec/video.c
> @@ -344,7 +344,11 @@ int InitVideoDec( decoder_t *p_dec, AVCodecContext *p_context,
>  #ifdef HAVE_AVCODEC_MT
>      int i_thread_count = var_InheritInteger( p_dec, "ffmpeg-threads" );
>      if( i_thread_count <= 0 )
> +    {
>          i_thread_count = vlc_GetCPUCount();
> +        if( i_thread_count >= 12 )
> +            i_thread_count = 12;
> +    }

In any case there should be a comment explaining the limit

>      msg_Dbg( p_dec, "allowing %d thread(s) for decoding", i_thread_count );
>      p_sys->p_context->thread_count = i_thread_count;
>  #endif



From rafael.carre at gmail.com  Thu Dec 29 06:00:37 2011
From: rafael.carre at gmail.com (=?ISO-8859-1?Q?Rafa=EBl_Carr=E9?=)
Date: Thu, 29 Dec 2011 00:00:37 -0500
Subject: [vlc-devel] Athlon 64 CPU problem
In-Reply-To: <1325129763.63208.YahooMailNeo@web125705.mail.ne1.yahoo.com>
References: <1325129763.63208.YahooMailNeo@web125705.mail.ne1.yahoo.com>
Message-ID: <4EFBF3F5.8000804@gmail.com>

Le 2011-12-28 22:36, zhu shi song a ?crit :
>      Following the instructions on WEB
> http://wiki.videolan.org/Win32CompileMSYSNew#Precompiled_contribs, I
> have compiled VLC 1.1.5 successfully.

Hello,

You should build 1.2 or 1.3 from git


From kaarlo.raiha at gmail.com  Thu Dec 29 07:19:39 2011
From: kaarlo.raiha at gmail.com (=?ISO-8859-1?B?S2FhcmxvIFLkaWjk?=)
Date: Thu, 29 Dec 2011 08:19:39 +0200
Subject: [vlc-devel] Athlon 64 CPU problem
In-Reply-To: <1325129763.63208.YahooMailNeo@web125705.mail.ne1.yahoo.com>
References: <1325129763.63208.YahooMailNeo@web125705.mail.ne1.yahoo.com>
Message-ID: <CAMKrNtzyRhRz7a8JLAeDA2c6V8_EQuJjSSqpqsyqrOYcOHRgrQ@mail.gmail.com>

2011/12/29 zhu shi song <zhushisongzhu at yahoo.com>

>      Following the instructions on WEB
> http://wiki.videolan.org/Win32CompileMSYSNew#Precompiled_contribs, I have
> compiled VLC 1.1.5 successfully.  I used contrib-20100608-win32-bin-gcc-4.4.4-runtime-3.17-only.tar.bz2.
>
>
> <http://people.videolan.org/%7Ejb/Contribs/contrib-20100608-win32-bin-gcc-4.4.4-runtime-3.17-only.tar.bz2>
>      I use it on Intel CPU computer , it works well.  But I test it on AMD
> Athlon 64 CPU computer, it can't play MP4(h.264+AAC) smoothly.  The time
> progress bar is moving, but video is frozen.<http://people.videolan.org/%7Ejb/Contribs/contrib-20100608-win32-bin-gcc-4.4.4-runtime-3.17-only.tar.bz2>
>     The CPU details:   AMD Athlon 64x2 Dual Core Processor 4400+.
>

http://wiki.videolan.org/WindowsFAQ-1.1.x#H.264.2FMPEG-4_AVC_playback_is_too_slow


>
>
>
>
> <http://people.videolan.org/%7Ejb/Contribs/contrib-20100608-win32-bin-gcc-4.4.4-runtime-3.17-only.tar.bz2>
>
> _______________________________________________
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
> http://mailman.videolan.org/listinfo/vlc-devel
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111229/bd80816b/attachment.html>

From typx at dinauz.org  Thu Dec 29 09:31:07 2011
From: typx at dinauz.org (Denis Charmet)
Date: Thu, 29 Dec 2011 09:31:07 +0100
Subject: [vlc-devel] [PATCH] Avcodec: limit the number of
	threads	automatically selected to 12
In-Reply-To: <4EFBF3C5.1060407@gmail.com>
References: <1325124481-30031-1-git-send-email-jb@videolan.org>
	<4EFBF3C5.1060407@gmail.com>
Message-ID: <20111229083107.GA18287@dinauz.org>

Le mercredi 28 d?cembre 2011 ? 11:59:49, Rafa?l Carr? a ?crit :
> Le 2011-12-28 21:08, Jean-Baptiste Kempf a ?crit :
> > It seems that over 16 doesn't behave well. Reduce it further for safety
> 
> What is the problem exactly ?
> 
IMO, as I said earlier I strongly suspect a race condition in picture
pool with two threads working on the same picture. Would'nt it be
cleaner and safer to just lock/unlock around decoder_NewPicture in
avcodec/video.c ?

Regards,

-- 
TypX
Le mauvais esprit est un art de vivre


From funman at videolan.org  Thu Dec 29 09:40:06 2011
From: funman at videolan.org (=?ISO-8859-1?Q?Rafa=EBl_Carr=E9?=)
Date: Thu, 29 Dec 2011 03:40:06 -0500
Subject: [vlc-devel] [PATCH] Avcodec: limit the number of threads
 automatically selected to 12
In-Reply-To: <20111229083107.GA18287@dinauz.org>
References: <1325124481-30031-1-git-send-email-jb@videolan.org>
	<4EFBF3C5.1060407@gmail.com> <20111229083107.GA18287@dinauz.org>
Message-ID: <4EFC2766.2080509@videolan.org>

Le 11-12-29 03:31, Denis Charmet a ?crit :
> Le mercredi 28 d?cembre 2011 ? 11:59:49, Rafa?l Carr? a ?crit :
>> Le 2011-12-28 21:08, Jean-Baptiste Kempf a ?crit :
>>> It seems that over 16 doesn't behave well. Reduce it further for safety
>>
>> What is the problem exactly ?
>>
> IMO, as I said earlier I strongly suspect a race condition in picture
> pool with two threads working on the same picture. Would'nt it be
> cleaner and safer to just lock/unlock around decoder_NewPicture in
> avcodec/video.c ?

How does using 12 or 16 threads only help in that regard?

VLC shouldn't use more threads than the number of cores anyway.


From jb at videolan.org  Thu Dec 29 10:25:45 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Thu, 29 Dec 2011 10:25:45 +0100
Subject: [vlc-devel] [PATCH] Avcodec: limit the number of threads
 automatically selected to 12
In-Reply-To: <4EFBF3C5.1060407@gmail.com>
References: <1325124481-30031-1-git-send-email-jb@videolan.org>
	<4EFBF3C5.1060407@gmail.com>
Message-ID: <20111229092545.GA14468@videolan.org>

On Wed, Dec 28, 2011 at 11:59:49PM -0500, Rafa?l Carr? wrote :
> Le 2011-12-28 21:08, Jean-Baptiste Kempf a ?crit :
> > It seems that over 16 doesn't behave well. Reduce it further for safety
> 
> What is the problem exactly ?

According to ringo, it doesn't work at all with 24 threads. According to
the source, 16 threads is the maximum anyway.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From beauze.h at gmail.com  Thu Dec 29 10:36:15 2011
From: beauze.h at gmail.com (=?UTF-8?Q?Hugo_Beauz=C3=A9e=2DLuyssen?=)
Date: Thu, 29 Dec 2011 10:36:15 +0100
Subject: [vlc-devel] [PATCH] Avcodec: limit the number of threads
 automatically selected to 12
In-Reply-To: <20111229092545.GA14468@videolan.org>
References: <1325124481-30031-1-git-send-email-jb@videolan.org>
	<4EFBF3C5.1060407@gmail.com> <20111229092545.GA14468@videolan.org>
Message-ID: <CAK-C75QjdgLHfnJoFsXGUhU2eeucwh09M=jTwF38DWQtJVc5hA@mail.gmail.com>

On Thu, Dec 29, 2011 at 10:25 AM, Jean-Baptiste Kempf <jb at videolan.org> wrote:
> On Wed, Dec 28, 2011 at 11:59:49PM -0500, Rafa?l Carr? wrote :
>> Le 2011-12-28 21:08, Jean-Baptiste Kempf a ?crit :
>> > It seems that over 16 doesn't behave well. Reduce it further for safety
>>
>> What is the problem exactly ?
>
> According to ringo, it doesn't work at all with 24 threads. According to
> the source, 16 threads is the maximum anyway.
>
> Best regards,
>

Doesn't work for me with ffmpeg-threads > 5, on a quad core.

Regards,

-- 
Hugo Beauz?e-Luyssen


From fyhuel at viotech.net  Thu Dec 29 12:14:03 2011
From: fyhuel at viotech.net (=?UTF-8?q?Fr=C3=A9d=C3=A9ric=20Yhuel?=)
Date: Thu, 29 Dec 2011 12:14:03 +0100
Subject: [vlc-devel] [PATCH] httplive.c: fix hls_Read (when caller skips
	data)
Message-ID: <1325157243-24181-1-git-send-email-fyhuel@viotech.net>

I have also fixed a comment
---
 modules/stream_filter/httplive.c |    6 ++++--
 1 files changed, 4 insertions(+), 2 deletions(-)

diff --git a/modules/stream_filter/httplive.c b/modules/stream_filter/httplive.c
index 4d60d8a..47b99a1 100644
--- a/modules/stream_filter/httplive.c
+++ b/modules/stream_filter/httplive.c
@@ -78,7 +78,8 @@ typedef struct hls_stream_s
     int         sequence;   /* media sequence number */
     int         duration;   /* maximum duration per segment (s) */
     uint64_t    bandwidth;  /* bandwidth usage of segments (bits per second)*/
-    uint64_t    size;       /* stream length (segment->duration * hls->bandwidth/8) */
+    uint64_t    size;       /* stream length = Sum for all segments(
+                                    segment->duration * hls->bandwidth/8 ) */
 
     vlc_array_t *segments;  /* list of segments */
     vlc_url_t   url;        /* uri to m3u8 */
@@ -1912,7 +1913,8 @@ static ssize_t hls_Read(stream_t *s, uint8_t *p_read, unsigned int i_read)
 
         if (len > 0)
         {
-            memcpy(p_read + copied, segment->data->p_buffer, len);
+            if( p_read ) /* otherwise caller skips data */
+                memcpy(p_read + copied, segment->data->p_buffer, len);
             segment->data->i_buffer -= len;
             segment->data->p_buffer += len;
             copied += len;
-- 
1.7.5.4



From fenrir at elivagar.org  Thu Dec 29 16:24:44 2011
From: fenrir at elivagar.org (Laurent Aimar)
Date: Thu, 29 Dec 2011 16:24:44 +0100
Subject: [vlc-devel] [PATCH] Avcodec: limit the number of
	threads	automatically selected to 12
In-Reply-To: <20111229083107.GA18287@dinauz.org>
References: <1325124481-30031-1-git-send-email-jb@videolan.org>
	<4EFBF3C5.1060407@gmail.com> <20111229083107.GA18287@dinauz.org>
Message-ID: <20111229152444.GA6575@elivagar.org>

On Thu, Dec 29, 2011 at 09:31:07AM +0100, Denis Charmet wrote:
> Le mercredi 28 d?cembre 2011 ? 11:59:49, Rafa?l Carr? a ?crit :
> > Le 2011-12-28 21:08, Jean-Baptiste Kempf a ?crit :
> > > It seems that over 16 doesn't behave well. Reduce it further for safety
> > 
> > What is the problem exactly ?
> > 
> IMO, as I said earlier I strongly suspect a race condition in picture
> pool with two threads working on the same picture. Would'nt it be
> cleaner and safer to just lock/unlock around decoder_NewPicture in
> avcodec/video.c ?
 A decoder goes through vout_GetPicture() which already protects the
pool used with a lock.
 I think the issue is that the pool might be too small when there is
too much thread as the latency increase.
 For a quick test, could you test by increasing the value dpb_size
in src/input/decoder.c by the number of threads (in the function
vout_new_buffer()). If it fixes the issue, decoder_t will have to be
extended to allow transmitting the value.

Regards,

-- 
fenrir


From edward.c.wang at compdigitec.com  Thu Dec 29 19:15:41 2011
From: edward.c.wang at compdigitec.com (Edward Wang)
Date: Thu, 29 Dec 2011 13:15:41 -0500
Subject: [vlc-devel] [PATCH] Qt4: Port convert dialog to QWizard
Message-ID: <1325182541-11177-1-git-send-email-edward.c.wang@compdigitec.com>

Porting the wizard has a few other advantages, one of them being that it
makes the interface consistent with the recently updated stream out dialog, secondly
it shows the MRL and allows user to change it (just like stream out
dialog).

Thanks,
    Edward Wang


---
 modules/gui/qt4/Modules.am          |    8 +-
 modules/gui/qt4/dialogs/convert.cpp |  121 +++++------------
 modules/gui/qt4/dialogs/convert.hpp |   16 +--
 modules/gui/qt4/ui/convert.ui       |  240 +++++++++++++++++++++++++++++++++++
 4 files changed, 288 insertions(+), 97 deletions(-)
 create mode 100644 modules/gui/qt4/ui/convert.ui

diff --git a/modules/gui/qt4/Modules.am b/modules/gui/qt4/Modules.am
index 56510f5..988148a 100644
--- a/modules/gui/qt4/Modules.am
+++ b/modules/gui/qt4/Modules.am
@@ -99,7 +99,8 @@ nodist_SOURCES_qt4 = \
 		ui/update.h \
 		dialogs/ml_configuration.moc.cpp \
 		components/playlist/ml_model.moc.cpp \
-		ui/sout.h
+		ui/sout.h \
+		ui/convert.h

 DEPS_res = \
 	pixmaps/arrow_down_dark.png \
@@ -425,6 +426,7 @@ EXTRA_DIST += \
 	ui/about.ui \
 	ui/update.ui \
 	ui/sout.ui \
+	ui/convert.ui \
 	ui/vlm.ui \
 	$(DEPS_res)

diff --git a/modules/gui/qt4/dialogs/convert.cpp b/modules/gui/qt4/dialogs/convert.cpp
index 4bed9fe..24a0b4b 100644
--- a/modules/gui/qt4/dialogs/convert.cpp
+++ b/modules/gui/qt4/dialogs/convert.cpp
@@ -39,120 +39,71 @@

 ConvertDialog::ConvertDialog( QWidget *parent, intf_thread_t *_p_intf,
                               const QString& inputMRL )
-              : QVLCDialog( parent, _p_intf )
+              : QWizard( parent )
 {
+    p_intf = _p_intf;
+    mrl = "";
+
     setWindowTitle( qtr( "Convert" ) );
     setWindowRole( "vlc-convert" );

-    QGridLayout *mainLayout = new QGridLayout( this );
-    SoutInputBox *inputBox = new SoutInputBox( this );
-    inputBox->setMRL( inputMRL );
-    mainLayout->addWidget( inputBox, 0, 0, 1, -1  );
+    ui.setupUi( this );
+    ui.helpEdit->setPlainText( qtr("This wizard will allow you to convert "
+            "your media into different formats.\n"
+            "You should start by checking that source matches what you want "
+            "your input to be and then press the \"Next\" "
+            "button to continue.\n") );
+
+    ui.mrlEdit->setToolTip ( qtr( "Transcoding string.\n"
+                "This is automatically generated "
+                 "when you change the above settings,\n"
+                 "but you can change it manually." ) ) ;

     /**
      * Destination
      **/
-    QGroupBox *destBox = new QGroupBox( qtr( "Destination" ) );
-    QGridLayout *destLayout = new QGridLayout( destBox );
-
-    QLabel *destLabel = new QLabel( qtr( "Destination file:" ) );
-    destLayout->addWidget( destLabel, 0, 0);
-
-    fileLine = new QLineEdit;
-    fileLine->setMinimumWidth( 300 );
-    fileLine->setFocus( Qt::ActiveWindowFocusReason );
-    destLabel->setBuddy( fileLine );
-
-    QPushButton *fileSelectButton = new QPushButton( qtr( "Browse" ) );
-    destLayout->addWidget( fileLine, 0, 1 );
-    destLayout->addWidget( fileSelectButton, 0, 2);
-    BUTTONACT( fileSelectButton, fileBrowse() );
-
-    displayBox = new QCheckBox( qtr( "Display the output" ) );
-    displayBox->setToolTip( qtr( "This display the resulting media, but can "
-                               "slow things down." ) );
-    destLayout->addWidget( displayBox, 2, 0, 1, -1 );
-
-    mainLayout->addWidget( destBox, 1, 0, 1, -1  );
-
-
-    /* Profile Editor */
-    QGroupBox *settingBox = new QGroupBox( qtr( "Settings" ) );
-    QGridLayout *settingLayout = new QGridLayout( settingBox );
-
-    profile = new VLCProfileSelector( this );
-    settingLayout->addWidget( profile, 0, 0, 1, -1 );
-
-    deinterBox = new QCheckBox( qtr( "Deinterlace" ) );
-    settingLayout->addWidget( deinterBox, 1, 0 );
-
-    dumpBox = new QCheckBox( qtr( "Dump raw input" ) );
-    settingLayout->addWidget( dumpBox, 1, 1 );
+    ui.inputBox->setMRL( inputMRL );

-    mainLayout->addWidget( settingBox, 3, 0, 1, -1  );
+    ui.fileLine->setFocus( Qt::ActiveWindowFocusReason );
+    BUTTONACT( ui.destSelectButton, destBrowse() );
+    CONNECT( ui.displayCheckBox, clicked(), this, updateMRL() );
+    CONNECT( ui.displayCheckBox, clicked(), this, updateMRL() );
+    CONNECT( ui.deinterBox, clicked(), this, updateMRL() );
+    CONNECT( ui.dumpBox, clicked(), this, updateMRL() );

-    /* Buttons */
-    QPushButton *okButton = new QPushButton( qtr( "&Start" ) );
-    QPushButton *cancelButton = new QPushButton( qtr( "&Cancel" ) );
-    QDialogButtonBox *buttonBox = new QDialogButtonBox;
-
-    okButton->setDefault( true );
-    buttonBox->addButton( okButton, QDialogButtonBox::AcceptRole );
-    buttonBox->addButton( cancelButton, QDialogButtonBox::RejectRole );
-
-    mainLayout->addWidget( buttonBox, 5, 3 );
-
-    BUTTONACT(okButton,close());
-    BUTTONACT(cancelButton,cancel());
-
-    CONNECT(dumpBox,toggled(bool),this,dumpChecked(bool));
+    setButtonText( QWizard::FinishButton, qtr("Convert") );
 }

-void ConvertDialog::fileBrowse()
+void ConvertDialog::destBrowse()
 {
     QString fileName = QFileDialog::getSaveFileName( this, qtr( "Save file..." ),
             "",
  qtr( "Containers (*.ps *.ts *.mpg *.ogg *.asf *.mp4 *.mov *.wav *.raw *.flv *.webm)" ) );
-    fileLine->setText( toNativeSeparators( fileName ) );
-}
-
-void ConvertDialog::cancel()
-{
-    reject();
+    ui.fileLine->setText( toNativeSeparators( fileName ) );
+    updateMRL();
 }

-void ConvertDialog::close()
+void ConvertDialog::updateMRL()
 {
-    hide();
-
-    if( dumpBox->isChecked() )
+    if( ui.dumpBox->isChecked() )
     {
-        mrl = "demux=dump :demuxdump-file=" + fileLine->text();
+        mrl = "demux=dump :demuxdump-file=" + ui.fileLine->text();
     }
     else
     {
-        mrl = "sout=#" + profile->getTranscode();
-        if( deinterBox->isChecked() )
+        mrl = "sout=#" + ui.profile->getTranscode();
+        if( ui.deinterBox->isChecked() )
         {
             mrl.remove( '}' );
             mrl += ",deinterlace}";
         }
         mrl += ":";
-        if( displayBox->isChecked() )
+        if( ui.displayCheckBox->isChecked() )
             mrl += "duplicate{dst=display,dst=";
-        mrl += "std{access=file,mux=" + profile->getMux()
-             + ",dst='" + fileLine->text() + "'}";
-        if( displayBox->isChecked() )
+        mrl += "std{access=file,mux=" + ui.profile->getMux()
+             + ",dst='" + ui.fileLine->text() + "'}";
+        if( ui.displayCheckBox->isChecked() )
             mrl += "}";
     }
-
-    msg_Warn( p_intf, "Transcode MRL: %s", qtu( mrl ) );
-    accept();
-}
-
-void ConvertDialog::dumpChecked( bool checked )
-{
-    deinterBox->setEnabled( !checked );
-    displayBox->setEnabled( !checked );
-    profile->setEnabled( !checked );
+    ui.mrlEdit->setPlainText( mrl );
 }
diff --git a/modules/gui/qt4/dialogs/convert.hpp b/modules/gui/qt4/dialogs/convert.hpp
index de87981..753929c 100644
--- a/modules/gui/qt4/dialogs/convert.hpp
+++ b/modules/gui/qt4/dialogs/convert.hpp
@@ -24,13 +24,14 @@
 #ifndef QVLC_CONVERT_DIALOG_H_
 #define QVLC_CONVERT_DIALOG_H_ 1

+#include "ui/convert.h"
 #include "util/qvlcframe.hpp"

 class QLineEdit;
 class QCheckBox;
 class VLCProfileSelector;

-class ConvertDialog : public QVLCDialog
+class ConvertDialog : public QWizard
 {
     Q_OBJECT
 public:
@@ -40,16 +41,13 @@ public:
     QString getMrl() {return mrl;}

 private:
-    QLineEdit *fileLine;
-
-    QCheckBox *displayBox, *deinterBox, *dumpBox;
-    VLCProfileSelector *profile;
+    Ui::Convert ui;
+    intf_thread_t* p_intf;
     QString mrl;
+
 private slots:
-    virtual void close();
-    virtual void cancel();
-    void fileBrowse();
-    void dumpChecked(bool);
+    virtual void updateMRL();
+    void destBrowse();
 };

