X-Git-Url: https://jfr.im/git/yt-dlp.git/blobdiff_plain/2eb0f72a0ea6bdbf6889efe34027a534a3473a03..3cfd000849208b58dab4f78d1486d3f24552009e:/youtube_dl/extractor/youtube.py

diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py
index 08e821362..21731188a 100644
--- a/youtube_dl/extractor/youtube.py
+++ b/youtube_dl/extractor/youtube.py
@@ -703,7 +703,7 @@ def _signature_cache_id(self, example_sig):
 
     def _extract_signature_function(self, video_id, player_url, example_sig):
         id_m = re.match(
-            r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?)?\.(?P<ext>[a-z]+)$',
+            r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|/base)?\.(?P<ext>[a-z]+)$',
             player_url)
         if not id_m:
             raise ExtractorError('Cannot identify player %r' % player_url)
@@ -1074,7 +1074,7 @@ def add_dash_mpd(video_info):
             age_gate = False
             video_info = None
             # Try looking directly into the video webpage
-            mobj = re.search(r';ytplayer\.config\s*=\s*({.*?});', video_webpage)
+            mobj = re.search(r';ytplayer\.config\s*=\s*({.*?});ytplayer', video_webpage)
             if mobj:
                 json_code = uppercase_escape(mobj.group(1))
                 ytplayer_config = json.loads(json_code)
@@ -1107,6 +1107,17 @@ def add_dash_mpd(video_info):
                     if not video_info:
                         video_info = get_video_info
                     if 'token' in get_video_info:
+                        # Different get_video_info requests may report different results, e.g.
+                        # some may report video unavailability, but some may serve it without
+                        # any complaint (see https://github.com/rg3/youtube-dl/issues/7362,
+                        # the original webpage as well as el=info and el=embedded get_video_info
+                        # requests report video unavailability due to geo restriction while
+                        # el=detailpage succeeds and returns valid data). This is probably
+                        # due to YouTube measures against IP ranges of hosting providers.
+                        # Working around by preferring the first succeeded video_info containing
+                        # the token if no such video_info yet was found.
+                        if 'token' not in video_info:
+                            video_info = get_video_info
                         break
         if 'token' not in video_info:
             if 'reason' in video_info:
@@ -1332,7 +1343,7 @@ def _map_to_format_list(urlmap):
                                 player_desc = 'flash player %s' % player_version
                             else:
                                 player_version = self._search_regex(
-                                    r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js',
+                                    [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js', r'(?:www|player)-([^/]+)/base\.js'],
                                     player_url,
                                     'html5 player', fatal=False)
                                 player_desc = 'html5 player %s' % player_version
@@ -1644,8 +1655,18 @@ class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
         'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
         'playlist_mincount': 91,
         'info_dict': {
-            'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
+            'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
+            'title': 'Uploads from lex will',
         }
+    }, {
+        'note': 'Age restricted channel',
+        # from https://www.youtube.com/user/DeusExOfficial
+        'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
+        'playlist_mincount': 64,
+        'info_dict': {
+            'id': 'UUs0ifCMCm1icqRbqhUINa0w',
+            'title': 'Uploads from Deus Ex',
+        },
     }]
 
     def _real_extract(self, url):
@@ -1666,7 +1687,7 @@ def _real_extract(self, url):
                 'channelId', channel_page, 'channel id', default=None)
             if not channel_playlist_id:
                 channel_playlist_id = self._search_regex(
-                    r'data-channel-external-id="([^"]+)"',
+                    r'data-(?:channel-external-|yt)id="([^"]+)"',
                     channel_page, 'channel id', default=None)
         if channel_playlist_id and channel_playlist_id.startswith('UC'):
             playlist_id = 'UU' + channel_playlist_id[2:]