[vlc-devel] Resubmitting [PATCH] npapi plugin slave-input support Add the ability to specify a sound source through slave-input, for example a v4l video capture device with alsa audio. Now with added buffer overflow protection.
Tord Andersson
tord.andersson at endian.se
Wed Mar 21 11:56:14 CET 2012
On 2012-03-20 15.41, Sergey Radionov wrote:
> 2012/3/20 Tord Andersson <tord.andersson at endian.se>:
>> ---
>> npapi/vlcplugin_base.cpp | 6 ++++++
>> 1 files changed, 6 insertions(+), 0 deletions(-)
>>
>> diff --git a/npapi/vlcplugin_base.cpp b/npapi/vlcplugin_base.cpp
>> index 1d9db49..3beee35 100644
>> --- a/npapi/vlcplugin_base.cpp
>> +++ b/npapi/vlcplugin_base.cpp
>> @@ -464,6 +464,12 @@ NPError VlcPluginBase::init(int argc, char* const argn[], char* const argv[])
>> {
>> set_enable_fs( boolValue(argv[i]) );
>> }
>> + else if( !strcmp( argn[i], "input-slave" ) )
>> + {
>> + char psz_input_slave_opts[80] = "--input-slave=";
> why exactly 80? why no use std::vector for example?
I tried to stay in line with the rest of the code.
80 characters was just an idea of how much would be needed.
I guess, some kind of limit is probably needed.
As also pointed out, the variable should be moved to the scope above.
>
>> + strncat(psz_input_slave_opts, argv[i], sizeof(psz_input_slave_opts) - strlen(psz_input_slave_opts) - 1);
> or that option can't be more than "sizeof(psz_input_slave_opts) -
> strlen(psz_input_slave_opts) - 1" ?
>
Yes. It's typically a string such as "--input-slave=alsa://hw:2,0"
>> + ppsz_argv[ppsz_argc++] = psz_input_slave_opts;
>> + }
>> else if( !strcmp( argn[i], "mute" ) )
>> {
>> if( boolValue(argv[i]) )
>> --
>> 1.7.5.4
>>
>> _______________________________________________
>> 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
More information about the vlc-devel
mailing list