 #endif
diff --git a/modules/gui/qt4/ui/convert.ui b/modules/gui/qt4/ui/convert.ui
new file mode 100644
index 0000000..0c43b18
--- /dev/null
+++ b/modules/gui/qt4/ui/convert.ui
@@ -0,0 +1,240 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<ui version="4.0">
+ <author>Edward Wang</author>
+ <class>Convert</class>
+ <widget class="QWizard" name="Convert">
+  <property name="geometry">
+   <rect>
+    <x>0</x>
+    <y>0</y>
+    <width>529</width>
+    <height>344</height>
+   </rect>
+  </property>
+  <property name="minimumSize">
+   <size>
+    <width>400</width>
+    <height>200</height>
+   </size>
+  </property>
+  <property name="windowTitle">
+   <string>Stream Output</string>
+  </property>
+  <property name="wizardStyle">
+   <enum>QWizard::ClassicStyle</enum>
+  </property>
+  <widget class="QWizardPage" name="sourceWizardPage">
+   <property name="title">
+    <string>Source</string>
+   </property>
+   <property name="subTitle">
+    <string>Set up media source to convert</string>
+   </property>
+   <layout class="QGridLayout" name="gridLayout">
+    <item row="0" column="0">
+     <layout class="QVBoxLayout" name="verticalLayout">
+      <item>
+       <widget class="QPlainTextEdit" name="helpEdit">
+        <property name="readOnly">
+         <bool>true</bool>
+        </property>
+       </widget>
+      </item>
+      <item>
+       <widget class="SoutInputBox" name="inputBox">
+        <property name="title">
+         <string/>
+        </property>
+        <property name="flat">
+         <bool>true</bool>
+        </property>
+        <property name="checkable">
+         <bool>false</bool>
+        </property>
+       </widget>
+      </item>
+     </layout>
+    </item>
+   </layout>
+  </widget>
+  <widget class="QWizardPage" name="transcodeWizardPage">
+   <property name="title">
+    <string>Transcoding Options</string>
+   </property>
+   <property name="subTitle">
+    <string>Select and choose transcoding profile</string>
+   </property>
+   <layout class="QVBoxLayout" name="verticalLayout_3">
+    <item>
+     <widget class="VLCProfileSelector" name="profile" native="true"/>
+    </item>
+    <item>
+     <layout class="QHBoxLayout" name="horizontalLayout_2">
+      <item>
+       <widget class="QCheckBox" name="deinterBox">
+        <property name="text">
+         <string>De-interlace</string>
+        </property>
+       </widget>
+      </item>
+      <item>
+       <widget class="QCheckBox" name="dumpBox">
+        <property name="toolTip">
+         <string>Copy the source to the destination with no processing</string>
+        </property>
+        <property name="text">
+         <string>Dump raw input</string>
+        </property>
+       </widget>
+      </item>
+     </layout>
+    </item>
+   </layout>
+  </widget>
+  <widget class="QWizardPage" name="destinationWizardPage">
+   <property name="title">
+    <string>Destination Setup</string>
+   </property>
+   <property name="subTitle">
+    <string>Select destination file</string>
+   </property>
+   <layout class="QVBoxLayout" name="verticalLayout_7">
+    <item>
+     <layout class="QVBoxLayout" name="verticalLayout_6">
+      <item>
+       <layout class="QHBoxLayout" name="horizontalLayout">
+        <property name="sizeConstraint">
+         <enum>QLayout::SetDefaultConstraint</enum>
+        </property>
+        <item>
+         <widget class="QLineEdit" name="fileLine"/>
+        </item>
+        <item>
+         <widget class="QPushButton" name="destSelectButton">
+          <property name="text">
+           <string>Browse...</string>
+          </property>
+         </widget>
+        </item>
+       </layout>
+      </item>
+      <item>
+       <widget class="QCheckBox" name="displayCheckBox">
+        <property name="toolTip">
+         <string>Display the media during conversion, but could cause the conversion to take longer.</string>
+        </property>
+        <property name="text">
+         <string>Display the output</string>
+        </property>
+       </widget>
+      </item>
+      <item>
+       <spacer name="verticalSpacer">
+        <property name="orientation">
+         <enum>Qt::Vertical</enum>
+        </property>
+        <property name="sizeHint" stdset="0">
+         <size>
+          <width>20</width>
+          <height>40</height>
+         </size>
+        </property>
+       </spacer>
+      </item>
+     </layout>
+    </item>
+   </layout>
+  </widget>
+  <widget class="QWizardPage" name="optionsWizardPage">
+   <property name="title">
+    <string>Review</string>
+   </property>
+   <property name="subTitle">
+    <string>Review generated transcoding string</string>
+   </property>
+   <layout class="QVBoxLayout" name="verticalLayout_2">
+    <item>
+     <widget class="QGroupBox" name="groupBox_4">
+      <property name="title">
+       <string>Generated stream output string</string>
+      </property>
+      <layout class="QVBoxLayout" name="verticalLayout_5">
+       <item>
+        <widget class="QTextEdit" name="mrlEdit">
+         <property name="acceptRichText">
+          <bool>false</bool>
+         </property>
+        </widget>
+       </item>
+      </layout>
+     </widget>
+    </item>
+   </layout>
+  </widget>
+ </widget>
+ <customwidgets>
+  <customwidget>
+   <class>VLCProfileSelector</class>
+   <extends>QWidget</extends>
+   <header>components/sout/profile_selector.hpp</header>
+   <container>1</container>
+  </customwidget>
+  <customwidget>
+   <class>SoutInputBox</class>
+   <extends>QGroupBox</extends>
+   <header>components/sout/sout_widgets.hpp</header>
+   <container>1</container>
+  </customwidget>
+ </customwidgets>
+ <resources/>
+ <connections>
+  <connection>
+   <sender>dumpBox</sender>
+   <signal>toggled(bool)</signal>
+   <receiver>profile</receiver>
+   <slot>setDisabled(bool)</slot>
+   <hints>
+    <hint type="sourcelabel">
+     <x>386</x>
+     <y>109</y>
+    </hint>
+    <hint type="destinationlabel">
+     <x>264</x>
+     <y>86</y>
+    </hint>
+   </hints>
+  </connection>
+  <connection>
+   <sender>dumpBox</sender>
+   <signal>toggled(bool)</signal>
+   <receiver>deinterBox</receiver>
+   <slot>setDisabled(bool)</slot>
+   <hints>
+    <hint type="sourcelabel">
+     <x>386</x>
+     <y>109</y>
+    </hint>
+    <hint type="destinationlabel">
+     <x>142</x>
+     <y>109</y>
+    </hint>
+   </hints>
+  </connection>
+  <connection>
+   <sender>dumpBox</sender>
+   <signal>toggled(bool)</signal>
+   <receiver>displayCheckBox</receiver>
+   <slot>setDisabled(bool)</slot>
+   <hints>
+    <hint type="sourcelabel">
+     <x>386</x>
+     <y>109</y>
+    </hint>
+    <hint type="destinationlabel">
+     <x>264</x>
+     <y>128</y>
+    </hint>
+   </hints>
+  </connection>
+ </connections>
+</ui>
--
1.7.5.4



From funman at videolan.org  Thu Dec 29 21:07:20 2011
From: funman at videolan.org (=?UTF-8?B?UmFmYcOrbCBDYXJyw6k=?=)
Date: Thu, 29 Dec 2011 15:07:20 -0500
Subject: [vlc-devel] [PATCH] Avcodec: limit the number of threads
 automatically selected to 12
In-Reply-To: <CAK-C75QjdgLHfnJoFsXGUhU2eeucwh09M=jTwF38DWQtJVc5hA@mail.gmail.com>
References: <1325124481-30031-1-git-send-email-jb@videolan.org>
	<4EFBF3C5.1060407@gmail.com> <20111229092545.GA14468@videolan.org>
	<CAK-C75QjdgLHfnJoFsXGUhU2eeucwh09M=jTwF38DWQtJVc5hA@mail.gmail.com>
Message-ID: <4EFCC878.5030203@videolan.org>

Le 11-12-29 04:36, Hugo Beauz?e-Luyssen a ?crit :
> On Thu, Dec 29, 2011 at 10:25 AM, Jean-Baptiste Kempf <jb at videolan.org> wrote:
>> On Wed, Dec 28, 2011 at 11:59:49PM -0500, Rafa?l Carr? wrote :
>>> Le 2011-12-28 21:08, Jean-Baptiste Kempf a ?crit :
>>>> It seems that over 16 doesn't behave well. Reduce it further for safety
>>>
>>> What is the problem exactly ?
>>
>> According to ringo, it doesn't work at all with 24 threads. According to
>> the source, 16 threads is the maximum anyway.
>>
>> Best regards,
>>
> 
> Doesn't work for me with ffmpeg-threads > 5, on a quad core.

5 does work?

What is the error, the assertion that Denis reported?

IIRC ringo had a different problem and he couldn't give much details


From edward.c.wang at compdigitec.com  Thu Dec 29 21:08:29 2011
From: edward.c.wang at compdigitec.com (Edward Wang)
Date: Thu, 29 Dec 2011 15:08:29 -0500
Subject: [vlc-devel] [PATCH] Qt4: Enable tabs in sout dialog
Message-ID: <1325189309-14244-1-git-send-email-edward.c.wang@compdigitec.com>

This patch will enable tabs in the sout dialog if possible (given the right Qt version).
Thanks,
	Edward Wang

---
 modules/gui/qt4/Modules.am       |    2 ++
 modules/gui/qt4/dialogs/sout.cpp |   13 ++++++++-----
 modules/gui/qt4/dialogs/sout.hpp |    5 ++++-
 modules/gui/qt4/ui/sout.ui       |    8 +++++++-
 4 files changed, 21 insertions(+), 7 deletions(-)

diff --git a/modules/gui/qt4/Modules.am b/modules/gui/qt4/Modules.am
index 988148a..57439a7 100644
--- a/modules/gui/qt4/Modules.am
+++ b/modules/gui/qt4/Modules.am
@@ -73,6 +73,7 @@ nodist_SOURCES_qt4 = \
 		util/searchlineedit.moc.cpp \
 		util/qmenuview.moc.cpp \
 		util/qvlcapp.moc.cpp \
+		util/qvlctabwidget.moc.cpp \
 		util/pictureflow.moc.cpp \
 		util/buttons/RoundButton.moc.cpp \
 		util/buttons/DeckButtonsLayout.moc.cpp \
@@ -395,6 +396,7 @@ noinst_HEADERS = \
 	util/searchlineedit.hpp \
 	util/qvlcframe.hpp \
 	util/qvlcapp.hpp \
+	util/qvlctabwidget.hpp \
 	util/qmenuview.hpp \
 	util/qt_dirs.hpp \
 	util/registry.hpp \
diff --git a/modules/gui/qt4/dialogs/sout.cpp b/modules/gui/qt4/dialogs/sout.cpp
index d4a6a1e..8b5b65a 100644
--- a/modules/gui/qt4/dialogs/sout.cpp
+++ b/modules/gui/qt4/dialogs/sout.cpp
@@ -35,6 +35,7 @@
 #include <QFileDialog>
 #include <QToolButton>
 #include <QSpinBox>
+#include <QtGlobal>
 #include <assert.h>
 
 SoutDialog::SoutDialog( QWidget *parent, intf_thread_t *_p_intf, const QString& inputMRL )
@@ -60,10 +61,12 @@ SoutDialog::SoutDialog( QWidget *parent, intf_thread_t *_p_intf, const QString&
                  "when you change the above settings,\n"
                  "but you can change it manually." ) ) ;
 
-#if 0
+#if QT_VERSION >= 0x040500
     /* This needs Qt4.5 to be cool */
     ui.destTab->setTabsClosable( true );
+    ui.destTab->getTabBar()->tabButton(0, QTabBar::RightSide)->hide();
     CONNECT( ui.destTab, tabCloseRequested( int ), this, closeTab( int ) );
+    closeTabButton = NULL;
 #else
     closeTabButton = new QToolButton( this );
     ui.destTab->setCornerWidget( closeTabButton );
@@ -108,15 +111,14 @@ SoutDialog::SoutDialog( QWidget *parent, intf_thread_t *_p_intf, const QString&
 
 void SoutDialog::tabChanged( int i )
 {
-    closeTabButton->setVisible( (i != 0) );
+    if( closeTabButton ) closeTabButton->setVisible( (i != 0) );
 }
 
-void SoutDialog::closeTab()
+void SoutDialog::closeTab( int i )
 {
-    int i = ui.destTab->currentIndex();
     if( i == 0 ) return;
 
-    QWidget *temp = ui.destTab->currentWidget();
+    QWidget* temp = ui.destTab->widget( i );
     ui.destTab->removeTab( i );
     delete temp;
     updateMRL();
@@ -163,6 +165,7 @@ void SoutDialog::addDest( )
             break;
         default:
             assert(0);
+            return;
     }
 
     int index = ui.destTab->addTab( db, caption );
diff --git a/modules/gui/qt4/dialogs/sout.hpp b/modules/gui/qt4/dialogs/sout.hpp
index b15c1ee..43375db 100644
--- a/modules/gui/qt4/dialogs/sout.hpp
+++ b/modules/gui/qt4/dialogs/sout.hpp
@@ -135,7 +135,10 @@ public slots:
 private slots:
     void ok();
     void cancel();
-    void closeTab();
+    void closeTab() {
+        closeTab( ui.destTab->currentIndex() );
+    }
+    void closeTab( int );
     void tabChanged( int );
     void addDest();
 };
diff --git a/modules/gui/qt4/ui/sout.ui b/modules/gui/qt4/ui/sout.ui
index 33b835f..84a8e21 100644
--- a/modules/gui/qt4/ui/sout.ui
+++ b/modules/gui/qt4/ui/sout.ui
@@ -66,7 +66,7 @@
    </property>
    <layout class="QVBoxLayout" name="verticalLayout_4">
     <item>
-     <widget class="QTabWidget" name="destTab">
+     <widget class="QVLCTabWidget" name="destTab">
       <property name="currentIndex">
        <number>0</number>
       </property>
@@ -243,6 +243,12 @@
    <header>components/sout/sout_widgets.hpp</header>
    <container>1</container>
   </customwidget>
+  <customwidget>
+   <class>QVLCTabWidget</class>
+   <extends>QTabWidget</extends>
+   <header>util/qvlctabwidget.hpp</header>
+   <container>1</container>
+  </customwidget>
  </customwidgets>
  <resources/>
  <connections/>
-- 
1.7.5.4



From rafael.carre at gmail.com  Thu Dec 29 21:10:12 2011
From: rafael.carre at gmail.com (=?ISO-8859-1?Q?Rafa=EBl_Carr=E9?=)
Date: Thu, 29 Dec 2011 15:10:12 -0500
Subject: [vlc-devel] [PATCH] Qt4: Enable tabs in sout dialog
In-Reply-To: <1325189309-14244-1-git-send-email-edward.c.wang@compdigitec.com>
References: <1325189309-14244-1-git-send-email-edward.c.wang@compdigitec.com>
Message-ID: <4EFCC924.7010809@gmail.com>

Le 2011-12-29 15:08, Edward Wang a ?crit :
> This patch will enable tabs in the sout dialog if possible (given the right Qt version).
> Thanks,

> @@ -163,6 +165,7 @@ void SoutDialog::addDest( )
>              break;
>          default:
>              assert(0);
> +            return;
>      }
>  
>      int index = ui.destTab->addTab( db, caption );

Looks unrelated to the patch


From edward.c.wang at compdigitec.com  Thu Dec 29 21:12:22 2011
From: edward.c.wang at compdigitec.com (Edward Wang)
Date: Thu, 29 Dec 2011 15:12:22 -0500
Subject: [vlc-devel] [PATCH] Qt4: Enable tabs in sout dialog
In-Reply-To: <4EFCC924.7010809@gmail.com>
References: <1325189309-14244-1-git-send-email-edward.c.wang@compdigitec.com>
	<4EFCC924.7010809@gmail.com>
Message-ID: <4EFCC9A6.5070507@compdigitec.com>

On 11-12-29 03:10 PM, Rafa?l Carr? wrote:
> Le 2011-12-29 15:08, Edward Wang a ?crit :
>> This patch will enable tabs in the sout dialog if possible (given the right Qt version).
>> Thanks,
>> @@ -163,6 +165,7 @@ void SoutDialog::addDest( )
>>               break;
>>           default:
>>               assert(0);
>> +            return;
>>       }
>>
>>       int index = ui.destTab->addTab( db, caption );
> Looks unrelated to the patch
>
Well, it prevents a warning about use of an uninitialised variable (db). 
It can be removed if really bothers you that much.

Thanks,
         Edward Wang


From rafael.carre at gmail.com  Thu Dec 29 21:14:27 2011
From: rafael.carre at gmail.com (=?ISO-8859-1?Q?Rafa=EBl_Carr=E9?=)
Date: Thu, 29 Dec 2011 15:14:27 -0500
Subject: [vlc-devel] [PATCH] Qt4: Enable tabs in sout dialog
In-Reply-To: <4EFCC9A6.5070507@compdigitec.com>
References: <1325189309-14244-1-git-send-email-edward.c.wang@compdigitec.com>
	<4EFCC924.7010809@gmail.com> <4EFCC9A6.5070507@compdigitec.com>
Message-ID: <4EFCCA23.8000906@gmail.com>

Le 2011-12-29 15:12, Edward Wang a ?crit :
> On 11-12-29 03:10 PM, Rafa?l Carr? wrote:
>> Le 2011-12-29 15:08, Edward Wang a ?crit :
>>> This patch will enable tabs in the sout dialog if possible (given the right Qt version).
>>> Thanks,
>>> @@ -163,6 +165,7 @@ void SoutDialog::addDest( )
>>>               break;
>>>           default:
>>>               assert(0);
>>> +            return;
>>>       }
>>>
>>>       int index = ui.destTab->addTab( db, caption );
>> Looks unrelated to the patch
>>
> Well, it prevents a warning about use of an uninitialised variable (db). 
> It can be removed if really bothers you that much.

I am not sure if it really fixes that warning.

Anyway you should have mentioned it in the commit log or made 2 patches;
and I think you should just remove it


From edward.c.wang at compdigitec.com  Thu Dec 29 21:22:49 2011
From: edward.c.wang at compdigitec.com (Edward Wang)
Date: Thu, 29 Dec 2011 15:22:49 -0500
Subject: [vlc-devel] [PATCH] Qt4: Enable tabs in sout dialog
In-Reply-To: <4EFCCA23.8000906@gmail.com>
References: <1325189309-14244-1-git-send-email-edward.c.wang@compdigitec.com>
	<4EFCC924.7010809@gmail.com> <4EFCC9A6.5070507@compdigitec.com>
	<4EFCCA23.8000906@gmail.com>
Message-ID: <4EFCCC19.8030302@compdigitec.com>

On 11-12-29 03:14 PM, Rafa?l Carr? wrote:
> I am not sure if it really fixes that warning. Anyway you should have 
> mentioned it in the commit log or made 2 patches; and I think you 
> should just remove it 
Amended.

---
  modules/gui/qt4/Modules.am       |    2 ++
  modules/gui/qt4/dialogs/sout.cpp |   12 +++++++-----
  modules/gui/qt4/dialogs/sout.hpp |    5 ++++-
  modules/gui/qt4/ui/sout.ui       |    8 +++++++-
  4 files changed, 20 insertions(+), 7 deletions(-)

diff --git a/modules/gui/qt4/Modules.am b/modules/gui/qt4/Modules.am
index 988148a..57439a7 100644
--- a/modules/gui/qt4/Modules.am
+++ b/modules/gui/qt4/Modules.am
@@ -73,6 +73,7 @@ nodist_SOURCES_qt4 = \
          util/searchlineedit.moc.cpp \
          util/qmenuview.moc.cpp \
          util/qvlcapp.moc.cpp \
+        util/qvlctabwidget.moc.cpp \
          util/pictureflow.moc.cpp \
          util/buttons/RoundButton.moc.cpp \
          util/buttons/DeckButtonsLayout.moc.cpp \
@@ -395,6 +396,7 @@ noinst_HEADERS = \
      util/searchlineedit.hpp \
      util/qvlcframe.hpp \
      util/qvlcapp.hpp \
+    util/qvlctabwidget.hpp \
      util/qmenuview.hpp \
      util/qt_dirs.hpp \
      util/registry.hpp \
diff --git a/modules/gui/qt4/dialogs/sout.cpp 
b/modules/gui/qt4/dialogs/sout.cpp
index d4a6a1e..13e107f 100644
--- a/modules/gui/qt4/dialogs/sout.cpp
+++ b/modules/gui/qt4/dialogs/sout.cpp
@@ -35,6 +35,7 @@
  #include <QFileDialog>
  #include <QToolButton>
  #include <QSpinBox>
+#include <QtGlobal>
  #include <assert.h>

  SoutDialog::SoutDialog( QWidget *parent, intf_thread_t *_p_intf, const 
QString& inputMRL )
@@ -60,10 +61,12 @@ SoutDialog::SoutDialog( QWidget *parent, 
intf_thread_t *_p_intf, const QString&
                   "when you change the above settings,\n"
                   "but you can change it manually." ) ) ;

-#if 0
+#if QT_VERSION >= 0x040500
      /* This needs Qt4.5 to be cool */
      ui.destTab->setTabsClosable( true );
+    ui.destTab->getTabBar()->tabButton(0, QTabBar::RightSide)->hide();
      CONNECT( ui.destTab, tabCloseRequested( int ), this, closeTab( int 
) );
+    closeTabButton = NULL;
  #else
      closeTabButton = new QToolButton( this );
      ui.destTab->setCornerWidget( closeTabButton );
@@ -108,15 +111,14 @@ SoutDialog::SoutDialog( QWidget *parent, 
intf_thread_t *_p_intf, const QString&

  void SoutDialog::tabChanged( int i )
  {
-    closeTabButton->setVisible( (i != 0) );
+    if( closeTabButton ) closeTabButton->setVisible( (i != 0) );
  }

-void SoutDialog::closeTab()
+void SoutDialog::closeTab( int i )
  {
-    int i = ui.destTab->currentIndex();
      if( i == 0 ) return;

-    QWidget *temp = ui.destTab->currentWidget();
+    QWidget* temp = ui.destTab->widget( i );
      ui.destTab->removeTab( i );
      delete temp;
      updateMRL();
diff --git a/modules/gui/qt4/dialogs/sout.hpp 
b/modules/gui/qt4/dialogs/sout.hpp
index b15c1ee..43375db 100644
--- a/modules/gui/qt4/dialogs/sout.hpp
+++ b/modules/gui/qt4/dialogs/sout.hpp
@@ -135,7 +135,10 @@ public slots:
  private slots:
      void ok();
      void cancel();
-    void closeTab();
+    void closeTab() {
+        closeTab( ui.destTab->currentIndex() );
+    }
+    void closeTab( int );
      void tabChanged( int );
      void addDest();
  };
diff --git a/modules/gui/qt4/ui/sout.ui b/modules/gui/qt4/ui/sout.ui
index 33b835f..84a8e21 100644
--- a/modules/gui/qt4/ui/sout.ui
+++ b/modules/gui/qt4/ui/sout.ui
@@ -66,7 +66,7 @@
</property>
<layout class="QVBoxLayout" name="verticalLayout_4">
<item>
- <widget class="QTabWidget" name="destTab">
+ <widget class="QVLCTabWidget" name="destTab">
<property name="currentIndex">
<number>0</number>
</property>
@@ -243,6 +243,12 @@
<header>components/sout/sout_widgets.hpp</header>
<container>1</container>
</customwidget>
+ <customwidget>
+ <class>QVLCTabWidget</class>
+ <extends>QTabWidget</extends>
+ <header>util/qvlctabwidget.hpp</header>
+ <container>1</container>
+ </customwidget>
</customwidgets>
<resources/>
<connections/>


From sergio at ammirata.net  Thu Dec 29 21:25:11 2011
From: sergio at ammirata.net (Sergio M. Ammirata, Ph.D.)
Date: Thu, 29 Dec 2011 15:25:11 -0500
Subject: [vlc-devel] [PATCH] Avcodec: limit the number of threads
 automatically selected to 12
In-Reply-To: <4EFCC878.5030203@videolan.org>
Message-ID: <CB223586.1F25E%sergio@ammirata.net>

Just to clarify:

I just retested on a 2 CPU machine (core i7, 6 cores each with
multithreading enabled) which basically gives me 24 cores listed under
/proc/cpuinfo.

When I try to transcode mpeg2 to h264, I get this error repeating over and
over again in the debug log:

0x80f65f0] avcodec generic warning: cannot decode one frame (36036 bytes)
too many threads
mpeg_decode_postinit() failure
[0x80f65f0] avcodec generic warning: cannot decode one frame (2920 bytes)
too many threads
mpeg_decode_postinit() failure
[0x80f65f0] avcodec generic warning: cannot decode one frame (6972 bytes)
too many threads
mpeg_decode_postinit() failure


If I use --ffmpeg-threads to limit the number of threads then it works
according to this table:

Using --ffmpeg-threads=16 or less, it works.
Using --ffmpeg-threads=17 or more, it does not work and I get the error
listed above.

This is consistent with J-B's statement regarding ffmpeg limit of 16
threads.

Sergio

On 12/29/11 3:07 PM, "Rafa?l Carr?" <funman at videolan.org> wrote:

>Le 11-12-29 04:36, Hugo Beauz?e-Luyssen a ?crit :
>> On Thu, Dec 29, 2011 at 10:25 AM, Jean-Baptiste Kempf <jb at videolan.org>
>>wrote:
>>> On Wed, Dec 28, 2011 at 11:59:49PM -0500, Rafa?l Carr? wrote :
>>>> Le 2011-12-28 21:08, Jean-Baptiste Kempf a ?crit :
>>>>> It seems that over 16 doesn't behave well. Reduce it further for
>>>>>safety
>>>>
>>>> What is the problem exactly ?
>>>
>>> According to ringo, it doesn't work at all with 24 threads. According
>>>to
>>> the source, 16 threads is the maximum anyway.
>>>
>>> Best regards,
>>>
>> 
>> Doesn't work for me with ffmpeg-threads > 5, on a quad core.
>
>5 does work?
>
>What is the error, the assertion that Denis reported?
>
>IIRC ringo had a different problem and he couldn't give much details
>_______________________________________________
>vlc-devel mailing list
>To unsubscribe or modify your subscription options:
>http://mailman.videolan.org/listinfo/vlc-devel




From rafael.carre at gmail.com  Thu Dec 29 21:36:26 2011
From: rafael.carre at gmail.com (=?ISO-8859-1?Q?Rafa=EBl_Carr=E9?=)
Date: Thu, 29 Dec 2011 15:36:26 -0500
Subject: [vlc-devel] [PATCH] Avcodec: limit the number of threads
 automatically selected to 12
In-Reply-To: <CB223586.1F25E%sergio@ammirata.net>
References: <CB223586.1F25E%sergio@ammirata.net>
Message-ID: <4EFCCF4A.50908@gmail.com>

Le 2011-12-29 15:25, Sergio M. Ammirata, Ph.D. a ?crit :
> Just to clarify:
> 
> I just retested on a 2 CPU machine (core i7, 6 cores each with
> multithreading enabled) which basically gives me 24 cores listed under
> /proc/cpuinfo.
> 
> When I try to transcode mpeg2 to h264, I get this error repeating over and
> over again in the debug log:
> 
> 0x80f65f0] avcodec generic warning: cannot decode one frame (36036 bytes)
> too many threads
> mpeg_decode_postinit() failure
> [0x80f65f0] avcodec generic warning: cannot decode one frame (2920 bytes)
> too many threads
> mpeg_decode_postinit() failure
> [0x80f65f0] avcodec generic warning: cannot decode one frame (6972 bytes)
> too many threads
> mpeg_decode_postinit() failure
> 
> 
> If I use --ffmpeg-threads to limit the number of threads then it works
> according to this table:
> 
> Using --ffmpeg-threads=16 or less, it works.
> Using --ffmpeg-threads=17 or more, it does not work and I get the error
> listed above.
> 
> This is consistent with J-B's statement regarding ffmpeg limit of 16
> threads.

