[vlc-devel] [PATCH 3/4] ytdl: extract playlist non-recursively ("flat")

[Rémi Denis-Courmont]

This uses the extract-flat mode of YoutubeDL, which skips parsing
individual items within a playlist. We define a dedicated MRL scheme
to track a YoutubeDL playlist item, and parse it only when the item is
actually opened.

This has two benefits:
1) Extracting a playlist is dramatically faster.
2) Expiring media URL can be played even if the playlist is long.

This does *not* solve the remaining problem that expiring URLs cannot
be saved and replayed later.
---
 modules/stream_filter/ytdl.c | 30 +++++++++++++++++++++---------
 share/ytdl-extract.py        | 31 ++++++++++++++++++++++++++++++-
 2 files changed, 51 insertions(+), 10 deletions(-)

diff --git a/modules/stream_filter/ytdl.c b/modules/stream_filter/ytdl.c
index 6da83e0c08..10af6c23f7 100644
--- a/modules/stream_filter/ytdl.c
+++ b/modules/stream_filter/ytdl.c
@@ -132,19 +132,11 @@ static void Close(vlc_object_t *obj)
     vlc_waitpid(sys->pid);
 }
 
-static int OpenFilter(vlc_object_t *obj)
+static int OpenCommon(vlc_object_t *obj)
 {
     stream_t *s = (stream_t *)obj;
     ssize_t val;
 
-    if (s->s->pf_readdir != NULL || s->psz_url == NULL)
-        return VLC_EGENERIC;
-   if (strncasecmp(s->psz_url, "http:", 5)
-    && strncasecmp(s->psz_url, "https:", 6))
-       return VLC_EGENERIC;
-    if (!var_InheritBool(s, "ytdl"))
-        return VLC_EGENERIC;
-
     struct ytdl_playlist *sys = vlc_obj_malloc(obj, sizeof (*sys));
     if (unlikely(sys == NULL))
         return VLC_EGENERIC;
@@ -192,6 +184,21 @@ static int OpenFilter(vlc_object_t *obj)
     return VLC_SUCCESS;
 }
 
+static int OpenFilter(vlc_object_t *obj)
+{
+    stream_t *s = (stream_t *)obj;
+
+    if (s->s->pf_readdir != NULL || s->psz_url == NULL)
+        return VLC_EGENERIC;
+    if (strncasecmp(s->psz_url, "http:", 5)
+     && strncasecmp(s->psz_url, "https:", 6))
+        return VLC_EGENERIC;
+    if (!var_InheritBool(obj, "ytdl"))
+        return VLC_EGENERIC;
+
+    return OpenCommon(obj);
+}
+
 vlc_module_begin()
     set_shortname("YT-DL")
     set_description("YoutubeDL")
@@ -201,4 +208,9 @@ vlc_module_begin()
     set_callbacks(OpenFilter, Close)
     add_bool("ytdl", true, N_("Enable YT-DL"), N_("Enable YT-DL"), true)
         change_safe()
+
+    add_submodule()
+    set_capability("access", 0)
+    add_shortcut("ytdl")
+    set_callbacks(OpenCommon, Close)
 vlc_module_end()
diff --git a/share/ytdl-extract.py b/share/ytdl-extract.py
index 74b8385d55..f166384b60 100755
--- a/share/ytdl-extract.py
+++ b/share/ytdl-extract.py
@@ -17,6 +17,7 @@
 # Inc., 51 Franklin Street, Fifth Floor, Boston MA 02110-1301, USA.
 
 import sys
+import urllib.parse
 import youtube_dl
 
 class logger(object):
@@ -68,6 +69,11 @@ def entry_extract(entry):
     if 'title' in entry:
         print('#EXTINF:,,' + entry['title'].splitlines()[0])
 
+    if 'ie_key' in fmt and fmt['ie_key']:
+        # Flat-extracted playlist entry
+        print('ytdl:///?' + urllib.parse.urlencode(fmt))
+        return
+
     if 'url' in fmt:
         print('#EXTVLCOPT:no-ytdl') # don't parse recursively
         print(fmt['url'])
@@ -76,6 +82,7 @@ def entry_extract(entry):
 
 def url_extract(url):
     opts = {
+        'extract_flat': True,
         'logger': logger(),
         'youtube_include_dash_manifest': False,
     }
@@ -98,5 +105,27 @@ def url_extract(url):
         # URL is a single media
         entry_extract(infos)
 
+def url_process(ie_url):
+    opts = {
+        'logger': logger(),
+        'youtube_include_dash_manifest': False,
+    }
+
+    dl = youtube_dl.YoutubeDL(opts)
+
+    # Rebuild the original IE entry
+    entry = { }
+
+    for p in urllib.parse.parse_qsl(url[9:]):
+        entry[p[0]] = p[1]
+
+    entry = dl.process_ie_result(entry, download=False)
+    print('#EXTM3U')
+    entry_extract(entry)
+
+url = sys.argv[1]
 
-url_extract(sys.argv[1])
+if url.startswith('ytdl:///?'):
+    url_process(url)
+else:
+    url_extract(url)
-- 
2.28.0