Yeah I think the limit should be 16 if libav/ffmpeg doesn't change it
itself.

The problems with threads > number of CPU threads is something else, I
think Laurent gave a way to investigate/solve it.

You don't see it in your build (assertion failure when releasing
pictures) because you commented/worked around that problem in your
build, right?


From Christoph.Miebach at web.de  Thu Dec 29 21:49:03 2011
From: Christoph.Miebach at web.de (Christoph Miebach)
Date: Thu, 29 Dec 2011 21:49:03 +0100 (CET)
Subject: [vlc-devel] Typo
Message-ID: <401492634.1466568.1325191743910.JavaMail.fmail@mwmweb023>

Hello!

Another one in vlc 1.2.0-pre3, POT-Creation-Date: 2011-12-28 03:13+0100

#: modules/demux/mp4/mp4.c:967
msgid "Composr"
=>
"Composer"


When do we have to deliver the translations for 1.2.0?


I would swap "horizontally/vertically" here:
tools->filters->video->basic->transform->
flip horizontally/vertically
But that's just me looking at vlc 1.1.12 ;-)

Regards

Christoph


From edward.c.wang at compdigitec.com  Thu Dec 29 21:52:07 2011
From: edward.c.wang at compdigitec.com (Edward Wang)
Date: Thu, 29 Dec 2011 15:52:07 -0500
Subject: [vlc-devel] [PATCH] Re:  Typo
In-Reply-To: <401492634.1466568.1325191743910.JavaMail.fmail@mwmweb023>
References: <401492634.1466568.1325191743910.JavaMail.fmail@mwmweb023>
Message-ID: <4EFCD2F7.4020309@compdigitec.com>

On 11-12-29 03:49 PM, Christoph Miebach wrote:
> Hello!
>
> Another one in vlc 1.2.0-pre3, POT-Creation-Date: 2011-12-28 03:13+0100
>
> #: modules/demux/mp4/mp4.c:967
> msgid "Composr"
> =>
> "Composer"
>
---
  modules/demux/mp4/mp4.c |    2 +-
  1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/modules/demux/mp4/mp4.c b/modules/demux/mp4/mp4.c
index e9acee3..96b2c66 100644
--- a/modules/demux/mp4/mp4.c
+++ b/modules/demux/mp4/mp4.c
@@ -964,7 +964,7 @@ static int Control( demux_t *p_demux, int i_query, 
va_list args )
                  static const struct { uint32_t xa9_type; char 
metadata[25]; } xa9typetoextrameta[] =
                  {
                      { ATOM_0xa9wrt, N_("Writer") },
-                    { ATOM_0xa9com, N_("Composr") },
+                    { ATOM_0xa9com, N_("Composer") },
                      { ATOM_0xa9prd, N_("Producer") },
                      { ATOM_0xa9inf, N_("Information") },
                      { ATOM_0xa9dir, N_("Director") },



From jb at videolan.org  Thu Dec 29 23:27:11 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Thu, 29 Dec 2011 23:27:11 +0100
Subject: [vlc-devel] [PATCH] Qt4: Enable tabs in sout dialog
In-Reply-To: <1325189309-14244-1-git-send-email-edward.c.wang@compdigitec.com>
References: <1325189309-14244-1-git-send-email-edward.c.wang@compdigitec.com>
Message-ID: <20111229222711.GB507@videolan.org>

On Thu, Dec 29, 2011 at 03:08:29PM -0500, Edward Wang wrote :
> This patch will enable tabs in the sout dialog if possible (given the right Qt version).
> Thanks,

I already dismissed this kind of patch because it added the close on the
first tab. But it seems you found a way..

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From edward.c.wang at compdigitec.com  Thu Dec 29 23:45:01 2011
From: edward.c.wang at compdigitec.com (Edward Wang)
Date: Thu, 29 Dec 2011 17:45:01 -0500
Subject: [vlc-devel] [PATCH] Qt4: Enable tabs in sout dialog
In-Reply-To: <20111229222711.GB507@videolan.org>
References: <1325189309-14244-1-git-send-email-edward.c.wang@compdigitec.com>
	<20111229222711.GB507@videolan.org>
Message-ID: <4EFCED6D.5090304@compdigitec.com>

On 11-12-29 05:27 PM, Jean-Baptiste Kempf wrote:
> I already dismissed this kind of patch because it added the close on 
> the first tab. But it seems you found a way..
Sorry, the previous patch had missed a file, you can now safely commit 
this patch...

---
  modules/gui/qt4/Modules.am             |    2 +
  modules/gui/qt4/dialogs/sout.cpp       |   12 +++++----
  modules/gui/qt4/dialogs/sout.hpp       |    5 +++-
  modules/gui/qt4/dialogs_provider.cpp   |    2 +-
  modules/gui/qt4/ui/sout.ui             |    8 +++++-
  modules/gui/qt4/util/qvlctabwidget.hpp |   41 
++++++++++++++++++++++++++++++++
  6 files changed, 62 insertions(+), 8 deletions(-)
  create mode 100644 modules/gui/qt4/util/qvlctabwidget.hpp

diff --git a/modules/gui/qt4/Modules.am b/modules/gui/qt4/Modules.am
index 988148a..57439a7 100644
--- a/modules/gui/qt4/Modules.am
+++ b/modules/gui/qt4/Modules.am
@@ -73,6 +73,7 @@ nodist_SOURCES_qt4 = \
          util/searchlineedit.moc.cpp \
          util/qmenuview.moc.cpp \
          util/qvlcapp.moc.cpp \
+        util/qvlctabwidget.moc.cpp \
          util/pictureflow.moc.cpp \
          util/buttons/RoundButton.moc.cpp \
          util/buttons/DeckButtonsLayout.moc.cpp \
@@ -395,6 +396,7 @@ noinst_HEADERS = \
      util/searchlineedit.hpp \
      util/qvlcframe.hpp \
      util/qvlcapp.hpp \
+    util/qvlctabwidget.hpp \
      util/qmenuview.hpp \
      util/qt_dirs.hpp \
      util/registry.hpp \
diff --git a/modules/gui/qt4/dialogs/sout.cpp 
b/modules/gui/qt4/dialogs/sout.cpp
index d4a6a1e..13e107f 100644
--- a/modules/gui/qt4/dialogs/sout.cpp
+++ b/modules/gui/qt4/dialogs/sout.cpp
@@ -35,6 +35,7 @@
  #include <QFileDialog>
  #include <QToolButton>
  #include <QSpinBox>
+#include <QtGlobal>
  #include <assert.h>

  SoutDialog::SoutDialog( QWidget *parent, intf_thread_t *_p_intf, const 
QString& inputMRL )
@@ -60,10 +61,12 @@ SoutDialog::SoutDialog( QWidget *parent, 
intf_thread_t *_p_intf, const QString&
                   "when you change the above settings,\n"
                   "but you can change it manually." ) ) ;

-#if 0
+#if QT_VERSION >= 0x040500
      /* This needs Qt4.5 to be cool */
      ui.destTab->setTabsClosable( true );
+    ui.destTab->getTabBar()->tabButton(0, QTabBar::RightSide)->hide();
      CONNECT( ui.destTab, tabCloseRequested( int ), this, closeTab( int 
) );
+    closeTabButton = NULL;
  #else
      closeTabButton = new QToolButton( this );
      ui.destTab->setCornerWidget( closeTabButton );
@@ -108,15 +111,14 @@ SoutDialog::SoutDialog( QWidget *parent, 
intf_thread_t *_p_intf, const QString&

  void SoutDialog::tabChanged( int i )
  {
-    closeTabButton->setVisible( (i != 0) );
+    if( closeTabButton ) closeTabButton->setVisible( (i != 0) );
  }

-void SoutDialog::closeTab()
+void SoutDialog::closeTab( int i )
  {
-    int i = ui.destTab->currentIndex();
      if( i == 0 ) return;

-    QWidget *temp = ui.destTab->currentWidget();
+    QWidget* temp = ui.destTab->widget( i );
      ui.destTab->removeTab( i );
      delete temp;
      updateMRL();
diff --git a/modules/gui/qt4/dialogs/sout.hpp 
b/modules/gui/qt4/dialogs/sout.hpp
index b15c1ee..43375db 100644
--- a/modules/gui/qt4/dialogs/sout.hpp
+++ b/modules/gui/qt4/dialogs/sout.hpp
@@ -135,7 +135,10 @@ public slots:
  private slots:
      void ok();
      void cancel();
-    void closeTab();
+    void closeTab() {
+        closeTab( ui.destTab->currentIndex() );
+    }
+    void closeTab( int );
      void tabChanged( int );
      void addDest();
  };
diff --git a/modules/gui/qt4/dialogs_provider.cpp 
b/modules/gui/qt4/dialogs_provider.cpp
index b8f5cf4..ae36c05 100644
--- a/modules/gui/qt4/dialogs_provider.cpp
+++ b/modules/gui/qt4/dialogs_provider.cpp
@@ -449,7 +449,7 @@ void DialogsProvider::addFromSimple( bool pl, bool go)
      {
          QString url = toURI( toNativeSeparators( file ) );
          playlist_Add( THEPL, qtu( url ), NULL, PLAYLIST_APPEND | mode,
-                      PLAYLIST_END, pl, pl_Unlocked );
+                      3, pl, pl_Unlocked );
          RecentsMRL::getInstance( p_intf )->addRecent( url );
          mode = PLAYLIST_PREPARSE;
      }
diff --git a/modules/gui/qt4/ui/sout.ui b/modules/gui/qt4/ui/sout.ui
index 33b835f..84a8e21 100644
--- a/modules/gui/qt4/ui/sout.ui
+++ b/modules/gui/qt4/ui/sout.ui
@@ -66,7 +66,7 @@
</property>
<layout class="QVBoxLayout" name="verticalLayout_4">
<item>
- <widget class="QTabWidget" name="destTab">
+ <widget class="QVLCTabWidget" name="destTab">
<property name="currentIndex">
<number>0</number>
</property>
@@ -243,6 +243,12 @@
<header>components/sout/sout_widgets.hpp</header>
<container>1</container>
</customwidget>
+ <customwidget>
+ <class>QVLCTabWidget</class>
+ <extends>QTabWidget</extends>
+ <header>util/qvlctabwidget.hpp</header>
+ <container>1</container>
+ </customwidget>
</customwidgets>
<resources/>
<connections/>
diff --git a/modules/gui/qt4/util/qvlctabwidget.hpp 
b/modules/gui/qt4/util/qvlctabwidget.hpp
new file mode 100644
index 0000000..17587a0
--- /dev/null
+++ b/modules/gui/qt4/util/qvlctabwidget.hpp
@@ -0,0 +1,41 @@
+/*****************************************************************************
+ * qvlctabwidget.hpp : Tab widget subclass to allow access to tabBar()
+ 
*****************************************************************************
+ * Copyright (C) 2011 the VideoLAN team
+ * $Id$
+ *
+ * Authors: Edward Wang <edward.c.wang at compdigitec.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston MA 
02110-1301, USA.
+ 
*****************************************************************************/
+
+#ifndef _QVLC_TABWIDGET_H_
+#define _QVLC_TABWIDGET_H_
+
+#include <QTabWidget>
+
+class QVLCTabWidget : public QTabWidget
+{
+    Q_OBJECT
+public:
+    QVLCTabWidget() : QTabWidget() {};
+    QVLCTabWidget( QWidget* parent )
+    : QTabWidget(parent) {
+
+    }
+    QTabBar* getTabBar() { return tabBar(); };
+};
+
+#endif
-- 
1.7.5.4



From jb at videolan.org  Fri Dec 30 02:13:26 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Fri, 30 Dec 2011 02:13:26 +0100
Subject: [vlc-devel] [PATCH] Qt4: Enable tabs in sout dialog
In-Reply-To: <4EFCED6D.5090304@compdigitec.com>
References: <1325189309-14244-1-git-send-email-edward.c.wang@compdigitec.com>
	<20111229222711.GB507@videolan.org>
	<4EFCED6D.5090304@compdigitec.com>
Message-ID: <20111230011326.GA1856@videolan.org>

On Thu, Dec 29, 2011 at 05:45:01PM -0500, Edward Wang wrote :
> +#include <QtGlobal>

No need.

> -#if 0
> +#if QT_VERSION >= 0x040500

Useless, we use already 4.6

> -                      PLAYLIST_END, pl, pl_Unlocked );
> +                      3, pl, pl_Unlocked );

Why?

> +class QVLCTabWidget : public QTabWidget
> +{
> +    Q_OBJECT
> +public:
> +    QVLCTabWidget() : QTabWidget() {};
> +    QVLCTabWidget( QWidget* parent )
> +    : QTabWidget(parent) {
> +
> +    }
> +    QTabBar* getTabBar() { return tabBar(); };
> +};

A new class, just for a getter, used only once?

Why not using something simpler, like QObject::findChild ?

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Fri Dec 30 02:14:16 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Fri, 30 Dec 2011 02:14:16 +0100
Subject: [vlc-devel] [PATCH] motion control: fix for recent kernel
In-Reply-To: <1325134677-8443-1-git-send-email-funman@videolan.org>
References: <1325134677-8443-1-git-send-email-funman@videolan.org>
Message-ID: <20111230011416.GB1856@videolan.org>

On Wed, Dec 28, 2011 at 11:57:57PM -0500, Rafa?l Carr? wrote :
> What to do : support older kernels too?

What kernel does it cahnge in?

> Btw the --motion-use-rotate doesn't work anymore.
> These days it's forbidden to find another module by name (video filter "rotate")

It is an issue for the mp4 demuxer too.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Fri Dec 30 02:18:01 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Fri, 30 Dec 2011 02:18:01 +0100
Subject: [vlc-devel] Typo
In-Reply-To: <401492634.1466568.1325191743910.JavaMail.fmail@mwmweb023>
References: <401492634.1466568.1325191743910.JavaMail.fmail@mwmweb023>
Message-ID: <20111230011801.GA2908@videolan.org>

On Thu, Dec 29, 2011 at 09:49:03PM +0100, Christoph Miebach wrote :
> Another one in vlc 1.2.0-pre3, POT-Creation-Date: 2011-12-28 03:13+0100
> 
> #: modules/demux/mp4/mp4.c:967
> msgid "Composr"
> =>
> "Composer"

Fixed.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Fri Dec 30 02:18:54 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Fri, 30 Dec 2011 02:18:54 +0100
Subject: [vlc-devel] [PATCH] Avcodec: limit the number of threads
 automatically selected to 12
In-Reply-To: <CB223586.1F25E%sergio@ammirata.net>
References: <4EFCC878.5030203@videolan.org>
	<CB223586.1F25E%sergio@ammirata.net>
Message-ID: <20111230011854.GB2908@videolan.org>

On Thu, Dec 29, 2011 at 03:25:11PM -0500, Sergio M. Ammirata, Ph.D. wrote :
> This is consistent with J-B's statement regarding ffmpeg limit of 16
> threads.

The bug is upstream then.

What do we do for it? Do we block at 16 threads?

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Fri Dec 30 02:21:51 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Fri, 30 Dec 2011 02:21:51 +0100
Subject: [vlc-devel] [PATCH] Change to npapi for giving access to media
 meta thru new input.meta(i) method.
In-Reply-To: <1325112114-6218-1-git-send-email-jean-michel.lambert7@wanadoo.fr>
References: <1325112114-6218-1-git-send-email-jean-michel.lambert7@wanadoo.fr>
Message-ID: <20111230012151.GC2908@videolan.org>

On Wed, Dec 28, 2011 at 11:41:54PM +0100, JM Lambert wrote :
> +
Why?

> +
> +

Idem?

> -    "none",
> +    "meta",

That doesn't break anything? Instead of adding it?


> +        VlcPlugin* p_plugin = getPrivate<VlcPlugin>();
> +        libvlc_media_player_t *p_md = p_plugin->getMD();
> +	libvlc_media_t * p_media = libvlc_media_player_get_media(p_md);

No tabs.
If p_md was NULL, wouldn't this be bad?

> +        if( !p_md )
> +            RETURN_ON_ERROR;
>          switch( index )
>          {
> -            case ID_none:
> -                return INVOKERESULT_NO_SUCH_METHOD;
> +            case ID_input_meta:
> +		if( argCount == 1)
tabs

> +                {
> +                    char *info;
const?

> +                    int i_metaID,  i;

i is used?

Btw, is this the best way to expose this API?

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From edward.c.wang at compdigitec.com  Fri Dec 30 02:26:31 2011
From: edward.c.wang at compdigitec.com (Edward Wang)
Date: Thu, 29 Dec 2011 20:26:31 -0500
Subject: [vlc-devel] [PATCH] Qt4: Enable tabs in sout dialog
In-Reply-To: <20111230011326.GA1856@videolan.org>
References: <1325189309-14244-1-git-send-email-edward.c.wang@compdigitec.com>
	<20111229222711.GB507@videolan.org>
	<4EFCED6D.5090304@compdigitec.com>
	<20111230011326.GA1856@videolan.org>
Message-ID: <4EFD1347.7030804@compdigitec.com>

On 11-12-29 08:13 PM, Jean-Baptiste Kempf wrote:
> A new class, just for a getter, used only once? Why not using 
> something simpler, like QObject::findChild ?
Can't do that, it is the QTabBar in question is protected and can only 
be accessed via a subclass, reference QT docs: 
http://developer.qt.nokia.com/doc/qt-4.8/QTabWidget.html

All other issues addressed. Sorry about the playlist thing, that was a 
fragment of my next patch that somehow got mixed in there...

---
  modules/gui/qt4/Modules.am             |    2 +
  modules/gui/qt4/dialogs/sout.cpp       |   23 ++---------------
  modules/gui/qt4/dialogs/sout.hpp       |    6 +++-
  modules/gui/qt4/ui/sout.ui             |    8 +++++-
  modules/gui/qt4/util/qvlctabwidget.hpp |   41 
++++++++++++++++++++++++++++++++
  5 files changed, 57 insertions(+), 23 deletions(-)
  create mode 100644 modules/gui/qt4/util/qvlctabwidget.hpp

diff --git a/modules/gui/qt4/Modules.am b/modules/gui/qt4/Modules.am
index 988148a..57439a7 100644
--- a/modules/gui/qt4/Modules.am
+++ b/modules/gui/qt4/Modules.am
@@ -73,6 +73,7 @@ nodist_SOURCES_qt4 = \
          util/searchlineedit.moc.cpp \
          util/qmenuview.moc.cpp \
          util/qvlcapp.moc.cpp \
+        util/qvlctabwidget.moc.cpp \
          util/pictureflow.moc.cpp \
          util/buttons/RoundButton.moc.cpp \
          util/buttons/DeckButtonsLayout.moc.cpp \
@@ -395,6 +396,7 @@ noinst_HEADERS = \
      util/searchlineedit.hpp \
      util/qvlcframe.hpp \
      util/qvlcapp.hpp \
+    util/qvlctabwidget.hpp \
      util/qmenuview.hpp \
      util/qt_dirs.hpp \
      util/registry.hpp \
diff --git a/modules/gui/qt4/dialogs/sout.cpp 
b/modules/gui/qt4/dialogs/sout.cpp
index d4a6a1e..c3a1538 100644
--- a/modules/gui/qt4/dialogs/sout.cpp
+++ b/modules/gui/qt4/dialogs/sout.cpp
@@ -60,20 +60,9 @@ SoutDialog::SoutDialog( QWidget *parent, 
intf_thread_t *_p_intf, const QString&
                   "when you change the above settings,\n"
                   "but you can change it manually." ) ) ;

-#if 0
-    /* This needs Qt4.5 to be cool */
      ui.destTab->setTabsClosable( true );
+    ui.destTab->getTabBar()->tabButton(0, QTabBar::RightSide)->hide();
      CONNECT( ui.destTab, tabCloseRequested( int ), this, closeTab( int 
) );
-#else
-    closeTabButton = new QToolButton( this );
-    ui.destTab->setCornerWidget( closeTabButton );
-    closeTabButton->hide();
-    closeTabButton->setAutoRaise( true );
-    closeTabButton->setIcon( QIcon( ":/toolbar/clear" ) );
-    closeTabButton->setToolTip( qtr("Clear") );
-    BUTTONACT( closeTabButton, closeTab() );
-#endif
-    CONNECT( ui.destTab, currentChanged( int ), this, tabChanged( int ) );
      ui.destTab->setTabIcon( 0, QIcon( 
":/buttons/playlist/playlist_add" ) );

      ui.destBox->addItem( qtr( "File" ) );
@@ -106,17 +95,11 @@ SoutDialog::SoutDialog( QWidget *parent, 
intf_thread_t *_p_intf, const QString&
  #undef CB
  }

-void SoutDialog::tabChanged( int i )
-{
-    closeTabButton->setVisible( (i != 0) );
-}
-
-void SoutDialog::closeTab()
+void SoutDialog::closeTab( int i )
  {
-    int i = ui.destTab->currentIndex();
      if( i == 0 ) return;

-    QWidget *temp = ui.destTab->currentWidget();
+    QWidget* temp = ui.destTab->widget( i );
      ui.destTab->removeTab( i );
      delete temp;
      updateMRL();
diff --git a/modules/gui/qt4/dialogs/sout.hpp 
b/modules/gui/qt4/dialogs/sout.hpp
index b15c1ee..d61b2e4 100644
--- a/modules/gui/qt4/dialogs/sout.hpp
+++ b/modules/gui/qt4/dialogs/sout.hpp
@@ -135,8 +135,10 @@ public slots:
  private slots:
      void ok();
      void cancel();
-    void closeTab();
-    void tabChanged( int );
+    void closeTab() {
+        closeTab( ui.destTab->currentIndex() );
+    }
+    void closeTab( int );
      void addDest();
  };

diff --git a/modules/gui/qt4/ui/sout.ui b/modules/gui/qt4/ui/sout.ui
index 33b835f..84a8e21 100644
--- a/modules/gui/qt4/ui/sout.ui
+++ b/modules/gui/qt4/ui/sout.ui
@@ -66,7 +66,7 @@
</property>
<layout class="QVBoxLayout" name="verticalLayout_4">
<item>
- <widget class="QTabWidget" name="destTab">
+ <widget class="QVLCTabWidget" name="destTab">
<property name="currentIndex">
<number>0</number>
</property>
@@ -243,6 +243,12 @@
<header>components/sout/sout_widgets.hpp</header>
<container>1</container>
</customwidget>
+ <customwidget>
+ <class>QVLCTabWidget</class>
+ <extends>QTabWidget</extends>
+ <header>util/qvlctabwidget.hpp</header>
+ <container>1</container>
+ </customwidget>
</customwidgets>
<resources/>
<connections/>
diff --git a/modules/gui/qt4/util/qvlctabwidget.hpp 
b/modules/gui/qt4/util/qvlctabwidget.hpp
new file mode 100644
index 0000000..17587a0
--- /dev/null
+++ b/modules/gui/qt4/util/qvlctabwidget.hpp
@@ -0,0 +1,41 @@
+/*****************************************************************************
+ * qvlctabwidget.hpp : Tab widget subclass to allow access to tabBar()
+ 
*****************************************************************************
+ * Copyright (C) 2011 the VideoLAN team
+ * $Id$
+ *
+ * Authors: Edward Wang <edward.c.wang at compdigitec.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston MA 
02110-1301, USA.
+ 
*****************************************************************************/
+
+#ifndef _QVLC_TABWIDGET_H_
+#define _QVLC_TABWIDGET_H_
+
+#include <QTabWidget>
+
+class QVLCTabWidget : public QTabWidget
+{
+    Q_OBJECT
+public:
+    QVLCTabWidget() : QTabWidget() {};
+    QVLCTabWidget( QWidget* parent )
+    : QTabWidget(parent) {
+
+    }
+    QTabBar* getTabBar() { return tabBar(); };
+};
+
+#endif
-- 
1.7.5.4





From jb at videolan.org  Fri Dec 30 02:26:43 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Fri, 30 Dec 2011 02:26:43 +0100
Subject: [vlc-devel] [PATCH] Qt4: Make playlist window drag&drop to
 current selected PLSelItem
In-Reply-To: <1325087980-30045-1-git-send-email-edward.c.wang@compdigitec.com>
References: <1325087980-30045-1-git-send-email-edward.c.wang@compdigitec.com>
Message-ID: <20111230012643.GD2908@videolan.org>

On Wed, Dec 28, 2011 at 10:59:40AM -0500, Edward Wang wrote :
> +    bool is_pl;
> +    int x = selector->currentItem()->data( 0, SPECIAL_ROLE ).toInt();
> +    if( x == IS_PL ) is_pl = true;
> +    else if( x == IS_ML ) is_pl = false;
> +    else return;

This should be in a method of the selector.

>  //#include <vlc_playlist.h>
> +#include "selector.hpp"
No need, use forward declaration.

> + * \param b_playlist true to add to playlist, false to add to media library

Do you manage correctly the SD?

Else, seems ok

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Fri Dec 30 02:27:51 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Fri, 30 Dec 2011 02:27:51 +0100
Subject: [vlc-devel] [PATCH] Qt4: Enable tabs in sout dialog
In-Reply-To: <4EFD1347.7030804@compdigitec.com>
References: <1325189309-14244-1-git-send-email-edward.c.wang@compdigitec.com>
	<20111229222711.GB507@videolan.org>
	<4EFCED6D.5090304@compdigitec.com>
	<20111230011326.GA1856@videolan.org>
	<4EFD1347.7030804@compdigitec.com>
Message-ID: <20111230012751.GE2908@videolan.org>

On Thu, Dec 29, 2011 at 08:26:31PM -0500, Edward Wang wrote :
> On 11-12-29 08:13 PM, Jean-Baptiste Kempf wrote:
> >A new class, just for a getter, used only once? Why not using
> >something simpler, like QObject::findChild ?
> Can't do that, it is the QTabBar in question is protected and can
> only be accessed via a subclass, reference QT docs:
> http://developer.qt.nokia.com/doc/qt-4.8/QTabWidget.html

That doesn't mean it isn't accessible with QObject::findChild. Did you
try it and it fails?

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Fri Dec 30 02:32:24 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Fri, 30 Dec 2011 02:32:24 +0100
Subject: [vlc-devel] [PATCH] Qt4: Add dialog to set specific playback
 speed
In-Reply-To: <1325122142-23755-1-git-send-email-edward.c.wang@compdigitec.com>
References: <1325122142-23755-1-git-send-email-edward.c.wang@compdigitec.com>
Message-ID: <20111230013223.GA6248@videolan.org>

On Wed, Dec 28, 2011 at 08:29:02PM -0500, Edward Wang wrote :
> This patch adds a dialog to be able to set a specific playback speed (e.g. 0.97x or 1.01x) instead of fumbling around with the small slider to get a desired value.
> Fixes bug #5734.

I would think that this is a bit complex for just adjusting a float...
Can't we have a doublespinbox instead on double-click? or one in the
popup menu?

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From edward.c.wang at compdigitec.com  Fri Dec 30 02:34:54 2011
From: edward.c.wang at compdigitec.com (Edward Wang)
Date: Thu, 29 Dec 2011 20:34:54 -0500
Subject: [vlc-devel] [PATCH] Qt4: Add dialog to set specific playback
 speed
In-Reply-To: <20111230013223.GA6248@videolan.org>
References: <1325122142-23755-1-git-send-email-edward.c.wang@compdigitec.com>
	<20111230013223.GA6248@videolan.org>
Message-ID: <4EFD153E.4030505@compdigitec.com>

On 11-12-29 08:32 PM, Jean-Baptiste Kempf wrote:
> I would think that this is a bit complex for just adjusting a float... 
> Can't we have a doublespinbox instead on double-click? or one in the 
> popup menu? Best regards, 
That could be done, although the current setup is more consistent to the 
goto time dialog. Also, I plan to add an option to launch the Set Speed 
dialog from the menus as well, something that will be hard to do with 
the embedded controls.

On a side note, will I get points for re-doing it as an embedded control 
or no? I'll still most likely do it anyways, but it would be nice to get 
points :)

Thanks,
     Edward Wang


From edward.c.wang at compdigitec.com  Fri Dec 30 02:37:42 2011
From: edward.c.wang at compdigitec.com (Edward Wang)
Date: Thu, 29 Dec 2011 20:37:42 -0500
Subject: [vlc-devel] [PATCH] Qt4: Enable tabs in sout dialog
In-Reply-To: <20111230012751.GE2908@videolan.org>
References: <1325189309-14244-1-git-send-email-edward.c.wang@compdigitec.com>
	<20111229222711.GB507@videolan.org>
	<4EFCED6D.5090304@compdigitec.com>
	<20111230011326.GA1856@videolan.org>
	<4EFD1347.7030804@compdigitec.com>
	<20111230012751.GE2908@videolan.org>
Message-ID: <4EFD15E6.5030901@compdigitec.com>

On 11-12-29 08:27 PM, Jean-Baptiste Kempf wrote:
> That doesn't mean it isn't accessible with QObject::findChild. Did you 
> try it and it fails? Best regards, 
Thank you very much for the tip, I didn't know you could access private 
and protected objects in Qt this way! Although it sorta defeats the 
point of OOP...

---
  modules/gui/qt4/dialogs/sout.cpp |   24 ++++--------------------
  modules/gui/qt4/dialogs/sout.hpp |    6 ++++--
  2 files changed, 8 insertions(+), 22 deletions(-)

diff --git a/modules/gui/qt4/dialogs/sout.cpp 
b/modules/gui/qt4/dialogs/sout.cpp
index d4a6a1e..9c29389 100644
--- a/modules/gui/qt4/dialogs/sout.cpp
+++ b/modules/gui/qt4/dialogs/sout.cpp
@@ -60,20 +60,10 @@ SoutDialog::SoutDialog( QWidget *parent, 
intf_thread_t *_p_intf, const QString&
                   "when you change the above settings,\n"
                   "but you can change it manually." ) ) ;

-#if 0
-    /* This needs Qt4.5 to be cool */
      ui.destTab->setTabsClosable( true );
+    QTabBar* tb = ui.destTab->findChild<QTabBar*>();
+    tb->tabButton(0, QTabBar::RightSide)->hide();
      CONNECT( ui.destTab, tabCloseRequested( int ), this, closeTab( int 
) );
-#else
-    closeTabButton = new QToolButton( this );
-    ui.destTab->setCornerWidget( closeTabButton );
-    closeTabButton->hide();
-    closeTabButton->setAutoRaise( true );
-    closeTabButton->setIcon( QIcon( ":/toolbar/clear" ) );
-    closeTabButton->setToolTip( qtr("Clear") );
-    BUTTONACT( closeTabButton, closeTab() );
-#endif
-    CONNECT( ui.destTab, currentChanged( int ), this, tabChanged( int ) );
      ui.destTab->setTabIcon( 0, QIcon( 
":/buttons/playlist/playlist_add" ) );

      ui.destBox->addItem( qtr( "File" ) );
@@ -106,17 +96,11 @@ SoutDialog::SoutDialog( QWidget *parent, 
intf_thread_t *_p_intf, const QString&
  #undef CB
  }

-void SoutDialog::tabChanged( int i )
-{
-    closeTabButton->setVisible( (i != 0) );
-}
-
-void SoutDialog::closeTab()
+void SoutDialog::closeTab( int i )
  {
-    int i = ui.destTab->currentIndex();
      if( i == 0 ) return;

-    QWidget *temp = ui.destTab->currentWidget();
+    QWidget* temp = ui.destTab->widget( i );
      ui.destTab->removeTab( i );
      delete temp;
      updateMRL();
diff --git a/modules/gui/qt4/dialogs/sout.hpp 
b/modules/gui/qt4/dialogs/sout.hpp
index b15c1ee..d61b2e4 100644
--- a/modules/gui/qt4/dialogs/sout.hpp
+++ b/modules/gui/qt4/dialogs/sout.hpp
@@ -135,8 +135,10 @@ public slots:
  private slots:
      void ok();
      void cancel();
-    void closeTab();
-    void tabChanged( int );
+    void closeTab() {
+        closeTab( ui.destTab->currentIndex() );
+    }
+    void closeTab( int );
      void addDest();
  };

-- 
1.7.5.4


From jb at videolan.org  Fri Dec 30 02:41:19 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Fri, 30 Dec 2011 02:41:19 +0100
Subject: [vlc-devel] [PATCH] Qt4: Enable tabs in sout dialog
In-Reply-To: <4EFD15E6.5030901@compdigitec.com>
References: <1325189309-14244-1-git-send-email-edward.c.wang@compdigitec.com>
	<20111229222711.GB507@videolan.org>
	<4EFCED6D.5090304@compdigitec.com>
	<20111230011326.GA1856@videolan.org>
	<4EFD1347.7030804@compdigitec.com>
	<20111230012751.GE2908@videolan.org>
	<4EFD15E6.5030901@compdigitec.com>
Message-ID: <20111230014119.GA7919@videolan.org>

On Thu, Dec 29, 2011 at 08:37:42PM -0500, Edward Wang wrote :
> +    QTabBar* tb = ui.destTab->findChild<QTabBar*>();
> +    tb->tabButton(0, QTabBar::RightSide)->hide();

You MUST check for NULLity after findChild

> +    void closeTab() {
> +        closeTab( ui.destTab->currentIndex() );
> +    }
> +    void closeTab( int );

Do we really need those 2?

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From edward.c.wang at compdigitec.com  Fri Dec 30 02:44:47 2011
From: edward.c.wang at compdigitec.com (Edward Wang)
Date: Thu, 29 Dec 2011 20:44:47 -0500
Subject: [vlc-devel] [PATCH] Qt4: Enable tabs in sout dialog
In-Reply-To: <20111230014119.GA7919@videolan.org>
References: <1325189309-14244-1-git-send-email-edward.c.wang@compdigitec.com>
	<20111229222711.GB507@videolan.org>
	<4EFCED6D.5090304@compdigitec.com>
	<20111230011326.GA1856@videolan.org>
	<4EFD1347.7030804@compdigitec.com>
	<20111230012751.GE2908@videolan.org>
	<4EFD15E6.5030901@compdigitec.com>
	<20111230014119.GA7919@videolan.org>
Message-ID: <4EFD178F.4020606@compdigitec.com>

On 11-12-29 08:41 PM, Jean-Baptiste Kempf wrote:
> Do we really need those 2? Best regards, 
Done

Thanks,
     Edward Wang
---
  modules/gui/qt4/dialogs/sout.cpp |   24 ++++--------------------
  modules/gui/qt4/dialogs/sout.hpp |    3 +--
  2 files changed, 5 insertions(+), 22 deletions(-)

diff --git a/modules/gui/qt4/dialogs/sout.cpp 
b/modules/gui/qt4/dialogs/sout.cpp
index d4a6a1e..c091744 100644
--- a/modules/gui/qt4/dialogs/sout.cpp
+++ b/modules/gui/qt4/dialogs/sout.cpp
@@ -60,20 +60,10 @@ SoutDialog::SoutDialog( QWidget *parent, 
intf_thread_t *_p_intf, const QString&
                   "when you change the above settings,\n"
                   "but you can change it manually." ) ) ;

-#if 0
-    /* This needs Qt4.5 to be cool */
      ui.destTab->setTabsClosable( true );
+    QTabBar* tb = ui.destTab->findChild<QTabBar*>();
+    if( tb != NULL ) tb->tabButton(0, QTabBar::RightSide)->hide();
      CONNECT( ui.destTab, tabCloseRequested( int ), this, closeTab( int 
) );
-#else
-    closeTabButton = new QToolButton( this );
-    ui.destTab->setCornerWidget( closeTabButton );
-    closeTabButton->hide();
-    closeTabButton->setAutoRaise( true );
-    closeTabButton->setIcon( QIcon( ":/toolbar/clear" ) );
-    closeTabButton->setToolTip( qtr("Clear") );
-    BUTTONACT( closeTabButton, closeTab() );
-#endif
-    CONNECT( ui.destTab, currentChanged( int ), this, tabChanged( int ) );
      ui.destTab->setTabIcon( 0, QIcon( 
":/buttons/playlist/playlist_add" ) );

      ui.destBox->addItem( qtr( "File" ) );
@@ -106,17 +96,11 @@ SoutDialog::SoutDialog( QWidget *parent, 
intf_thread_t *_p_intf, const QString&
  #undef CB
  }

-void SoutDialog::tabChanged( int i )
-{
-    closeTabButton->setVisible( (i != 0) );
-}
-
-void SoutDialog::closeTab()
+void SoutDialog::closeTab( int i )
  {
-    int i = ui.destTab->currentIndex();
      if( i == 0 ) return;

-    QWidget *temp = ui.destTab->currentWidget();
+    QWidget* temp = ui.destTab->widget( i );
      ui.destTab->removeTab( i );
      delete temp;
      updateMRL();
diff --git a/modules/gui/qt4/dialogs/sout.hpp 
b/modules/gui/qt4/dialogs/sout.hpp
index b15c1ee..40d7242 100644
--- a/modules/gui/qt4/dialogs/sout.hpp
+++ b/modules/gui/qt4/dialogs/sout.hpp
@@ -135,8 +135,7 @@ public slots:
  private slots:
      void ok();
      void cancel();
-    void closeTab();
-    void tabChanged( int );
+    void closeTab( int );
      void addDest();
  };

-- 
1.7.5.4


From funman at videolan.org  Fri Dec 30 02:45:05 2011
From: funman at videolan.org (=?ISO-8859-1?Q?Rafa=EBl_Carr=E9?=)
Date: Thu, 29 Dec 2011 20:45:05 -0500
Subject: [vlc-devel] [PATCH] motion control: fix for recent kernel
In-Reply-To: <20111230011416.GB1856@videolan.org>
References: <1325134677-8443-1-git-send-email-funman@videolan.org>
	<20111230011416.GB1856@videolan.org>
Message-ID: <4EFD17A1.7010306@videolan.org>

Le 11-12-29 20:14, Jean-Baptiste Kempf a ?crit :
> On Wed, Dec 28, 2011 at 11:57:57PM -0500, Rafa?l Carr? wrote :
>> What to do : support older kernels too?
> 
> What kernel does it cahnge in?

Dunno, the file itself doesn't set that path.

I think we should check when it started to be a platform device perhaps

>> Btw the --motion-use-rotate doesn't work anymore.
>> These days it's forbidden to find another module by name (video filter "rotate")
> 
> It is an issue for the mp4 demuxer too.
> 
> Best regards,

courmisch?

should we just remove that code?


From funman at videolan.org  Fri Dec 30 02:46:32 2011
From: funman at videolan.org (=?ISO-8859-1?Q?Rafa=EBl_Carr=E9?=)
Date: Thu, 29 Dec 2011 20:46:32 -0500
Subject: [vlc-devel] [PATCH] Avcodec: limit the number of threads
 automatically selected to 12
In-Reply-To: <20111230011854.GB2908@videolan.org>
References: <4EFCC878.5030203@videolan.org>
	<CB223586.1F25E%sergio@ammirata.net>
	<20111230011854.GB2908@videolan.org>
Message-ID: <4EFD17F8.9040900@videolan.org>

Le 11-12-29 20:18, Jean-Baptiste Kempf a ?crit :
> On Thu, Dec 29, 2011 at 03:25:11PM -0500, Sergio M. Ammirata, Ph.D. wrote :
>> This is consistent with J-B's statement regarding ffmpeg limit of 16
>> threads.
> 
> The bug is upstream then.
> 
> What do we do for it? Do we block at 16 threads?

I think it makes sense but it should probably limited in ffmpeg/libav too


From jb at videolan.org  Fri Dec 30 02:54:14 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Fri, 30 Dec 2011 02:54:14 +0100
Subject: [vlc-devel] [PATCH] Qt4: Enable tabs in sout dialog
In-Reply-To: <4EFD178F.4020606@compdigitec.com>
References: <1325189309-14244-1-git-send-email-edward.c.wang@compdigitec.com>
	<20111229222711.GB507@videolan.org>
	<4EFCED6D.5090304@compdigitec.com>
	<20111230011326.GA1856@videolan.org>
	<4EFD1347.7030804@compdigitec.com>
	<20111230012751.GE2908@videolan.org>
	<4EFD15E6.5030901@compdigitec.com>
	<20111230014119.GA7919@videolan.org>
	<4EFD178F.4020606@compdigitec.com>
Message-ID: <20111230015414.GA8770@videolan.org>

On Thu, Dec 29, 2011 at 08:44:47PM -0500, Edward Wang wrote :
> On 11-12-29 08:41 PM, Jean-Baptiste Kempf wrote:
> >Do we really need those 2? Best regards,
> Done

Didn't apply. Modified and merged. Thanks.

I prefer this shorter version, to be honest.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Fri Dec 30 02:55:01 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Fri, 30 Dec 2011 02:55:01 +0100
Subject: [vlc-devel] [PATCH] Avcodec: limit the number of threads
 automatically selected to 12
In-Reply-To: <4EFD17F8.9040900@videolan.org>
References: <4EFCC878.5030203@videolan.org>
	<CB223586.1F25E%sergio@ammirata.net>
	<20111230011854.GB2908@videolan.org>
	<4EFD17F8.9040900@videolan.org>
Message-ID: <20111230015501.GA11178@videolan.org>

On Thu, Dec 29, 2011 at 08:46:32PM -0500, Rafa?l Carr? wrote :
> Le 11-12-29 20:18, Jean-Baptiste Kempf a ?crit :
> > On Thu, Dec 29, 2011 at 03:25:11PM -0500, Sergio M. Ammirata, Ph.D. wrote :
> >> This is consistent with J-B's statement regarding ffmpeg limit of 16
> >> threads.
> > 
> > The bug is upstream then.
> > 
> > What do we do for it? Do we block at 16 threads?
> 
> I think it makes sense but it should probably limited in ffmpeg/libav too

It is. But the limitation doesn't work, which is the issue here.


-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Fri Dec 30 02:55:33 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Fri, 30 Dec 2011 02:55:33 +0100
Subject: [vlc-devel] [PATCH] Don't keep VLC on top when paused
In-Reply-To: <1325039452-22490-1-git-send-email-edward.c.wang@compdigitec.com>
References: <1325039452-22490-1-git-send-email-edward.c.wang@compdigitec.com>
Message-ID: <20111230015533.GB11178@videolan.org>

On Tue, Dec 27, 2011 at 09:30:52PM -0500, Edward Wang wrote :
> Have the "always on top" preference only take effect while a movie is playing. When not playing VLC will act like any other window.
> Fixes #4689 if committed.

This works for me. Anyone knows if that is the proper way? Laurent?

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Fri Dec 30 02:56:32 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Fri, 30 Dec 2011 02:56:32 +0100
Subject: [vlc-devel] [PATCH] Qt4: Add dialog to set specific playback
 speed
In-Reply-To: <4EFD153E.4030505@compdigitec.com>
References: <1325122142-23755-1-git-send-email-edward.c.wang@compdigitec.com>
	<20111230013223.GA6248@videolan.org>
	<4EFD153E.4030505@compdigitec.com>
Message-ID: <20111230015632.GC11178@videolan.org>

On Thu, Dec 29, 2011 at 08:34:54PM -0500, Edward Wang wrote :
> On 11-12-29 08:32 PM, Jean-Baptiste Kempf wrote:
> >I would think that this is a bit complex for just adjusting a
> >float... Can't we have a doublespinbox instead on double-click? or
> >one in the popup menu? Best regards,
> That could be done, although the current setup is more consistent to
> the goto time dialog. Also, I plan to add an option to launch the

Yes, but I dislike the goto dialog in the same way.

> Set Speed dialog from the menus as well, something that will be hard
> to do with the embedded controls.

I don't see the point, to be honest, but maybe I am wrong.

> On a side note, will I get points for re-doing it as an embedded
> control or no? I'll still most likely do it anyways, but it would be
> nice to get points :)

I don't think that would be fair.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From zhushisongzhu at yahoo.com  Fri Dec 30 03:50:42 2011
From: zhushisongzhu at yahoo.com (zhu shi song)
Date: Thu, 29 Dec 2011 18:50:42 -0800 (PST)
Subject: [vlc-devel] VLC playing problem on Athlon computer
In-Reply-To: <mailman.2074.1325209602.1630.vlc-devel@videolan.org>
References: <mailman.2074.1325209602.1630.vlc-devel@videolan.org>
Message-ID: <1325213442.81547.YahooMailNeo@web125702.mail.ne1.yahoo.com>

???? Following the instructions on WEB http://wiki.videolan.org/Win32CompileMSYSNew#Precompiled_contribs, I have compiled VLC 1.1.5 successfully.? I used contrib-20100608-win32-bin-gcc-4.4.4-runtime-3.17-only.tar.bz2.? 

???? I use it on Intel CPU computer , it works well.? But I test it on AMD Athlon 64 CPU 
computer, it can't play MP4(h.264+AAC) smoothly.? The time progress bar 
is moving, but video is frozen.
??? The CPU details:?? AMD Athlon 64x2 Dual Core Processor 4400+.
???? OS: WindowXP Home Edition SP2
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111229/cbaa59ce/attachment.html>

From jean-michel.lambert7 at wanadoo.fr  Fri Dec 30 10:45:50 2011
From: jean-michel.lambert7 at wanadoo.fr (jean-michel Lambert)
Date: Fri, 30 Dec 2011 10:45:50 +0100
Subject: [vlc-devel] [PATCH] Change to npapi for giving access to media
 meta thru new input.meta(i) method.
In-Reply-To: <20111230012151.GC2908@videolan.org>
References: <1325112114-6218-1-git-send-email-jean-michel.lambert7@wanadoo.fr>
	<20111230012151.GC2908@videolan.org>
Message-ID: <4EFD884E.1010606@wanadoo.fr>

Regarding the way to expose the meta info, do you prefer me to add a 
property "meta" and sub properties for each meta type  (ie 
input.meta.title, meta.currentlyPlaying, etc ... ?). I liked it this way 
before it allows to process new fields in the future without recompiling 
everything.

thanks,

On 30/12/2011 02:21, Jean-Baptiste Kempf wrote:
> On Wed, Dec 28, 2011 at 11:41:54PM +0100, JM Lambert wrote :
>> +
> Why?
>
>> +
>> +
> Idem?
>
>> -    "none",
>> +    "meta",
> That doesn't break anything? Instead of adding it?
>
>
>> +        VlcPlugin* p_plugin = getPrivate<VlcPlugin>();
>> +        libvlc_media_player_t *p_md = p_plugin->getMD();
>> +	libvlc_media_t * p_media = libvlc_media_player_get_media(p_md);
> No tabs.
> If p_md was NULL, wouldn't this be bad?
>
>> +        if( !p_md )
>> +            RETURN_ON_ERROR;
>>           switch( index )
>>           {
>> -            case ID_none:
>> -                return INVOKERESULT_NO_SUCH_METHOD;
>> +            case ID_input_meta:
>> +		if( argCount == 1)
> tabs
>
>> +                {
>> +                    char *info;
> const?
>
>> +                    int i_metaID,  i;
> i is used?
>
> Btw, is this the best way to expose this API?
>
> Best regards,
>



From fyhuel at viotech.net  Fri Dec 30 11:07:15 2011
From: fyhuel at viotech.net (Frederic YHUEL)
Date: Fri, 30 Dec 2011 11:07:15 +0100
Subject: [vlc-devel] [PATCH] httplive.c: fix hls_Read (when caller skips
	data)
In-Reply-To: <1325157243-24181-1-git-send-email-fyhuel@viotech.net>
References: <1325157243-24181-1-git-send-email-fyhuel@viotech.net>
Message-ID: <CAKHNsC2OOBuvZK2GkNyce3SdonStpvFeFhiq72_rt8d-RF=6QQ@mail.gmail.com>

2011/12/29 Fr?d?ric Yhuel <fyhuel at viotech.net>:
> I have also fixed a comment
> ---
> ?modules/stream_filter/httplive.c | ? ?6 ++++--
> ?1 files changed, 4 insertions(+), 2 deletions(-)
>
> diff --git a/modules/stream_filter/httplive.c b/modules/stream_filter/httplive.c
> index 4d60d8a..47b99a1 100644
> --- a/modules/stream_filter/httplive.c
> +++ b/modules/stream_filter/httplive.c
> @@ -78,7 +78,8 @@ typedef struct hls_stream_s
> ? ? int ? ? ? ? sequence; ? /* media sequence number */
> ? ? int ? ? ? ? duration; ? /* maximum duration per segment (s) */
> ? ? uint64_t ? ?bandwidth; ?/* bandwidth usage of segments (bits per second)*/
> - ? ?uint64_t ? ?size; ? ? ? /* stream length (segment->duration * hls->bandwidth/8) */
> + ? ?uint64_t ? ?size; ? ? ? /* stream length = Sum for all segments(
> + ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?segment->duration * hls->bandwidth/8 ) */
>
> ? ? vlc_array_t *segments; ?/* list of segments */
> ? ? vlc_url_t ? url; ? ? ? ?/* uri to m3u8 */
> @@ -1912,7 +1913,8 @@ static ssize_t hls_Read(stream_t *s, uint8_t *p_read, unsigned int i_read)
>
> ? ? ? ? if (len > 0)
> ? ? ? ? {
> - ? ? ? ? ? ?memcpy(p_read + copied, segment->data->p_buffer, len);
> + ? ? ? ? ? ?if( p_read ) /* otherwise caller skips data */
> + ? ? ? ? ? ? ? ?memcpy(p_read + copied, segment->data->p_buffer, len);
> ? ? ? ? ? ? segment->data->i_buffer -= len;
> ? ? ? ? ? ? segment->data->p_buffer += len;
> ? ? ? ? ? ? copied += len;
> --
> 1.7.5.4
>

Sorry this patch is wrong, I'll send a new one.

-- 
Fr?d?ric


From fyhuel at viotech.net  Fri Dec 30 11:18:20 2011
From: fyhuel at viotech.net (fyhuel at viotech.net)
Date: Fri, 30 Dec 2011 11:18:20 +0100
Subject: [vlc-devel] [PATCH 1/2] httplive: fix Read func. (when caller skips
	data)
Message-ID: <4efd900a.0d12980a.3671.30bb@mx.google.com>

From: Fr?d?ric Yhuel <fyhuel at viotech.net>

---
 modules/stream_filter/httplive.c |   12 ++----------
 1 files changed, 2 insertions(+), 10 deletions(-)

diff --git a/modules/stream_filter/httplive.c b/modules/stream_filter/httplive.c
index 4d60d8a..58dc1db 100644
--- a/modules/stream_filter/httplive.c
+++ b/modules/stream_filter/httplive.c
@@ -1912,7 +1912,8 @@ static ssize_t hls_Read(stream_t *s, uint8_t *p_read, unsigned int i_read)
 
         if (len > 0)
         {
-            memcpy(p_read + copied, segment->data->p_buffer, len);
+            if( p_read ) /* otherwise caller skips data */
+                memcpy(p_read + copied, segment->data->p_buffer, len);
             segment->data->i_buffer -= len;
             segment->data->p_buffer += len;
             copied += len;
@@ -1935,15 +1936,6 @@ static int Read(stream_t *s, void *buffer, unsigned int i_read)
     if (p_sys->b_error)
         return 0;
 
-    if (buffer == NULL)
-    {
-        /* caller skips data, get big enough buffer */
-        msg_Warn(s, "buffer is NULL (allocate %d)", i_read);
-        buffer = calloc(1, i_read);
-        if (buffer == NULL)
-            return 0; /* NO MEMORY left*/
-    }
-
     length = hls_Read(s, (uint8_t*) buffer, i_read);
     if (length < 0)
         return 0;
-- 
1.7.5.4



From fyhuel at viotech.net  Fri Dec 30 11:18:21 2011
From: fyhuel at viotech.net (fyhuel at viotech.net)
Date: Fri, 30 Dec 2011 11:18:21 +0100
Subject: [vlc-devel] [PATCH 2/2] httplive: fix a comment
In-Reply-To: <1325240301-6067-1-git-send-email-y>
References: <1325240301-6067-1-git-send-email-y>
Message-ID: <4efd9020.0d12980a.3671.30c7@mx.google.com>

From: Fr?d?ric Yhuel <fyhuel at viotech.net>

---
 modules/stream_filter/httplive.c |    3 ++-
 1 files changed, 2 insertions(+), 1 deletions(-)

diff --git a/modules/stream_filter/httplive.c b/modules/stream_filter/httplive.c
index 58dc1db..e635db3 100644
--- a/modules/stream_filter/httplive.c
+++ b/modules/stream_filter/httplive.c
@@ -78,7 +78,8 @@ typedef struct hls_stream_s
     int         sequence;   /* media sequence number */
     int         duration;   /* maximum duration per segment (s) */
     uint64_t    bandwidth;  /* bandwidth usage of segments (bits per second)*/
-    uint64_t    size;       /* stream length (segment->duration * hls->bandwidth/8) */
+    uint64_t    size;       /* stream length = Sum for all segments(
+                                    segment->duration * hls->bandwidth/8 ) */
 
     vlc_array_t *segments;  /* list of segments */
     vlc_url_t   url;        /* uri to m3u8 */
-- 
1.7.5.4



From rsatom at gmail.com  Fri Dec 30 11:46:33 2011
From: rsatom at gmail.com (Sergey Radionov)
Date: Fri, 30 Dec 2011 17:46:33 +0700
Subject: [vlc-devel] [PATCH 1/4] npapi win32: added protection from multiple
	calling of create_windows(); 
Message-ID: <1325241996-2844-1-git-send-email-RSATom@gmail.com>

---
 npapi/vlcplugin_win.cpp |    3 +++
 1 files changed, 3 insertions(+), 0 deletions(-)

diff --git a/npapi/vlcplugin_win.cpp b/npapi/vlcplugin_win.cpp
index 5b40f7a..745cfec 100644
--- a/npapi/vlcplugin_win.cpp
+++ b/npapi/vlcplugin_win.cpp
@@ -101,6 +101,9 @@ bool VlcPluginWin::create_windows()
 {
     HWND drawable = (HWND) (getWindow().window);
 
+    if( GetWindowLongPtr(drawable, GWLP_USERDATA) )
+        return false;
+
     /* attach our plugin object */
     SetWindowLongPtr(drawable, GWLP_USERDATA,
                      (LONG_PTR)this);
-- 
1.7.7.1.msysgit.0



From rsatom at gmail.com  Fri Dec 30 11:46:34 2011
From: rsatom at gmail.com (Sergey Radionov)
Date: Fri, 30 Dec 2011 17:46:34 +0700
Subject: [vlc-devel] [PATCH 2/4] npapi win32: added window proc replacing
	for browser drawable window.
In-Reply-To: <1325241996-2844-1-git-send-email-RSATom@gmail.com>
References: <1325241996-2844-1-git-send-email-RSATom@gmail.com>
Message-ID: <1325241996-2844-2-git-send-email-RSATom@gmail.com>

some browsers (mainly opera) have not good realization of NPApi,
and we will need track some events via native window messages,
rather than via NPP_SetWindow.
---
 npapi/vlcplugin_win.cpp |   26 +++++++++++++++++++++++++-
 npapi/vlcplugin_win.h   |    2 ++
 2 files changed, 27 insertions(+), 1 deletions(-)

diff --git a/npapi/vlcplugin_win.cpp b/npapi/vlcplugin_win.cpp
index 745cfec..13322d2 100644
--- a/npapi/vlcplugin_win.cpp
+++ b/npapi/vlcplugin_win.cpp
@@ -50,8 +50,17 @@ BOOL WINAPI DllMain(HANDLE hModule, DWORD fdwReason, LPVOID lpReserved )
 };
 
 
+LRESULT CALLBACK VlcPluginWin::NPWndProcR( HWND p_hwnd, UINT i_msg, WPARAM wpar, LPARAM lpar )
+{
+    VlcPluginWin *p_plugin = reinterpret_cast<VlcPluginWin *>(GetWindowLongPtr(p_hwnd, GWLP_USERDATA));
+
+    /* delegate to default handler */
+    return CallWindowProc( p_plugin->_NPWndProc, p_hwnd,
+                           i_msg, wpar, lpar );
+}
+
 VlcPluginWin::VlcPluginWin(NPP instance, NPuint16_t mode) :
-    VlcPluginBase(instance, mode),
+    VlcPluginBase(instance, mode), _NPWndProc(0),
     _WindowsManager(DllGetModule())
 {
 }
@@ -108,6 +117,12 @@ bool VlcPluginWin::create_windows()
     SetWindowLongPtr(drawable, GWLP_USERDATA,
                      (LONG_PTR)this);
 
+    /* install our WNDPROC */
+    _NPWndProc =
+        (WNDPROC) SetWindowLongPtr( drawable,
+                                    GWLP_WNDPROC,
+                                    (LONG_PTR)NPWndProcR );
+
     /* change window style to our liking */
     LONG style = GetWindowLong(drawable, GWL_STYLE);
     style |= WS_CLIPCHILDREN | WS_CLIPSIBLINGS;
@@ -133,6 +148,15 @@ bool VlcPluginWin::resize_windows()
 bool VlcPluginWin::destroy_windows()
 {
     _WindowsManager.DestroyWindows();
+
+    HWND hWnd = (HWND)npwindow.window;
+    if( hWnd && _NPWndProc){
+        /* reset WNDPROC */
+        SetWindowLongPtr( hWnd, GWLP_WNDPROC, (LONG_PTR)_NPWndProc );
+    }
+    _NPWndProc = 0;
+    npwindow.window = 0;
+
     return true;
 }
 
diff --git a/npapi/vlcplugin_win.h b/npapi/vlcplugin_win.h
index bbc66c9..7415bed 100644
--- a/npapi/vlcplugin_win.h
+++ b/npapi/vlcplugin_win.h
@@ -63,8 +63,10 @@ protected:
     virtual void on_media_player_release();
 
 private:
+    static LRESULT CALLBACK NPWndProcR( HWND p_hwnd, UINT i_msg, WPARAM wpar, LPARAM lpar );
     void set_player_window(){};
 
+    WNDPROC _NPWndProc;
     VLCWindowsManager _WindowsManager;
 };
 
-- 
1.7.7.1.msysgit.0



From rsatom at gmail.com  Fri Dec 30 11:46:35 2011
From: rsatom at gmail.com (Sergey Radionov)
Date: Fri, 30 Dec 2011 17:46:35 +0700
Subject: [vlc-devel] [PATCH 3/4] npapi win32: detach VlcPlugin from drawable
	on windows destroying.
In-Reply-To: <1325241996-2844-1-git-send-email-RSATom@gmail.com>
References: <1325241996-2844-1-git-send-email-RSATom@gmail.com>
Message-ID: <1325241996-2844-3-git-send-email-RSATom@gmail.com>

---
 npapi/vlcplugin_win.cpp |    2 ++
 1 files changed, 2 insertions(+), 0 deletions(-)

diff --git a/npapi/vlcplugin_win.cpp b/npapi/vlcplugin_win.cpp
index 13322d2..d36b078 100644
--- a/npapi/vlcplugin_win.cpp
+++ b/npapi/vlcplugin_win.cpp
@@ -157,6 +157,8 @@ bool VlcPluginWin::destroy_windows()
     _NPWndProc = 0;
     npwindow.window = 0;
 
+    SetWindowLongPtr(hWnd, GWLP_USERDATA, 0);
+
     return true;
 }
 
-- 
1.7.7.1.msysgit.0



From rsatom at gmail.com  Fri Dec 30 11:49:32 2011
From: rsatom at gmail.com (Sergey Radionov)
Date: Fri, 30 Dec 2011 17:49:32 +0700
Subject: [vlc-devel] [PATCH 4/4] npapi win32: call destroy_windows on
	drawable destroying (opera compatibility fix)
Message-ID: <1325242172-1968-4-git-send-email-RSATom@gmail.com>

---
 npapi/vlcplugin_win.cpp |    6 ++++++
 1 files changed, 6 insertions(+), 0 deletions(-)

diff --git a/npapi/vlcplugin_win.cpp b/npapi/vlcplugin_win.cpp
index d36b078..a4a1f33 100644
--- a/npapi/vlcplugin_win.cpp
+++ b/npapi/vlcplugin_win.cpp
@@ -54,6 +54,12 @@ LRESULT CALLBACK VlcPluginWin::NPWndProcR( HWND p_hwnd, UINT i_msg, WPARAM wpar,
 {
     VlcPluginWin *p_plugin = reinterpret_cast<VlcPluginWin *>(GetWindowLongPtr(p_hwnd, GWLP_USERDATA));
 
+    switch( i_msg ){
+        case WM_DESTROY:
+            p_plugin->destroy_windows();
+            break;
+    }
+
     /* delegate to default handler */
     return CallWindowProc( p_plugin->_NPWndProc, p_hwnd,
                            i_msg, wpar, lpar );
-- 
1.7.7.1.msysgit.0



From beauze.h at gmail.com  Fri Dec 30 16:09:44 2011
From: beauze.h at gmail.com (=?UTF-8?Q?Hugo_Beauz=C3=A9e=2DLuyssen?=)
Date: Fri, 30 Dec 2011 16:09:44 +0100
Subject: [vlc-devel] [PATCH] Avcodec: limit the number of threads
 automatically selected to 12
In-Reply-To: <4EFCC878.5030203@videolan.org>
References: <1325124481-30031-1-git-send-email-jb@videolan.org>
	<4EFBF3C5.1060407@gmail.com> <20111229092545.GA14468@videolan.org>
	<CAK-C75QjdgLHfnJoFsXGUhU2eeucwh09M=jTwF38DWQtJVc5hA@mail.gmail.com>
	<4EFCC878.5030203@videolan.org>
Message-ID: <CAK-C75RxJkAzQ_TC4wSG7d3te2hibD5T1eu0y6ZUSf2NLWvFsA@mail.gmail.com>

On Thu, Dec 29, 2011 at 9:07 PM, Rafa?l Carr? <funman at videolan.org> wrote:
> Le 11-12-29 04:36, Hugo Beauz?e-Luyssen a ?crit :
>> On Thu, Dec 29, 2011 at 10:25 AM, Jean-Baptiste Kempf <jb at videolan.org> wrote:
>>> On Wed, Dec 28, 2011 at 11:59:49PM -0500, Rafa?l Carr? wrote :
>>>> Le 2011-12-28 21:08, Jean-Baptiste Kempf a ?crit :
>>>>> It seems that over 16 doesn't behave well. Reduce it further for safety
>>>>
>>>> What is the problem exactly ?
>>>
>>> According to ringo, it doesn't work at all with 24 threads. According to
>>> the source, 16 threads is the maximum anyway.
>>>
>>> Best regards,
>>>
>>
>> Doesn't work for me with ffmpeg-threads > 5, on a quad core.
>
> 5 does work?
>
> What is the error, the assertion that Denis reported?
>

5 seems to work for me.

The error is indeed what TypX described :
[0x1519f28] main video output error: pictures leaked, trying to
workaround[0x1519f28] main video output error: pictures leaked, trying
to workaround
....vlc: ../../src/misc/picture_pool.c:266: Release: Assertion
`picture->i_refcount > 0' failed.
-- 
Hugo Beauz?e-Luyssen


From jb at videolan.org  Fri Dec 30 17:51:33 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Fri, 30 Dec 2011 17:51:33 +0100
Subject: [vlc-devel] Blockers for VLC 1.2
Message-ID: <20111230165133.GA15175@videolan.org>

Hello,

I've done some bug triaging and a bit of forum-walking to find the
remaining important issues for VLC 1.2.0.

We have:
 - UI, UI, UI, UI
 - DVB is broken on Win32 (#5627)
 - libvlc_video_get_size is still not working: is it the window size or
   the video size? (#3679)
 - MKV subtitles across segments (#5625)
 - Vobsub subtitle placement (#5683)
 - UI, UI, UI

Any good input would be valued.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From edward.c.wang at compdigitec.com  Fri Dec 30 17:53:16 2011
From: edward.c.wang at compdigitec.com (Edward Wang)
Date: Fri, 30 Dec 2011 11:53:16 -0500
Subject: [vlc-devel] Blockers for VLC 1.2
In-Reply-To: <20111230165133.GA15175@videolan.org>
References: <20111230165133.GA15175@videolan.org>
Message-ID: <4EFDEC7C.4070709@compdigitec.com>

On 11-12-30 11:51 AM, Jean-Baptiste Kempf wrote:
> Hello,
>
> I've done some bug triaging and a bit of forum-walking to find the
> remaining important issues for VLC 1.2.0.
>
> We have:
>   - UI, UI, UI, UI
>   - DVB is broken on Win32 (#5627)
>   - libvlc_video_get_size is still not working: is it the window size or
>     the video size? (#3679)
>   - MKV subtitles across segments (#5625)
>   - Vobsub subtitle placement (#5683)
>   - UI, UI, UI
>
Apologies if this is a stupid question, but what do you mean when you 
say "UI, UI, UI"? Would it be something I can work on or no?

Thanks,
     Edward Wang


From jb at videolan.org  Fri Dec 30 17:54:42 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Fri, 30 Dec 2011 17:54:42 +0100
Subject: [vlc-devel] Blockers for VLC 1.2
In-Reply-To: <4EFDEC7C.4070709@compdigitec.com>
References: <20111230165133.GA15175@videolan.org>
	<4EFDEC7C.4070709@compdigitec.com>
Message-ID: <20111230165442.GA16585@videolan.org>

On Fri, Dec 30, 2011 at 11:53:16AM -0500, Edward Wang wrote :
> Apologies if this is a stupid question, but what do you mean when
> you say "UI, UI, UI"? Would it be something I can work on or no?

OSX UI and Qt UI have issues. And yes, you can work on things, like
#5588 or other quirks (come on IRC to discuss).

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From rafael.carre at gmail.com  Fri Dec 30 17:55:07 2011
From: rafael.carre at gmail.com (=?ISO-8859-1?Q?Rafa=EBl_Carr=E9?=)
Date: Fri, 30 Dec 2011 11:55:07 -0500
Subject: [vlc-devel] Blockers for VLC 1.2
In-Reply-To: <20111230165133.GA15175@videolan.org>
References: <20111230165133.GA15175@videolan.org>
Message-ID: <4EFDECEB.6090505@gmail.com>

Le 2011-12-30 11:51, Jean-Baptiste Kempf a ?crit :
> Hello,
> 
> I've done some bug triaging and a bit of forum-walking to find the
> remaining important issues for VLC 1.2.0.
> 
> We have:
>  - UI, UI, UI, UI
>  - DVB is broken on Win32 (#5627)
>  - libvlc_video_get_size is still not working: is it the window size or
>    the video size? (#3679)

Why is it a blocker? It's been disabled for a long time already in
1.1.0-pre1-104-g64da2d6

I agree that it was present in 1.1 though.

Laurent?

>  - MKV subtitles across segments (#5625)
>  - Vobsub subtitle placement (#5683)
>  - UI, UI, UI

ncurses UI is fine, let's make that the default on win32/osx ;-)

> Any good input would be valued.
> 
> Best regards,
> 



From jb at videolan.org  Fri Dec 30 17:57:08 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Fri, 30 Dec 2011 17:57:08 +0100
Subject: [vlc-devel] Blockers for VLC 1.2
In-Reply-To: <4EFDECEB.6090505@gmail.com>
References: <4EFDECEB.6090505@gmail.com>
Message-ID: <20111230165708.GA17150@videolan.org>

On Fri, Dec 30, 2011 at 11:55:07AM -0500, Rafa?l Carr? wrote :
> Why is it a blocker?

Maybe not a blocker, but a very important issue, I think, notably if we
want to promote libVLC...

> ncurses UI is fine, let's make that the default on win32/osx ;-)

Ok.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From edward.c.wang at compdigitec.com  Fri Dec 30 23:48:51 2011
From: edward.c.wang at compdigitec.com (Edward Wang)
Date: Fri, 30 Dec 2011 17:48:51 -0500
Subject: [vlc-devel] [PATCH] Qt4: Show fullscreen docking button in defaults
Message-ID: <1325285331-25517-1-git-send-email-edward.c.wang@compdigitec.com>

---
 modules/gui/qt4/dialogs/toolbar.hpp |    4 ++--
 1 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/modules/gui/qt4/dialogs/toolbar.hpp b/modules/gui/qt4/dialogs/toolbar.hpp
index 374f42e..d9c0a5a 100644
--- a/modules/gui/qt4/dialogs/toolbar.hpp
+++ b/modules/gui/qt4/dialogs/toolbar.hpp
@@ -31,13 +31,13 @@
 #include <QCheckBox>
 
 #define PROFILE_NAME_1 "VLC 1.1.x Style"
-#define VALUE_1 "0|64;39;64;38;65;|0-2;64;3;1;4;64;7;9;64;10;20;19;64-4;37;65;35-4;|12;11;13;14;|5-1;33;6-1;|0-2;64;3;1;4;64;37;64;38;64;8;65;35-4;34;"
+#define VALUE_1 "0|64;39;64;38;65;|0-2;64;3;1;4;64;7;9;64;10;20;19;64-4;37;65;35-4;|12;11;13;14;|5-1;33;6-1;|0-2;64;3;1;4;64;37;64;38;64;8;65;25;35-4;34;"
 #define PROFILE_NAME_2 "VLC 0.8.x Style"
 #define VALUE_2 "1|64;39-1;64;38;|2-1;32-4;0-5;1-5;32-5;3-5;5-5;6-5;4-5;32-5;10-1;64-1;35-1;65;|12-1;11-1;13-1;14-1;|33;37-4;|0-5;1-5;32-1;1-5;5-1;6-1;4-5;32-1;12-5;11-1;65;34-4;35-1;"
 #define PROFILE_NAME_3 "Minimalist Style"
 #define VALUE_3 "0|64;65;|0-7;64;3-1;1-5;4-1;64;12-5;64-5;37-5;38-5;64-4;10-1;65;36-4;|11-5;13-5;14-5;|5-1;33;6-1;|0-5;64;3-5;1-5;4-5;64;12-5;65;34-4;35-1;"
 #define PROFILE_NAME_4 "One-Liner Style"
-#define VALUE_4 "0|64;38;65;|0-4;64;3;1;4;64;7;10;9;64-4;39;64-4;37;65;36-4;|12;11;14;13;|5-1;33;6-1;|0-2;64;3;1;4;64;37;64;38;64;8;65;35-4;34;"
+#define VALUE_4 "0|64;38;65;|0-4;64;3;1;4;64;7;10;9;64-4;39;64-4;37;65;36-4;|12;11;14;13;|5-1;33;6-1;|0-2;64;3;1;4;64;37;64;38;64;8;65;25;35-4;34;"
 #define PROFILE_NAME_5 "Simplest Style"
 #define VALUE_5 "0||36-4;65-4;0-7;4-5;1-5;65-4;7-5;|12-4;11;13;||36-4;65-4;0-5;4-5;1-5;65;8-5;"
 
-- 
1.7.5.4



From typx at dinauz.org  Sat Dec 31 00:21:32 2011
From: typx at dinauz.org (Denis Charmet)
Date: Sat, 31 Dec 2011 00:21:32 +0100
Subject: [vlc-devel] Blockers for VLC 1.2
In-Reply-To: <20111230165133.GA15175@videolan.org>
References: <20111230165133.GA15175@videolan.org>
Message-ID: <20111230232132.GA8116@dinauz.org>

Le vendredi 30 d?cembre 2011 ? 05:51:33, Jean-Baptiste Kempf a ?crit :
>  - MKV subtitles across segments (#5625)

It works if you set a language preference, to do it automatically I fear
my previous patch is needed. I'll investigate further.

Plus the picture pool "refcount > 0" assertion failed with ffmpeg-mt
should be added as blocking stuff.

-- 
TypX
Le mauvais esprit est un art de vivre


From jerome at taodyne.com  Sat Dec 31 00:48:01 2011
From: jerome at taodyne.com (=?iso-8859-1?Q?J=E9r=F4me_FORISSIER?=)
Date: Sat, 31 Dec 2011 00:48:01 +0100
Subject: [vlc-devel] Blockers for VLC 1.2
In-Reply-To: <20111230165708.GA17150@videolan.org>
References: <4EFDECEB.6090505@gmail.com> <20111230165708.GA17150@videolan.org>
Message-ID: <2D1F8A3A-8264-4DB0-B5F9-E7F7EA350144@taodyne.com>


On 30 d?c. 2011, at 17:57, Jean-Baptiste Kempf wrote:

> On Fri, Dec 30, 2011 at 11:55:07AM -0500, Rafa?l Carr? wrote :
>> Why is it a blocker?
> 
> Maybe not a blocker, but a very important issue, I think, notably if we
> want to promote libVLC...

+1

On a side note, in my code libvlc_media_get_tracks_info() returns zero u.video.i_width and  u.video.i_height for some videos.

-- 
J?r?me



From jerome at taodyne.com  Sat Dec 31 00:52:49 2011
From: jerome at taodyne.com (=?iso-8859-1?Q?J=E9r=F4me_FORISSIER?=)
Date: Sat, 31 Dec 2011 00:52:49 +0100
Subject: [vlc-devel] Blockers for VLC 1.2
In-Reply-To: <20111230232132.GA8116@dinauz.org>
References: <20111230165133.GA15175@videolan.org>
	<20111230232132.GA8116@dinauz.org>
Message-ID: <89A1F096-3F7E-44E9-927D-E8C738B4C8D8@taodyne.com>


On 31 d?c. 2011, at 00:21, Denis Charmet wrote:

> Le vendredi 30 d?cembre 2011 ? 05:51:33, Jean-Baptiste Kempf a ?crit :
>> - MKV subtitles across segments (#5625)
> 
> It works if you set a language preference, to do it automatically I fear
> my previous patch is needed. I'll investigate further.
> 
> Plus the picture pool "refcount > 0" assertion failed with ffmpeg-mt
> should be added as blocking stuff.

+1

This assertion has triggered with my code, on Windows 7. I've not observed the problem on MacOSX, though.

-- 
J?r?me



From mapei_ at hotmail.com  Sat Dec 31 02:34:18 2011
From: mapei_ at hotmail.com (Michael E)
Date: Sat, 31 Dec 2011 01:34:18 +0000
Subject: [vlc-devel] Blockers for VLC 1.2
In-Reply-To: <20111230165133.GA15175@videolan.org>
References: <20111230165133.GA15175@videolan.org>
Message-ID: <BLU130-W422BF0F432F9A7B2F86D1E0930@phx.gbl>




> Date: Fri, 30 Dec 2011 17:51:33 +0100
> From: jb at videolan.org
> To: vlc-devel at videolan.org
> Subject: [vlc-devel] Blockers for VLC 1.2
> 
> Hello,
> 
>  - libvlc_video_get_size is still not working: is it the window size or
>    the video size? (#3679)

When using the vmem interface without a desktop window (as I do), one would assume that  libvlc_video_get_size() would return the size of the video.

> Any good input would be valued.
> 
> Best regards,
> 
> -- 
> Jean-Baptiste Kempf

 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111231/2c19b8d2/attachment.html>

From jb at videolan.org  Sat Dec 31 02:38:50 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Sat, 31 Dec 2011 02:38:50 +0100
Subject: [vlc-devel] Blockers for VLC 1.2
In-Reply-To: <20111230232132.GA8116@dinauz.org>
References: <20111230165133.GA15175@videolan.org>
	<20111230232132.GA8116@dinauz.org>
Message-ID: <20111231013850.GA18157@videolan.org>

On Sat, Dec 31, 2011 at 12:21:32AM +0100, Denis Charmet wrote :
> Plus the picture pool "refcount > 0" assertion failed with ffmpeg-mt
> should be added as blocking stuff.

Please trac it.


-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From akiki at free.fr  Fri Dec 30 12:14:24 2011
From: akiki at free.fr (akiki at free.fr)
Date: Fri, 30 Dec 2011 12:14:24 +0100 (CET)
Subject: [vlc-devel] [VLC] #5741: http Playlist items authentication
In-Reply-To: <060.795fff077bf6a7ab8de004c2fe1ab658@videolan.org>
Message-ID: <3916a9e7-66bc-4394-8eef-d7eb5c195112@zimbra76-e14.priv.proxad.net>

Hi,

I am very surprised for your Notation of the problem

What do you think of a playlist where I must to enter at each song my identification with a password
You never use it later and recommand Windows Media Player to play such a list.

I am sure to be right - I am far from paradize ...

Sorry

----- Mail original -----
De: "VLC" <trac at videolan.org>
Envoy?: Vendredi 30 D?cembre 2011 00:35:44
Objet: Re: [VLC] #5741: http Playlist items authentication

#5741: http Playlist items authentication
------------------------------------+-----------------------------
   Reporter:  akiki                 |        Owner:  jb
       Type:  defect                |       Status:  new
   Priority:  normal                |    Milestone:  Bugs paradize
  Component:  Demuxers: Playlist    |      Version:  master git
   Severity:  normal                |   Resolution:
   Keywords:  playlist access http  |   Difficulty:  unknown
Platform(s):  all                   |  Work status:  Not started
------------------------------------+-----------------------------
Changes (by jb):

 * priority:  high => normal
 * platform:  Win32 => all
 * severity:  critical => normal
 * milestone:  1.1.13 => Bugs paradize


Comment:

 Clearly not critical...

-- 
Ticket URL: <http://trac.videolan.org/vlc/ticket/5741#comment:1>
VLC <http://www.videolan.org/vlc/>
VLC media player


From jb at videolan.org  Sat Dec 31 03:27:08 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Sat, 31 Dec 2011 03:27:08 +0100
Subject: [vlc-devel] Blockers for VLC 1.2
In-Reply-To: <20111230165133.GA15175@videolan.org>
References: <20111230165133.GA15175@videolan.org>
Message-ID: <20111231022708.GA29034@videolan.org>

On Fri, Dec 30, 2011 at 05:51:33PM +0100, Jean-Baptiste Kempf wrote :
> We have:
>  - UI, UI, UI, UI
>  - DVB is broken on Win32 (#5627)
>  - libvlc_video_get_size is still not working: is it the window size or
>    the video size? (#3679)
>  - MKV subtitles across segments (#5625)
>  - Vobsub subtitle placement (#5683)
>  - UI, UI, UI

- We have a resize red quality issue on Win32, #5751 too.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Sat Dec 31 03:33:30 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Sat, 31 Dec 2011 03:33:30 +0100
Subject: [vlc-devel] [PATCH] Qt4: Show fullscreen docking button in
	defaults
In-Reply-To: <1325285331-25517-1-git-send-email-edward.c.wang@compdigitec.com>
References: <1325285331-25517-1-git-send-email-edward.c.wang@compdigitec.com>
Message-ID: <4EFE747A.6020608@videolan.org>

On 30/12/2011 23:48, Edward Wang wrote:
> ---
>   modules/gui/qt4/dialogs/toolbar.hpp |    4 ++--
>   1 files changed, 2 insertions(+), 2 deletions(-)
>

Applied.


-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Sat Dec 31 03:50:32 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Sat, 31 Dec 2011 03:50:32 +0100
Subject: [vlc-devel] [PATCH 1/4] npapi win32: added protection from
 multiple calling of create_windows(); 
In-Reply-To: <1325241996-2844-1-git-send-email-RSATom@gmail.com>
References: <1325241996-2844-1-git-send-email-RSATom@gmail.com>
Message-ID: <20111231025032.GA31182@videolan.org>

On Fri, Dec 30, 2011 at 05:46:33PM +0700, Sergey Radionov wrote :
>  npapi/vlcplugin_win.cpp |    3 +++

Applied.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Sat Dec 31 03:50:41 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Sat, 31 Dec 2011 03:50:41 +0100
Subject: [vlc-devel] [PATCH 2/4] npapi win32: added window proc
 replacing for browser drawable window.
In-Reply-To: <1325241996-2844-2-git-send-email-RSATom@gmail.com>
References: <1325241996-2844-1-git-send-email-RSATom@gmail.com>
	<1325241996-2844-2-git-send-email-RSATom@gmail.com>
Message-ID: <20111231025041.GB31182@videolan.org>

On Fri, Dec 30, 2011 at 05:46:34PM +0700, Sergey Radionov wrote :
> some browsers (mainly opera) have not good realization of NPApi,
> and we will need track some events via native window messages,
> rather than via NPP_SetWindow.

Applied.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Sat Dec 31 03:50:51 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Sat, 31 Dec 2011 03:50:51 +0100
Subject: [vlc-devel] [PATCH 3/4] npapi win32: detach VlcPlugin from
 drawable on windows destroying.
In-Reply-To: <1325241996-2844-3-git-send-email-RSATom@gmail.com>
References: <1325241996-2844-1-git-send-email-RSATom@gmail.com>
	<1325241996-2844-3-git-send-email-RSATom@gmail.com>
Message-ID: <20111231025051.GC31182@videolan.org>

On Fri, Dec 30, 2011 at 05:46:35PM +0700, Sergey Radionov wrote :
>  npapi/vlcplugin_win.cpp |    2 ++

Applied.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From jb at videolan.org  Sat Dec 31 03:51:02 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Sat, 31 Dec 2011 03:51:02 +0100
Subject: [vlc-devel] [PATCH 4/4] npapi win32: call destroy_windows on
 drawable destroying (opera compatibility fix)
In-Reply-To: <1325242172-1968-4-git-send-email-RSATom@gmail.com>
References: <1325242172-1968-4-git-send-email-RSATom@gmail.com>
Message-ID: <20111231025102.GD31182@videolan.org>

On Fri, Dec 30, 2011 at 05:49:32PM +0700, Sergey Radionov wrote :
>  npapi/vlcplugin_win.cpp |    6 ++++++

Applied.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From edward.c.wang at compdigitec.com  Sat Dec 31 03:58:12 2011
From: edward.c.wang at compdigitec.com (Edward Wang)
Date: Fri, 30 Dec 2011 21:58:12 -0500
Subject: [vlc-devel] [PATCH] Qt4: Make playlist window drag&drop to current
	selected PLSelItem (updated)
Message-ID: <1325300292-31859-1-git-send-email-edward.c.wang@compdigitec.com>

This patch updated to latest git will make playlist window drag and drop to current selection (either media library or playlist). This should make it somewhat more user friendly because it drops into where the user expects it to.

Fixes #4873
Thanks,
	Edward Wang


---
 modules/gui/qt4/components/playlist/playlist.cpp |    9 ++++++---
 modules/gui/qt4/components/playlist/playlist.hpp |    2 ++
 modules/gui/qt4/components/playlist/selector.cpp |    5 +++++
 modules/gui/qt4/components/playlist/selector.hpp |    1 +
 modules/gui/qt4/main_interface.cpp               |   15 ++++++++++++---
 modules/gui/qt4/main_interface.hpp               |    3 ++-
 6 files changed, 28 insertions(+), 7 deletions(-)

diff --git a/modules/gui/qt4/components/playlist/playlist.cpp b/modules/gui/qt4/components/playlist/playlist.cpp
index 661f376..a0eb4cd 100644
--- a/modules/gui/qt4/components/playlist/playlist.cpp
+++ b/modules/gui/qt4/components/playlist/playlist.cpp
@@ -63,8 +63,8 @@ PlaylistWidget::PlaylistWidget( intf_thread_t *_p_i, QWidget *_par )
     leftSplitter = new QSplitter( Qt::Vertical, this );
 
     /* Source Selector */
-    PLSelector *selector = new PLSelector( this, p_intf );
-    leftSplitter->addWidget( selector);
+    selector = new PLSelector( this, p_intf );
+    leftSplitter->addWidget( selector );
 
     /* Create a Container for the Art Label
        in order to have a beautiful resizing for the selector above it */
@@ -211,8 +211,11 @@ PlaylistWidget::~PlaylistWidget()
 
 void PlaylistWidget::dropEvent( QDropEvent *event )
 {
+    if( !( selector->getCurrentItemCategory() == IS_PL ||
+           selector->getCurrentItemCategory() == IS_ML ) ) return;
+
     if( p_intf->p_sys->p_mi )
-        p_intf->p_sys->p_mi->dropEventPlay( event, false );
+        p_intf->p_sys->p_mi->dropEventPlay( event, false, (selector->getCurrentItemCategory() == IS_PL) );
 }
 void PlaylistWidget::dragEnterEvent( QDragEnterEvent *event )
 {
diff --git a/modules/gui/qt4/components/playlist/playlist.hpp b/modules/gui/qt4/components/playlist/playlist.hpp
index 7172f95..7d10eb6 100644
--- a/modules/gui/qt4/components/playlist/playlist.hpp
+++ b/modules/gui/qt4/components/playlist/playlist.hpp
@@ -46,6 +46,7 @@ class QSignalMapper;
 class SearchLineEdit;
 class QModelIndex;
 class QStackedWidget;
+class PLSelector;
 
 class PlaylistWidget : public QWidget
 {
@@ -61,6 +62,7 @@ private:
     QSplitter            *leftSplitter;
     QSplitter            *split;
     StandardPLPanel      *mainView;
+    PLSelector           *selector;
 
     QAction *viewActions[ 4 /* StandardPLPanel::VIEW_COUNT*/ ];
 
diff --git a/modules/gui/qt4/components/playlist/selector.cpp b/modules/gui/qt4/components/playlist/selector.cpp
index bb64382..693d454 100644
--- a/modules/gui/qt4/components/playlist/selector.cpp
+++ b/modules/gui/qt4/components/playlist/selector.cpp
@@ -534,6 +534,11 @@ void PLSelector::getCurrentSelectedItem( int* type, QString *string)
     *string = currentItem()->data( 0, NAME_ROLE ).toString();
 }
 
+int PLSelector::getCurrentItemCategory()
+{
+    return currentItem()->data( 0, SPECIAL_ROLE ).toInt();
+}
+
 void PLSelector::wheelEvent( QWheelEvent *e )
 {
     // Accept this event in order to prevent unwanted volume up/down changes
diff --git a/modules/gui/qt4/components/playlist/selector.hpp b/modules/gui/qt4/components/playlist/selector.hpp
index 7657fa2..21d2977 100644
--- a/modules/gui/qt4/components/playlist/selector.hpp
+++ b/modules/gui/qt4/components/playlist/selector.hpp
@@ -117,6 +117,7 @@ public:
     virtual ~PLSelector();
 
     void getCurrentSelectedItem( int *type, QString *name );
+    int getCurrentItemCategory();
 
 protected:
     virtual void drawBranches ( QPainter *, const QRect &, const QModelIndex & ) const;
diff --git a/modules/gui/qt4/main_interface.cpp b/modules/gui/qt4/main_interface.cpp
index 0e88231..19c65a8 100644
--- a/modules/gui/qt4/main_interface.cpp
+++ b/modules/gui/qt4/main_interface.cpp
@@ -1192,7 +1192,16 @@ void MainInterface::dropEvent(QDropEvent *event)
     dropEventPlay( event, true );
 }
 
-void MainInterface::dropEventPlay( QDropEvent *event, bool b_play )
+/**
+ * dropEventPlay
+ *
+ * Event called if something is dropped onto a VLC window
+ * \param event the event in question
+ * \param b_play whether to play the file immediately
+ * \param b_playlist true to add to playlist, false to add to media library
+ * \return nothing
+ */
+void MainInterface::dropEventPlay( QDropEvent *event, bool b_play, bool b_playlist )
 {
     if( event->possibleActions() & ( Qt::CopyAction | Qt::MoveAction ) )
        event->setDropAction( Qt::CopyAction );
@@ -1221,7 +1230,7 @@ void MainInterface::dropEventPlay( QDropEvent *event, bool b_play )
             QString mrl = toURI( url.toEncoded().constData() );
             playlist_Add( THEPL, qtu(mrl), NULL,
                           PLAYLIST_APPEND | (first ? PLAYLIST_GO: PLAYLIST_PREPARSE),
-                          PLAYLIST_END, true, pl_Unlocked );
+                          PLAYLIST_END, b_playlist, pl_Unlocked );
             first = false;
             RecentsMRL::getInstance( p_intf )->addRecent( mrl );
         }
@@ -1236,7 +1245,7 @@ void MainInterface::dropEventPlay( QDropEvent *event, bool b_play )
         QString mrl = toURI( mimeData->text() );
         playlist_Add( THEPL, qtu(mrl), NULL,
                       PLAYLIST_APPEND | (first ? PLAYLIST_GO: PLAYLIST_PREPARSE),
-                      PLAYLIST_END, true, pl_Unlocked );
+                      PLAYLIST_END, b_playlist, pl_Unlocked );
     }
     event->accept();
 }
diff --git a/modules/gui/qt4/main_interface.hpp b/modules/gui/qt4/main_interface.hpp
index 87e04d5..142d092 100644
--- a/modules/gui/qt4/main_interface.hpp
+++ b/modules/gui/qt4/main_interface.hpp
@@ -88,7 +88,8 @@ public:
     bool isInterfaceFullScreen() { return b_interfaceFullScreen; }
 
 protected:
-    void dropEventPlay( QDropEvent *, bool);
+    void dropEventPlay( QDropEvent* event, bool b_play ) { dropEventPlay(event, b_play, true); }
+    void dropEventPlay( QDropEvent *, bool, bool );
 #ifdef WIN32
     virtual bool winEvent( MSG *, long * );
 #endif
-- 
1.7.5.4



From jb at videolan.org  Sat Dec 31 04:03:41 2011
From: jb at videolan.org (Jean-Baptiste Kempf)
Date: Sat, 31 Dec 2011 04:03:41 +0100
Subject: [vlc-devel] [PATCH] Qt4: Make playlist window drag&drop to
 current selected PLSelItem (updated)
In-Reply-To: <1325300292-31859-1-git-send-email-edward.c.wang@compdigitec.com>
References: <1325300292-31859-1-git-send-email-edward.c.wang@compdigitec.com>
Message-ID: <20111231030341.GA3618@videolan.org>

On Fri, Dec 30, 2011 at 09:58:12PM -0500, Edward Wang wrote :
> This patch updated to latest git will make playlist window drag and drop to current selection (either media library or playlist). This should make it somewhat more user friendly because it drops into where the user expects it to.
> 
> Fixes #4873

Ok. Applied.

Best regards,

-- 
Jean-Baptiste Kempf
http://www.jbkempf.com/ - +33 672 704 734
Sent from my Electronic Device


From kaarlo.raiha at gmail.com  Sat Dec 31 08:58:25 2011
From: kaarlo.raiha at gmail.com (=?ISO-8859-1?B?S2FhcmxvIFLkaWjk?=)
Date: Sat, 31 Dec 2011 09:58:25 +0200
Subject: [vlc-devel] [VLC] #5741: http Playlist items authentication
In-Reply-To: <3916a9e7-66bc-4394-8eef-d7eb5c195112@zimbra76-e14.priv.proxad.net>
References: <060.795fff077bf6a7ab8de004c2fe1ab658@videolan.org>
	<3916a9e7-66bc-4394-8eef-d7eb5c195112@zimbra76-e14.priv.proxad.net>
Message-ID: <CAMKrNtxYcu4MS8HQtVw90zP58zY65FK0KXmcsTXW9wbPUnif5Q@mail.gmail.com>

Can we change the trac system, so that only certain people can change
priority and severity of tickets? Because the current system causes too
much stress/fighting when people (who don't read the descriptions) try to
cut the line.

2011/12/30 <akiki at free.fr>

> Hi,
>
> I am very surprised for your Notation of the problem
>
> What do you think of a playlist where I must to enter at each song my
> identification with a password
> You never use it later and recommand Windows Media Player to play such a
> list.
>
> I am sure to be right - I am far from paradize ...
>
> Sorry
>
> ----- Mail original -----
> De: "VLC" <trac at videolan.org>
> Envoy?: Vendredi 30 D?cembre 2011 00:35:44
> Objet: Re: [VLC] #5741: http Playlist items authentication
>
> #5741: http Playlist items authentication
> ------------------------------------+-----------------------------
>   Reporter:  akiki                 |        Owner:  jb
>       Type:  defect                |       Status:  new
>   Priority:  normal                |    Milestone:  Bugs paradize
>  Component:  Demuxers: Playlist    |      Version:  master git
>   Severity:  normal                |   Resolution:
>   Keywords:  playlist access http  |   Difficulty:  unknown
> Platform(s):  all                   |  Work status:  Not started
> ------------------------------------+-----------------------------
> Changes (by jb):
>
>  * priority:  high => normal
>  * platform:  Win32 => all
>  * severity:  critical => normal
>  * milestone:  1.1.13 => Bugs paradize
>
>
> Comment:
>
>  Clearly not critical...
>
> --
> Ticket URL: <http://trac.videolan.org/vlc/ticket/5741#comment:1>
> VLC <http://www.videolan.org/vlc/>
> VLC media player
> _______________________________________________
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
> http://mailman.videolan.org/listinfo/vlc-devel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20111231/d8442656/attachment.html>

From jpsaman at videolan.org  Sat Dec 31 10:32:38 2011
From: jpsaman at videolan.org (Jean-Paul Saman)
Date: Sat, 31 Dec 2011 10:32:38 +0100
Subject: [vlc-devel] [PATCH 1/2] httplive: fix Read func. (when caller
 skips data)
In-Reply-To: <4efd900a.0d12980a.3671.30bb@mx.google.com>
References: <4efd900a.0d12980a.3671.30bb@mx.google.com>
Message-ID: <CAK9US3CVF9sgfG=xE+v2wQqCHLemfzpDBNwZT18gNr-Zcp_1Kw@mail.gmail.com>

This patch is wrong p_read should have been checked earlier.

On Fri, Dec 30, 2011 at 11:18 AM,  <fyhuel at viotech.net> wrote:
> From: Fr?d?ric Yhuel <fyhuel at viotech.net>
>
> ---
> ?modules/stream_filter/httplive.c | ? 12 ++----------
> ?1 files changed, 2 insertions(+), 10 deletions(-)
>
> diff --git a/modules/stream_filter/httplive.c b/modules/stream_filter/httplive.c
> index 4d60d8a..58dc1db 100644
> --- a/modules/stream_filter/httplive.c
> +++ b/modules/stream_filter/httplive.c
> @@ -1912,7 +1912,8 @@ static ssize_t hls_Read(stream_t *s, uint8_t *p_read, unsigned int i_read)
>
> ? ? ? ? if (len > 0)
> ? ? ? ? {
> - ? ? ? ? ? ?memcpy(p_read + copied, segment->data->p_buffer, len);
> + ? ? ? ? ? ?if( p_read ) /* otherwise caller skips data */
> + ? ? ? ? ? ? ? ?memcpy(p_read + copied, segment->data->p_buffer, len);

Below you update the segment data buffer while p_read is NULL, which
is the wrong thing to do.

> ? ? ? ? ? ? segment->data->i_buffer -= len;
> ? ? ? ? ? ? segment->data->p_buffer += len;
> ? ? ? ? ? ? copied += len;
> @@ -1935,15 +1936,6 @@ static int Read(stream_t *s, void *buffer, unsigned int i_read)
> ? ? if (p_sys->b_error)
> ? ? ? ? return 0;
>
> - ? ?if (buffer == NULL)
> - ? ?{
> - ? ? ? ?/* caller skips data, get big enough buffer */
> - ? ? ? ?msg_Warn(s, "buffer is NULL (allocate %d)", i_read);
> - ? ? ? ?buffer = calloc(1, i_read);
> - ? ? ? ?if (buffer == NULL)
> - ? ? ? ? ? ?return 0; /* NO MEMORY left*/
> - ? ?}
> -
> ? ? length = hls_Read(s, (uint8_t*) buffer, i_read);
> ? ? if (length < 0)
> ? ? ? ? return 0;
> --
> 1.7.5.4
>
> _______________________________________________
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
> http://mailman.videolan.org/listinfo/vlc-devel

Kind regards,

Jean-Paul Saman


From fyhuel at viotech.net  Sat Dec 31 11:43:05 2011
From: fyhuel at viotech.net (=?utf-8?Q?Fr=C3=A9d=C3=A9ric_Yhuel?=)
Date: Sat, 31 Dec 2011 11:43:05 +0100
Subject: [vlc-devel] [PATCH 1/2] httplive: fix Read func. (when caller
	skips data)
In-Reply-To: <CAK9US3CVF9sgfG=xE+v2wQqCHLemfzpDBNwZT18gNr-Zcp_1Kw@mail.gmail.com>
References: <4efd900a.0d12980a.3671.30bb@mx.google.com>
	<CAK9US3CVF9sgfG=xE+v2wQqCHLemfzpDBNwZT18gNr-Zcp_1Kw@mail.gmail.com>
Message-ID: <B45F2F87-E1F0-455F-B4BD-46D8FFAA2BEB@viotech.net>

This patch is indeed wrong... I'll send a new one and I will properly test it!

Sorry I lost your time...

(and sorry for top posting)

--
Fr?d?ric


Le 31 d?c. 2011 ? 10:32, Jean-Paul Saman <jpsaman at videolan.org> a ?crit :

> This patch is wrong p_read should have been checked earlier.
> 
> On Fri, Dec 30, 2011 at 11:18 AM,  <fyhuel at viotech.net> wrote:
>> From: Fr?d?ric Yhuel <fyhuel at viotech.net>
>> 
>> ---
>>  modules/stream_filter/httplive.c |   12 ++----------
>>  1 files changed, 2 insertions(+), 10 deletions(-)
>> 
>> diff --git a/modules/stream_filter/httplive.c b/modules/stream_filter/httplive.c
>> index 4d60d8a..58dc1db 100644
>> --- a/modules/stream_filter/httplive.c
>> +++ b/modules/stream_filter/httplive.c
>> @@ -1912,7 +1912,8 @@ static ssize_t hls_Read(stream_t *s, uint8_t *p_read, unsigned int i_read)
>> 
>>         if (len > 0)
>>         {
>> -            memcpy(p_read + copied, segment->data->p_buffer, len);
>> +            if( p_read ) /* otherwise caller skips data */
>> +                memcpy(p_read + copied, segment->data->p_buffer, len);
> 
> Below you update the segment data buffer while p_read is NULL, which
> is the wrong thing to do.
> 
>>             segment->data->i_buffer -= len;
>>             segment->data->p_buffer += len;
>>             copied += len;
>> @@ -1935,15 +1936,6 @@ static int Read(stream_t *s, void *buffer, unsigned int i_read)
>>     if (p_sys->b_error)
>>         return 0;
>> 
>> -    if (buffer == NULL)
>> -    {
>> -        /* caller skips data, get big enough buffer */
>> -        msg_Warn(s, "buffer is NULL (allocate %d)", i_read);
>> -        buffer = calloc(1, i_read);
>> -        if (buffer == NULL)
>> -            return 0; /* NO MEMORY left*/
>> -    }
>> -
>>     length = hls_Read(s, (uint8_t*) buffer, i_read);
>>     if (length < 0)
>>         return 0;
>> --
>> 1.7.5.4
>> 
>> _______________________________________________
>> vlc-devel mailing list
>> To unsubscribe or modify your subscription options:
>> http://mailman.videolan.org/listinfo/vlc-devel
> 
> Kind regards,
> 
> Jean-Paul Saman
> _______________________________________________
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
> http://mailman.videolan.org/listinfo/vlc-devel


From komh78 at gmail.com  Sat Dec 31 12:19:46 2011
From: komh78 at gmail.com (KO Myung-Hun)
Date: Sat, 31 Dec 2011 20:19:46 +0900
Subject: [vlc-devel] [PATCH 2/3] Define the some macros to avoid name
	clashes on OS/2.
In-Reply-To: <4EF1C7CC.1050503@chollian.net>
References: <1324380587-225-1-git-send-email-komh@chollian.net>	<1324380587-225-3-git-send-email-komh@chollian.net>	<20111221003722.GA22705@videolan.org>
	<4EF1C7CC.1050503@chollian.net>
Message-ID: <4EFEEFD2.5050508@chollian.net>



KO Myung-Hun wrote:
> 
> 
> Jean-Baptiste Kempf wrote:
>> On Tue, Dec 20, 2011 at 08:29:46PM +0900, KO Myung-Hun wrote :
>>> They are
>>>
>>>     #define BITMAPINFOHEADER    VLC_BITMAPINFOHEADER
>>>     #define PBITMAPINFOHEADER   VLC_PBITMAPINFOHEADER
>>>     #define LPBITMAPINFOHEADER  VLC_LPBITMAPINFOHEADER
>>>     #define BITMAPINFO          VLC_BITMAPINFO
>>>     #define LPBITMAPINFO        VLC_LPBITMAPINFO
>>>
>>> OS/2 has already the types of the same name, but the different member name.
>>
>> I am really not at ease with that...
>>
>> BITMAPINFOHEADER is quite defined by MicroSoft. Any way to undef the
>> OS/2 defines?
>>
>>
> 
> Unfortunately, it is also defined by IBM.
> 
> I'm sorry I don't know how to undef 'typedef'.
> 

Ping ?
-- 
KO Myung-Hun

Using Mozilla SeaMonkey 2.0.14
Under OS/2 Warp 4 for Korean with FixPak #15
On AMD ThunderBird 1GHz with 512 MB RAM

Korean OS/2 User Community : http://www.ecomstation.co.kr



From nirankoon at nuol.edu.la  Sat Dec 31 10:03:02 2011
From: nirankoon at nuol.edu.la (Nirankoon SINGPRASEUTH)
Date: Sat, 31 Dec 2011 16:03:02 +0700
Subject: [vlc-devel] LAOS isn't not in list Localization
Message-ID: <CAPXBDCsXE13iRgUvF49-xozZEDSWpfJggFXci_52naY8UVAuJw@mail.gmail.com>

Dear Sir vlc-devel

I am Nirankoon SINGPRASEUTH, i am work at National University at Laos.
Laos is in South east Asia near Thailand,Cambodia,Vietnam,China and Myanmar
Laos Doesn't have in List Localization.
I am very interested in VLC Player.
VLC is Very Popular Player in Laos (Population in Laos around 6Million Peoples).
So want to Translate this application into Lao Language.
Please Advice me.

Thank.
Nirankoon SINGPRASEUTH


From typx at dinauz.org  Sat Dec 31 13:30:18 2011
From: typx at dinauz.org (Denis Charmet)
Date: Sat, 31 Dec 2011 13:30:18 +0100
Subject: [vlc-devel] Blockers for VLC 1.2
In-Reply-To: <20111231013850.GA18157@videolan.org>
References: <20111230165133.GA15175@videolan.org>
	<20111230232132.GA8116@dinauz.org>
	<20111231013850.GA18157@videolan.org>
Message-ID: <20111231123018.GA15836@dinauz.org>

Le samedi 31 d?cembre 2011 ? 02:38:50, Jean-Baptiste Kempf a ?crit :
> On Sat, Dec 31, 2011 at 12:21:32AM +0100, Denis Charmet wrote :
> > Plus the picture pool "refcount > 0" assertion failed with ffmpeg-mt
> > should be added as blocking stuff.
> 
> Please trac it.
>
Alright it's #5753
-- 
TypX
Le mauvais esprit est un art de vivre


From edward.c.wang at compdigitec.com  Sat Dec 31 16:18:16 2011
From: edward.c.wang at compdigitec.com (Edward Wang)
Date: Sat, 31 Dec 2011 10:18:16 -0500
Subject: [vlc-devel] [PATCH] Qt4: UI enhancements
Message-ID: <1325344696-13339-1-git-send-email-edward.c.wang@compdigitec.com>

This patch adds separate elapsed and total/remaining labels, as well as a
horizontal speed slider with doublespinbox.

Thanks,
	Edward Wang

---
 modules/gui/qt4/components/controller.cpp        |   16 ++++-
 modules/gui/qt4/components/controller.hpp        |    5 +-
 modules/gui/qt4/components/interface_widgets.cpp |   91 +++++++++++++++++----
 modules/gui/qt4/components/interface_widgets.hpp |   18 ++++-
 modules/gui/qt4/dialogs/toolbar.cpp              |   12 +++
 modules/gui/qt4/main_interface.cpp               |    2 +-
 6 files changed, 120 insertions(+), 24 deletions(-)

diff --git a/modules/gui/qt4/components/controller.cpp b/modules/gui/qt4/components/controller.cpp
index 22d3792..6a4bfac 100644
--- a/modules/gui/qt4/components/controller.cpp
+++ b/modules/gui/qt4/components/controller.cpp
@@ -380,7 +380,7 @@ QWidget *AbstractController::createWidget( buttonType_e button, int options )
         break;
     case TIME_LABEL:
         {
-            TimeLabel *timeLabel = new TimeLabel( p_intf );
+            TimeLabel *timeLabel = new TimeLabel( p_intf, TimeLabel::Both );
             widget = timeLabel;
         }
         break;
@@ -468,6 +468,20 @@ QWidget *AbstractController::createWidget( buttonType_e button, int options )
     case ASPECT_RATIO_COMBOBOX:
         widget = new AspectRatioComboBox( p_intf );
         break;
+    case SPEED_LABEL:
+        {
+            SpeedLabel* label = new SpeedLabel( p_intf, this );
+            label->setFrameStyle( QFrame::StyledPanel | QFrame::Raised );
+            label->setLineWidth( 1 );
+            widget = label;
+        }
+        break;
+    case TIME_LABEL_ELAPSED:
+        widget = new TimeLabel( p_intf, TimeLabel::Elapsed );
+        break;
+    case TIME_LABEL_REMAINING:
+        widget = new TimeLabel( p_intf, TimeLabel::Remaining );
+        break;
     default:
         msg_Warn( p_intf, "This should not happen %i", button );
         break;
diff --git a/modules/gui/qt4/components/controller.hpp b/modules/gui/qt4/components/controller.hpp
index 96d53c0..1f0743d 100644
--- a/modules/gui/qt4/components/controller.hpp
+++ b/modules/gui/qt4/components/controller.hpp
@@ -37,7 +37,7 @@
 #define MAIN_TB1_DEFAULT "64;39;64;38;65"
 #define MAIN_TB2_DEFAULT "0-2;64;3;1;4;64;7;9;64;10;20;19;64-4;37;65;35-4"
 #define ADV_TB_DEFAULT "12;11;13;14"
-#define INPT_TB_DEFAULT "5-1;15-1;33;6-1"
+#define INPT_TB_DEFAULT "43;33;44;42"
 #define FSC_TB_DEFAULT "0-2;64;3;1;4;64;37;64;38;64;8;65;25;35-4;34"
 
 #define I_PLAY_TOOLTIP N_("Play\nIf the playlist is empty, open a medium")
@@ -100,6 +100,9 @@ typedef enum buttonType_e
     ADVANCED_CONTROLLER,
     PLAYBACK_BUTTONS,
     ASPECT_RATIO_COMBOBOX,
+    SPEED_LABEL,
+    TIME_LABEL_ELAPSED,
+    TIME_LABEL_REMAINING,
     SPECIAL_MAX,
 
     WIDGET_SPACER = 0x40,
diff --git a/modules/gui/qt4/components/interface_widgets.cpp b/modules/gui/qt4/components/interface_widgets.cpp
index e384092..c83fd48 100644
--- a/modules/gui/qt4/components/interface_widgets.cpp
+++ b/modules/gui/qt4/components/interface_widgets.cpp
@@ -383,15 +383,15 @@ void SpeedLabel::setRate( float rate )
 SpeedControlWidget::SpeedControlWidget( intf_thread_t *_p_i, QWidget *_parent )
                     : QFrame( _parent ), p_intf( _p_i )
 {
-    QSizePolicy sizePolicy( QSizePolicy::Maximum, QSizePolicy::Fixed );
+    QSizePolicy sizePolicy( QSizePolicy::Fixed, QSizePolicy::Maximum );
     sizePolicy.setHorizontalStretch( 0 );
     sizePolicy.setVerticalStretch( 0 );
 
     speedSlider = new QSlider( this );
     speedSlider->setSizePolicy( sizePolicy );
-    speedSlider->setMaximumSize( QSize( 80, 200 ) );
-    speedSlider->setOrientation( Qt::Vertical );
-    speedSlider->setTickPosition( QSlider::TicksRight );
+    speedSlider->setMinimumSize( QSize( 200, 24 ) );
+    speedSlider->setOrientation( Qt::Horizontal );
+    speedSlider->setTickPosition( QSlider::TicksAbove );
 
     speedSlider->setRange( -34, 34 );
     speedSlider->setSingleStep( 1 );
@@ -408,11 +408,24 @@ SpeedControlWidget::SpeedControlWidget( intf_thread_t *_p_i, QWidget *_parent )
 
     CONNECT( normalSpeedButton, clicked(), this, resetRate() );
 
-    QVBoxLayout *speedControlLayout = new QVBoxLayout( this );
+    spinBox = new QDoubleSpinBox();
+    spinBox->setDecimals( 2 );
+    spinBox->setMaximum( 32 );
+    spinBox->setMinimum( 0.03F );
+    spinBox->setSingleStep( 0.10F );
+    spinBox->setAlignment( Qt::AlignRight );
+
+    CONNECT( spinBox, valueChanged( double ), this, updateSpinBoxRate( double ) );
+
+    QVBoxLayout* speedControlLayout = new QVBoxLayout( this );
+    QHBoxLayout* hboxLayout = new QHBoxLayout();
+    hboxLayout->addWidget( spinBox );
+    hboxLayout->addWidget( normalSpeedButton );
     speedControlLayout->setContentsMargins( 4, 4, 4, 4 );
     speedControlLayout->setSpacing( 4 );
     speedControlLayout->addWidget( speedSlider );
-    speedControlLayout->addWidget( normalSpeedButton );
+    speedControlLayout->addLayout( hboxLayout );
+    speedControlLayout->setAlignment( hboxLayout, Qt::AlignHCenter );
 
     lastValue = 0;
 
@@ -422,6 +435,7 @@ SpeedControlWidget::SpeedControlWidget( intf_thread_t *_p_i, QWidget *_parent )
 void SpeedControlWidget::activateOnState()
 {
     speedSlider->setEnabled( THEMIM->getIM()->hasInput() );
+    spinBox->setEnabled( THEMIM->getIM()->hasInput() );
 }
 
 void SpeedControlWidget::updateControls( float rate )
@@ -446,6 +460,7 @@ void SpeedControlWidget::updateControls( float rate )
     lastValue = sliderValue;
 
     speedSlider->setValue( sliderValue );
+    spinBox->setValue( rate );
 }
 
 void SpeedControlWidget::updateRate( int sliderValue )
@@ -457,6 +472,12 @@ void SpeedControlWidget::updateRate( int sliderValue )
     int rate = INPUT_RATE_DEFAULT / speed;
 
     THEMIM->getIM()->setRate(rate);
+    spinBox->setValue( var_InheritFloat( THEPL, "rate" ) );
+}
+
+void SpeedControlWidget::updateSpinBoxRate( double r )
+{
+    var_SetFloat( THEPL, "rate", r );
 }
 
 void SpeedControlWidget::resetRate()
@@ -512,17 +533,33 @@ void CoverArtLabel::askForUpdate()
     THEMIM->getIM()->requestArtUpdate();
 }
 
-TimeLabel::TimeLabel( intf_thread_t *_p_intf  )
+TimeLabel::TimeLabel( intf_thread_t *_p_intf, TimeLabel::Display _displayType  )
     : QLabel(), p_intf( _p_intf ), bufTimer( new QTimer(this) ),
-      buffering( false ), showBuffering(false), bufVal( -1 )
+      buffering( false ), showBuffering(false), bufVal( -1 ), displayType( _displayType )
 {
     b_remainingTime = false;
-    setText( " --:--/--:-- " );
+    switch( _displayType ) {
+        case TimeLabel::Elapsed:
+            setText( " --:-- " );
+            setToolTip( qtr("Elapsed time") );
+            break;
+        case TimeLabel::Remaining:
+            setText( " --:-- " );
+            setToolTip( qtr("Total/Remaining time")
+                        + QString("\n-")
+                        + qtr("Click to toggle between total and remaining time")
+                      );
+            break;
+        case TimeLabel::Both:
+            setText( " --:--/--:-- " );
+            setToolTip( QString( "- " )
+                + qtr( "Click to toggle between elapsed and remaining time" )
+                + QString( "\n- " )
+                + qtr( "Double click to jump to a chosen time position" ) );
+            break;
+    }
     setAlignment( Qt::AlignRight | Qt::AlignVCenter );
-    setToolTip( QString( "- " )
-        + qtr( "Click to toggle between elapsed and remaining time" )
-        + QString( "\n- " )
-        + qtr( "Double click to jump to a chosen time position" ) );
+
     bufTimer->setSingleShot( true );
 
     CONNECT( THEMIM->getIM(), positionUpdated( float, int64_t, int ),
@@ -530,6 +567,8 @@ TimeLabel::TimeLabel( intf_thread_t *_p_intf  )
     CONNECT( THEMIM->getIM(), cachingChanged( float ),
               this, updateBuffering( float ) );
     CONNECT( bufTimer, timeout(), this, updateBuffering() );
+
+    this->setContentsMargins( 4, 0, 4, 0 );
 }
 
 void TimeLabel::setDisplayPosition( float pos, int64_t t, int length )
@@ -539,7 +578,10 @@ void TimeLabel::setDisplayPosition( float pos, int64_t t, int length )
 
     if( pos == -1.f )
     {
-        setText( " --:--/--:-- " );
+        if( displayType == TimeLabel::Both )
+            setText( " --:--/--:-- " );
+        else
+            setText( " --:-- " );
         return;
     }
 
@@ -548,14 +590,27 @@ void TimeLabel::setDisplayPosition( float pos, int64_t t, int length )
     secstotimestr( psz_length, length );
     secstotimestr( psz_time, ( b_remainingTime && length ) ? length - time
                                                            : time );
-
-    QString timestr = QString( " %1%2/%3 " )
+    switch( displayType )
+    {
+        case TimeLabel::Elapsed:
+            setText( QString(" ") + QString( psz_time ) + QString(" ") );
+            break;
+        case TimeLabel::Remaining:
+            if( b_remainingTime )
+                setText( QString(" -") + QString( psz_time ) + QString(" ") );
+            else
+                setText( QString(" ") + QString( psz_length ) + QString(" ") );
+            break;
+        case TimeLabel::Both:
+        default:
+            QString timestr = QString( " %1%2/%3 " )
             .arg( QString( (b_remainingTime && length) ? "-" : "" ) )
             .arg( QString( psz_time ) )
             .arg( QString( ( !length && time ) ? "--:--" : psz_length ) );
 
-    setText( timestr );
-
+            setText( timestr );
+            break;
+    }
     cachedLength = length;
 }
 
diff --git a/modules/gui/qt4/components/interface_widgets.hpp b/modules/gui/qt4/components/interface_widgets.hpp
index a2ce18f..3026243 100644
--- a/modules/gui/qt4/components/interface_widgets.hpp
+++ b/modules/gui/qt4/components/interface_widgets.hpp
@@ -122,15 +122,24 @@ class TimeLabel : public QLabel
 {
     Q_OBJECT
 public:
-    TimeLabel( intf_thread_t *_p_intf );
+    enum Display
+    {
+        Elapsed,
+        Remaining,
+        Both
+    };
+
+    TimeLabel( intf_thread_t *_p_intf, TimeLabel::Display _displayType );
 protected:
     virtual void mousePressEvent( QMouseEvent *event )
     {
+        if( displayType == TimeLabel::Elapsed ) return;
         toggleTimeDisplay();
         event->accept();
     }
     virtual void mouseDoubleClickEvent( QMouseEvent *event )
     {
+        if( displayType != TimeLabel::Both ) return;
         event->accept();
         toggleTimeDisplay();
         emit timeLabelDoubleClicked();
@@ -144,6 +153,7 @@ private:
     bool buffering;
     bool showBuffering;
     float bufVal;
+    TimeLabel::Display displayType;
 
     char psz_length[MSTRTIME_MAX_SIZE];
     char psz_time[MSTRTIME_MAX_SIZE];
@@ -188,8 +198,9 @@ public:
     SpeedControlWidget( intf_thread_t *, QWidget * );
     void updateControls( float );
 private:
-    intf_thread_t *p_intf;
-    QSlider *speedSlider;
+    intf_thread_t* p_intf;
+    QSlider* speedSlider;
+    QDoubleSpinBox* spinBox;
     int lastValue;
 
 public slots:
@@ -197,6 +208,7 @@ public slots:
 
 private slots:
     void updateRate( int );
+    void updateSpinBoxRate( double );
     void resetRate();
 };
 
diff --git a/modules/gui/qt4/dialogs/toolbar.cpp b/modules/gui/qt4/dialogs/toolbar.cpp
index 43ebf09..bd1e982 100644
--- a/modules/gui/qt4/dialogs/toolbar.cpp
+++ b/modules/gui/qt4/dialogs/toolbar.cpp
@@ -449,6 +449,18 @@ WidgetListing::WidgetListing( intf_thread_t *p_intf, QWidget *_parent )
             widget = new AspectRatioComboBox( p_intf );
             widgetItem->setText( qtr("Aspect ratio selector") );
             break;
+        case SPEED_LABEL:
+            widget = new SpeedLabel( p_intf, this );
+            widgetItem->setText( qtr("Speed selector") );
+            break;
+        case TIME_LABEL_ELAPSED:
+            widget = new QLabel( "2:42", this );
+            widgetItem->setText( qtr("Elasped time") );
+            break;
+        case TIME_LABEL_REMAINING:
+            widget = new QLabel( "-2:42", this );
+            widgetItem->setText( qtr("Total/Remaining time") );
+            break;
         default:
             msg_Warn( p_intf, "This should not happen %i", i );
             break;
diff --git a/modules/gui/qt4/main_interface.cpp b/modules/gui/qt4/main_interface.cpp
index 19c65a8..21a8c9c 100644
--- a/modules/gui/qt4/main_interface.cpp
+++ b/modules/gui/qt4/main_interface.cpp
@@ -458,7 +458,7 @@ inline void MainInterface::createStatusBar()
     /* Widgets Creation*/
     QStatusBar *statusBarr = statusBar();
 
-    TimeLabel *timeLabel = new TimeLabel( p_intf );
+    TimeLabel *timeLabel = new TimeLabel( p_intf, TimeLabel::Both );
     nameLabel = new QLabel( this );
     nameLabel->setTextInteractionFlags( Qt::TextSelectableByMouse
                                       | Qt::TextSelectableByKeyboard );
-- 
1.7.5.4



From fyhuel at viotech.net  Sat Dec 31 19:18:20 2011
From: fyhuel at viotech.net (Frederic YHUEL)
Date: Sat, 31 Dec 2011 18:18:20 +0000
Subject: [vlc-devel] [PATCH 1/2] httplive: fix Read func. (when caller
 skips data)
In-Reply-To: <B45F2F87-E1F0-455F-B4BD-46D8FFAA2BEB@viotech.net>
References: <4efd900a.0d12980a.3671.30bb@mx.google.com>
	<CAK9US3CVF9sgfG=xE+v2wQqCHLemfzpDBNwZT18gNr-Zcp_1Kw@mail.gmail.com>
	<B45F2F87-E1F0-455F-B4BD-46D8FFAA2BEB@viotech.net>
Message-ID: <CAKHNsC1N+z2BxU7ZYEYMj+oDV5+pax6_pzrgu862MJEFNz+XeA@mail.gmail.com>

2011/12/31 Fr?d?ric Yhuel <fyhuel at viotech.net>:
> This patch is indeed wrong...
>

Jean-Paul, I reacted a tad too fast... Actually I think that patch is good.

>
>
> Le 31 d?c. 2011 ? 10:32, Jean-Paul Saman <jpsaman at videolan.org> a ?crit :
>
>> This patch is wrong p_read should have been checked earlier.
>>
>> On Fri, Dec 30, 2011 at 11:18 AM, ?<fyhuel at viotech.net> wrote:
>>> From: Fr?d?ric Yhuel <fyhuel at viotech.net>
>>>
>>> ---
>>> ?modules/stream_filter/httplive.c | ? 12 ++----------
>>> ?1 files changed, 2 insertions(+), 10 deletions(-)
>>>
>>> diff --git a/modules/stream_filter/httplive.c b/modules/stream_filter/httplive.c
>>> index 4d60d8a..58dc1db 100644
>>> --- a/modules/stream_filter/httplive.c
>>> +++ b/modules/stream_filter/httplive.c
>>> @@ -1912,7 +1912,8 @@ static ssize_t hls_Read(stream_t *s, uint8_t *p_read, unsigned int i_read)
>>>
>>> ? ? ? ? if (len > 0)
>>> ? ? ? ? {
>>> - ? ? ? ? ? ?memcpy(p_read + copied, segment->data->p_buffer, len);
>>> + ? ? ? ? ? ?if( p_read ) /* otherwise caller skips data */
>>> + ? ? ? ? ? ? ? ?memcpy(p_read + copied, segment->data->p_buffer, len);
>>
>> Below you update the segment data buffer while p_read is NULL, which
>> is the wrong thing to do.
>>

So what is the right thing to do? Do you agree that currently data is
read (instead of being skipped) when p_read is null?

Happy new year!

--
F?d?ric


From kelly at silka.with-linux.com  Sat Dec 31 23:26:14 2011
From: kelly at silka.with-linux.com (Kelly Anderson)
Date: Sat, 31 Dec 2011 15:26:14 -0700
Subject: [vlc-devel] [PATCH] Changes to support lua 5.2
Message-ID: <1325370374-28308-1-git-send-email-kelly@silka.with-linux.com>

---
 modules/lua/intf.c                      |    4 ----
 modules/lua/libs/acl.c                  |    3 ---
 modules/lua/libs/configuration.c        |    3 ---
 modules/lua/libs/dialog.c               |    3 ---
 modules/lua/libs/equalizer.c            |    3 ---
 modules/lua/libs/gettext.c              |    4 ----
 modules/lua/libs/httpd.c                |    4 ----
 modules/lua/libs/input.c                |    4 +---
 modules/lua/libs/input.h                |    2 ++
 modules/lua/libs/messages.c             |    4 ----
 modules/lua/libs/misc.c                 |    4 ----
 modules/lua/libs/net.c                  |    3 ---
 modules/lua/libs/objects.c              |    3 ---
 modules/lua/libs/osd.c                  |    3 ---
 modules/lua/libs/playlist.c             |    4 +---
 modules/lua/libs/sd.c                   |    4 ----
 modules/lua/libs/stream.c               |    3 ---
 modules/lua/libs/strings.c              |    3 ---
 modules/lua/libs/variables.c            |    4 ----
 modules/lua/libs/video.c                |    3 ---
 modules/lua/libs/vlm.c                  |    3 ---
 modules/lua/libs/volume.c               |    4 ----
 modules/lua/libs/xml.c                  |    3 ---
 modules/lua/vlc.c                       |    4 ----
 modules/lua/vlc.h                       |    6 ++++++
 share/lua/extensions/imdb.lua           |    2 +-
 share/lua/intf/cli.lua                  |    2 +-
 share/lua/intf/modules/httprequests.lua |    4 ++--
 share/lua/playlist/appletrailers.lua    |    6 +++---
 share/lua/playlist/extreme.lua          |    2 +-
 share/lua/playlist/katsomo.lua          |    2 +-
 share/lua/playlist/mpora.lua            |    2 +-
 share/lua/playlist/pinkbike.lua         |    4 ++--
 33 files changed, 22 insertions(+), 90 deletions(-)

diff --git a/modules/lua/intf.c b/modules/lua/intf.c
index 3b01727..61fe362 100644
--- a/modules/lua/intf.c
+++ b/modules/lua/intf.c
@@ -37,10 +37,6 @@
 #include <sys/types.h>
 #include <sys/stat.h>
 
-#include <lua.h>        /* Low level lua C API */
-#include <lauxlib.h>    /* Higher level C API */
-#include <lualib.h>     /* Lua libs */
-
 #include "vlc.h"
 #include "libs.h"
 
diff --git a/modules/lua/libs/acl.c b/modules/lua/libs/acl.c
index 4deb66e..693137d 100644
--- a/modules/lua/libs/acl.c
+++ b/modules/lua/libs/acl.c
@@ -35,9 +35,6 @@
 #include <vlc_common.h>
 #include <vlc_acl.h>
 
-#include <lua.h>        /* Low level lua C API */
-#include <lauxlib.h>    /* Higher level C API */
-
 #include "../vlc.h"
 #include "../libs.h"
 
diff --git a/modules/lua/libs/configuration.c b/modules/lua/libs/configuration.c
index 7a4785f..7b5f732 100644
--- a/modules/lua/libs/configuration.c
+++ b/modules/lua/libs/configuration.c
@@ -34,9 +34,6 @@
 
 #include <vlc_common.h>
 
-#include <lua.h>        /* Low level lua C API */
-#include <lauxlib.h>    /* Higher level C API */
-
 #include "../vlc.h"
 #include "../libs.h"
 
diff --git a/modules/lua/libs/dialog.c b/modules/lua/libs/dialog.c
index 3ca67b2..55468f3 100644
--- a/modules/lua/libs/dialog.c
+++ b/modules/lua/libs/dialog.c
@@ -35,9 +35,6 @@
 #include <vlc_common.h>
 #include <vlc_extensions.h>
 
-#include <lua.h>        /* Low level lua C API */
-#include <lauxlib.h>    /* Higher level C API */
-
 #include "../vlc.h"
 #include "../libs.h"
 
diff --git a/modules/lua/libs/equalizer.c b/modules/lua/libs/equalizer.c
index 75b2334..e698c6f 100644
--- a/modules/lua/libs/equalizer.c
+++ b/modules/lua/libs/equalizer.c
@@ -38,9 +38,6 @@
 #include <vlc_input.h>
 #include <vlc_charset.h>
 
-#include <lua.h>        /* Low level lua C API */
-#include <lauxlib.h>    /* Higher level C API */
-
 #include "input.h"
 #include "../libs.h"
 #include "../vlc.h"
diff --git a/modules/lua/libs/gettext.c b/modules/lua/libs/gettext.c
index b461e73..2781d35 100644
--- a/modules/lua/libs/gettext.c
+++ b/modules/lua/libs/gettext.c
@@ -32,10 +32,6 @@
 # include "config.h"
 #endif
 
-#include <lua.h>        /* Low level lua C API */
-#include <lauxlib.h>    /* Higher level C API */
-#include <lualib.h>     /* Lua libs */
-
 #include "../vlc.h"
 #include "../libs.h"
 
diff --git a/modules/lua/libs/httpd.c b/modules/lua/libs/httpd.c
index b4520d9..85a38cc 100644
--- a/modules/lua/libs/httpd.c
+++ b/modules/lua/libs/httpd.c
@@ -35,10 +35,6 @@
 #include <vlc_common.h>
 #include <vlc_httpd.h>
 
-#include <lua.h>        /* Low level lua C API */
-#include <lauxlib.h>    /* Higher level C API */
-#include <lualib.h>     /* Lua libs */
-
 #include "../vlc.h"
 #include "../libs.h"
 
diff --git a/modules/lua/libs/input.c b/modules/lua/libs/input.c
index 60211fd..7b413e2 100644
--- a/modules/lua/libs/input.c
+++ b/modules/lua/libs/input.c
@@ -37,13 +37,11 @@
 
 #include <vlc_playlist.h>
 
-#include <lua.h>        /* Low level lua C API */
-#include <lauxlib.h>    /* Higher level C API */
 #include <assert.h>
 
+#include "../vlc.h"
 #include "input.h"
 #include "playlist.h"
-#include "../vlc.h"
 #include "../libs.h"
 #include "../extension.h"
 
diff --git a/modules/lua/libs/input.h b/modules/lua/libs/input.h
index dbe76df..903134a 100644
--- a/modules/lua/libs/input.h
+++ b/modules/lua/libs/input.h
@@ -24,6 +24,8 @@
 #ifndef VLC_LUA_INPUT_H
 #define VLC_LUA_INPUT_H
 
+#include "../vlc.h"
+
 input_thread_t * vlclua_get_input_internal( lua_State * );
 
 #endif
diff --git a/modules/lua/libs/messages.c b/modules/lua/libs/messages.c
index 9c40e53..589d162 100644
--- a/modules/lua/libs/messages.c
+++ b/modules/lua/libs/messages.c
@@ -38,10 +38,6 @@
 #include <vlc_meta.h>
 #include <vlc_aout.h>
 
-#include <lua.h>        /* Low level lua C API */
-#include <lauxlib.h>    /* Higher level C API */
-#include <lualib.h>     /* Lua libs */
-
 #include "../vlc.h"
 #include "../libs.h"
 
diff --git a/modules/lua/libs/misc.c b/modules/lua/libs/misc.c
index f98d28a..5b793bd 100644
--- a/modules/lua/libs/misc.c
+++ b/modules/lua/libs/misc.c
@@ -41,10 +41,6 @@
 #include <vlc_interface.h>
 #include <vlc_keys.h>
 
-#include <lua.h>        /* Low level lua C API */
-#include <lauxlib.h>    /* Higher level C API */
-#include <lualib.h>     /* Lua libs */
-
 #include "../vlc.h"
 #include "../libs.h"
 
diff --git a/modules/lua/libs/net.c b/modules/lua/libs/net.c
index 1c7cd12..0e832cc 100644
--- a/modules/lua/libs/net.c
+++ b/modules/lua/libs/net.c
@@ -41,9 +41,6 @@
 #include <vlc_url.h>
 #include <vlc_fs.h>
 
-#include <lua.h>        /* Low level lua C API */
-#include <lauxlib.h>    /* Higher level C API */
-
 #ifdef HAVE_POLL
 #include <poll.h>       /* poll structures and defines */
 #endif
diff --git a/modules/lua/libs/objects.c b/modules/lua/libs/objects.c
index c3543d6..bcdb43e 100644
--- a/modules/lua/libs/objects.c
+++ b/modules/lua/libs/objects.c
@@ -35,9 +35,6 @@
 #include <vlc_common.h>
 #include <vlc_vout.h>
 
-#include <lua.h>        /* Low level lua C API */
-#include <lauxlib.h>    /* Higher level C API */
-
 #include "../vlc.h"
 #include "../libs.h"
 #include "objects.h"
diff --git a/modules/lua/libs/osd.c b/modules/lua/libs/osd.c
index fb36ed5..e06646b 100644
--- a/modules/lua/libs/osd.c
+++ b/modules/lua/libs/osd.c
@@ -35,9 +35,6 @@
 #include <vlc_vout.h>
 #include <vlc_osd.h>
 
-#include <lua.h>        /* Low level lua C API */
-#include <lauxlib.h>    /* Higher level C API */
-
 #include "../vlc.h"
 #include "../libs.h"
 #include "input.h"
diff --git a/modules/lua/libs/playlist.c b/modules/lua/libs/playlist.c
index ae610f3..4bd8024 100644
--- a/modules/lua/libs/playlist.c
+++ b/modules/lua/libs/playlist.c
@@ -37,9 +37,6 @@
 #include <vlc_interface.h>
 #include <vlc_playlist.h>
 
-#include <lua.h>        /* Low level lua C API */
-#include <lauxlib.h>    /* Higher level C API */
-
 #include "../vlc.h"
 #include "../libs.h"
 #include "playlist.h"
@@ -381,6 +378,7 @@ static const luaL_Reg vlclua_playlist_reg[] = {
     { "loop", vlclua_playlist_loop },
     { "random", vlclua_playlist_random },
     { "goto", vlclua_playlist_goto },
+    { "goto_", vlclua_playlist_goto },
     { "add", vlclua_playlist_add },
     { "enqueue", vlclua_playlist_enqueue },
     { "get", vlclua_playlist_get },
diff --git a/modules/lua/libs/sd.c b/modules/lua/libs/sd.c
index 7ca9476..d88cab2 100644
--- a/modules/lua/libs/sd.c
+++ b/modules/lua/libs/sd.c
@@ -38,10 +38,6 @@
 #include <vlc_playlist.h>
 #include <vlc_charset.h>
 
-#include <lua.h>        /* Low level lua C API */
-#include <lauxlib.h>    /* Higher level C API */
-#include <lualib.h>     /* Lua libs */
-
 #include "../vlc.h"
 #include "../libs.h"
 #include "playlist.h"
diff --git a/modules/lua/libs/stream.c b/modules/lua/libs/stream.c
index ef7187c..6a366f1 100644
--- a/modules/lua/libs/stream.c
+++ b/modules/lua/libs/stream.c
@@ -38,9 +38,6 @@
 #include <vlc_meta.h>
 #include <vlc_aout.h>
 
-#include <lua.h>        /* Low level lua C API */
-#include <lauxlib.h>    /* Higher level C API */
-
 #include "../vlc.h"
 #include "../libs.h"
 
diff --git a/modules/lua/libs/strings.c b/modules/lua/libs/strings.c
index ab8e435..88c5f0e 100644
--- a/modules/lua/libs/strings.c
+++ b/modules/lua/libs/strings.c
@@ -39,9 +39,6 @@
 #include <vlc_aout.h>
 #include <vlc_charset.h>
 
-#include <lua.h>        /* Low level lua C API */
-#include <lauxlib.h>    /* Higher level C API */
-
 #include "../vlc.h"
 #include "../libs.h"
 
diff --git a/modules/lua/libs/variables.c b/modules/lua/libs/variables.c
index f3becae..aad5ae1 100644
--- a/modules/lua/libs/variables.c
+++ b/modules/lua/libs/variables.c
@@ -34,10 +34,6 @@
 
 #include <vlc_common.h>
 
-#include <lua.h>        /* Low level lua C API */
-#include <lauxlib.h>    /* Higher level C API */
-#include <lualib.h>     /* Lua libs */
-
 #include "../vlc.h"
 #include "../libs.h"
 #include "variables.h"
diff --git a/modules/lua/libs/video.c b/modules/lua/libs/video.c
index eb914b5..258e0d2 100644
--- a/modules/lua/libs/video.c
+++ b/modules/lua/libs/video.c
@@ -34,9 +34,6 @@
 
 #include <vlc_vout.h>
 
-#include <lua.h>        /* Low level lua C API */
-#include <lauxlib.h>    /* Higher level C API */
-
 #include "../vlc.h"
 #include "../libs.h"
 #include "input.h"
diff --git a/modules/lua/libs/vlm.c b/modules/lua/libs/vlm.c
index d48755c..7b2b708 100644
--- a/modules/lua/libs/vlm.c
+++ b/modules/lua/libs/vlm.c
@@ -35,9 +35,6 @@
 #include <vlc_common.h>
 #include <vlc_vlm.h>
 
-#include <lua.h>        /* Low level lua C API */
-#include <lauxlib.h>    /* Higher level C API */
-
 #include "../vlc.h"
 #include "../libs.h"
 
diff --git a/modules/lua/libs/volume.c b/modules/lua/libs/volume.c
index 3aab542..bae1534 100644
--- a/modules/lua/libs/volume.c
+++ b/modules/lua/libs/volume.c
@@ -38,10 +38,6 @@
 #include <vlc_meta.h>
 #include <vlc_aout_intf.h>
 
-#include <lua.h>        /* Low level lua C API */
-#include <lauxlib.h>    /* Higher level C API */
-#include <lualib.h>     /* Lua libs */
-
 #include "../vlc.h"
 #include "../libs.h"
 #include "playlist.h"
diff --git a/modules/lua/libs/xml.c b/modules/lua/libs/xml.c
index f8b0afe..80f53fb 100644
--- a/modules/lua/libs/xml.c
+++ b/modules/lua/libs/xml.c
@@ -35,9 +35,6 @@
 #include <vlc_common.h>
 #include <vlc_xml.h>
 
-#include <lua.h>        /* Low level lua C API */
-#include <lauxlib.h>    /* Higher level C API */
-
 #include "../vlc.h"
 #include "../libs.h"
 
diff --git a/modules/lua/vlc.c b/modules/lua/vlc.c
index 5433eb8..a0132cf 100644
--- a/modules/lua/vlc.c
+++ b/modules/lua/vlc.c
@@ -45,10 +45,6 @@
 #include <vlc_stream.h>
 #include <sys/stat.h>
 
-#include <lua.h>        /* Low level lua C API */
-#include <lauxlib.h>    /* Higher level C API */
-#include <lualib.h>     /* Lua libs */
-
 #include "vlc.h"
 
 /*****************************************************************************
diff --git a/modules/lua/vlc.h b/modules/lua/vlc.h
index 5d87914..2b8fbff 100644
--- a/modules/lua/vlc.h
+++ b/modules/lua/vlc.h
@@ -36,9 +36,15 @@
 #include <vlc_strings.h>
 #include <vlc_stream.h>
 
+#define LUA_COMPAT_MODULE
 #include <lua.h>        /* Low level lua C API */
 #include <lauxlib.h>    /* Higher level C API */
 #include <lualib.h>     /* Lua libs */
+#if LUA_VERSION_NUM >= 502
+#define lua_equal(L,idx1,idx2)		lua_compare(L,(idx1),(idx2),LUA_OPEQ)
+#define lua_objlen(L,idx)			lua_rawlen(L,idx)
+#define lua_strlen(L,idx)			lua_rawlen(L,idx)
+#endif
 
 /*****************************************************************************
  * Module entry points
diff --git a/share/lua/extensions/imdb.lua b/share/lua/extensions/imdb.lua
index 829948d..194a219 100644
--- a/share/lua/extensions/imdb.lua
+++ b/share/lua/extensions/imdb.lua
@@ -237,7 +237,7 @@ function parse_resultspage(data)
                 if not link then break end -- this would not be normal behavior...
                 _, pos, thistitle = string.find(table, "<a href=\"" .. link .. "\"[^>]*>([^<]+)</a>", pos)
                 if not thistitle then break end -- this would not be normal behavior...
-                local _, _, year = string.find(table, "\((%d+)\)", pos)
+                local _, _, year = string.find(table, "%((%d+)%)", pos)
                 -- Add this title to the list
                 count = count + 1
                 local _, _, imdbID = string.find(link, "/([^/]+)/$")
diff --git a/share/lua/intf/cli.lua b/share/lua/intf/cli.lua
index 265b938..8a13194 100644
--- a/share/lua/intf/cli.lua
+++ b/share/lua/intf/cli.lua
@@ -535,7 +535,7 @@ commands_ordered = {
     { "stop"; { func = skip2(vlc.playlist.stop); help = "stop stream" } };
     { "next"; { func = skip2(vlc.playlist.next); help = "next playlist item" } };
     { "prev"; { func = skip2(vlc.playlist.prev); help = "previous playlist item" } };
-    { "goto"; { func = skip2(vlc.playlist.goto); help = "goto item at index" } };
+    { "goto"; { func = skip2(vlc.playlist.goto_); help = "goto item at index" } };
     { "repeat"; { func = skip2(vlc.playlist.repeat_); args = "[on|off]"; help = "toggle playlist repeat" } };
     { "loop"; { func = skip2(vlc.playlist.loop); args = "[on|off]"; help = "toggle playlist loop" } };
     { "random"; { func = skip2(vlc.playlist.random); args = "[on|off]"; help = "toggle playlist random" } };
diff --git a/share/lua/intf/modules/httprequests.lua b/share/lua/intf/modules/httprequests.lua
index 407ab3e..b90ea5a 100644
--- a/share/lua/intf/modules/httprequests.lua
+++ b/share/lua/intf/modules/httprequests.lua
@@ -93,14 +93,14 @@ processcommands = function ()
       if id == -1 then
         vlc.playlist.play()
       else
-        vlc.playlist.goto(id)
+        vlc.playlist.goto_(id)
       end
     elseif command == "pl_pause" then
       if vlc.playlist.status() == "stopped" then
         if id == -1 then
           vlc.playlist.play()
         else
-          vlc.playlist.goto(id)
+          vlc.playlist.goto_(id)
         end
       else
         vlc.playlist.pause()
diff --git a/share/lua/playlist/appletrailers.lua b/share/lua/playlist/appletrailers.lua
index 8bd4bea..4d0f5e7 100644
--- a/share/lua/playlist/appletrailers.lua
+++ b/share/lua/playlist/appletrailers.lua
@@ -67,7 +67,7 @@ function parse()
             description = find( line, "h%d.->(.-)</h%d") .. ' '
         end
         if string.match( line, 'img src=') then
-            for img in string.gmatch(line, '<img src="(http://.*\.jpg)" ') do
+            for img in string.gmatch(line, '<img src="(http://.*%.jpg)" ') do
                 art_url = img
             end
             for i,value in pairs(playlist) do
@@ -76,8 +76,8 @@ function parse()
                 else break end
             end
         end
-        if string.match( line, "class=\"hd\".-\.mov") then
-            for urlline,resolution in string.gmatch(line, "class=\"hd\".-href=\"(.-.mov)\".-(%d+.-p)") do
+        if string.match( line, 'class="hd".-%.mov') then
+            for urlline,resolution in string.gmatch(line, 'class="hd".-href="(.-%.mov)".-(%d+.-p)') do
                 urlline = string.gsub( urlline, "_"..resolution, "_h"..resolution )
                 table.insert( playlist, { path = urlline,
                                           name = description ..  '(' .. resolution .. ')',
diff --git a/share/lua/playlist/extreme.lua b/share/lua/playlist/extreme.lua
index 6fd7251..a71564b 100644
--- a/share/lua/playlist/extreme.lua
+++ b/share/lua/playlist/extreme.lua
@@ -41,7 +41,7 @@ end
 
 -- Parse function.
 function parse()
-    if (string.match( vlc.path, "extreme\.com/." ) or string.match( vlc.path, "freecaster\.tv/." )) and not string.match( vlc.path, "player.extreme.com/info/") then
+    if (string.match( vlc.path, "extreme.com/." ) or string.match( vlc.path, "freecaster.tv/." )) and not string.match( vlc.path, "player.extreme.com/info/") then
         while true do
             line = vlc.readline()
             if not line then break end
diff --git a/share/lua/playlist/katsomo.lua b/share/lua/playlist/katsomo.lua
index bf461c7..85c3c41 100644
--- a/share/lua/playlist/katsomo.lua
+++ b/share/lua/playlist/katsomo.lua
@@ -57,7 +57,7 @@ function parse()
         then
             arturl = "http://www.katsomo.fi"..find( line, " src=\"(.-)\" alt=" )
         end
-        for treeid,name in string.gmatch( line, "/\?treeId=(%d+)\">([^<]+)</a") do
+        for treeid,name in string.gmatch( line, '/?treeId=(%d+)">([^<]+)</a') do
             name = vlc.strings.resolve_xml_special_chars( name )
             name = vlc.strings.from_charset( "ISO_8859-1", name )
             path = "http://www.katsomo.fi/?treeId="..treeid
diff --git a/share/lua/playlist/mpora.lua b/share/lua/playlist/mpora.lua
index cfee5d0..81ffbcd 100644
--- a/share/lua/playlist/mpora.lua
+++ b/share/lua/playlist/mpora.lua
@@ -40,7 +40,7 @@ function parse()
             _,_,arturl = string.find( line, "image_src\" href=\"(.*)\" />" )
         end
         if string.match( line, "video_src" ) then
-            _,_,video = string.find( line, "href=\"http://video\.mpora\.com/ep/(.*).swf\" />" )
+            _,_,video = string.find( line, 'href="http://video.mpora.com/ep/(.*)%.swf" />' )
         end
 
     end
diff --git a/share/lua/playlist/pinkbike.lua b/share/lua/playlist/pinkbike.lua
index 06105d7..f6787c8 100644
--- a/share/lua/playlist/pinkbike.lua
+++ b/share/lua/playlist/pinkbike.lua
@@ -45,10 +45,10 @@ function parse()
 			end
 			-- Try to find server which has our video
 			if string.match( line, "<link rel=\"videothumbnail\" href=\"http://(.*)/vt/svt-") then
-				_,_,server = string.find (line, "<link rel=\"videothumbnail\"\ href=\"http://(.*)/vt/svt-" )
+				_,_,server = string.find (line, '<link rel="videothumbnail" href="http://(.*)/vt/svt-' )
 			end
 			if string.match( line, "<link rel=\"videothumbnail\" href=\"(.*)\" type=\"image/jpeg\"") then
-				_,_,arturl = string.find (line, "<link rel=\"videothumbnail\" href=\"(.*)\"\ type=\"image/jpeg\"")
+				_,_,arturl = string.find (line, '<link rel="videothumbnail" href="(.*)" type="image/jpeg"')
 			end
 		end
 
-- 
1.7.7