Merge branch 'polskie-radio-programme' of https://github.com/JakubAdamWieczorek/youtu...
[youtube-dl] / youtube_dl / extractor / youtube.py
index d5d5b7334e4b26a6bb9156857f05df68d0088985..5082cb589c430ed35fda45b97a9c7322aa66d594 100644 (file)
@@ -264,7 +264,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                          )
                      )?                                                       # all until now is optional -> you can pass the naked ID
                      ([0-9A-Za-z_-]{11})                                      # here is it! the YouTube video ID
-                     (?!.*?&list=)                                            # combined list/video URLs are handled by the playlist IE
+                     (?!.*?\blist=)                                            # combined list/video URLs are handled by the playlist IE
                      (?(1).+)?                                                # if we found the ID, everything can follow
                      $"""
     _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
@@ -844,6 +844,24 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
             # YouTube Red paid video (https://github.com/rg3/youtube-dl/issues/10059)
             'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
             'only_matching': True,
+        },
+        {
+            # Rental video preview
+            'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
+            'info_dict': {
+                'id': 'uGpuVWrhIzE',
+                'ext': 'mp4',
+                'title': 'Piku - Trailer',
+                'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
+                'upload_date': '20150811',
+                'uploader': 'FlixMatrix',
+                'uploader_id': 'FlixMatrixKaravan',
+                'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/FlixMatrixKaravan',
+                'license': 'Standard YouTube License',
+            },
+            'params': {
+                'skip_download': True,
+            },
         }
     ]
 
@@ -1254,6 +1272,12 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                     # Convert to the same format returned by compat_parse_qs
                     video_info = dict((k, [v]) for k, v in args.items())
                     add_dash_mpd(video_info)
+                # Rental video is not rented but preview is available (e.g.
+                # https://www.youtube.com/watch?v=yYr8q0y5Jfg,
+                # https://github.com/rg3/youtube-dl/issues/10532)
+                if not video_info and args.get('ypc_vid'):
+                    return self.url_result(
+                        args['ypc_vid'], YoutubeIE.ie_key(), video_id=args['ypc_vid'])
                 if args.get('livestream') == '1' or args.get('live_playback') == 1:
                     is_live = True
             if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
@@ -1754,11 +1778,14 @@ class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
     _VALID_URL = r"""(?x)(?:
                         (?:https?://)?
                         (?:\w+\.)?
-                        youtube\.com/
                         (?:
-                           (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)
-                           \? (?:.*?[&;])*? (?:p|a|list)=
-                        |  p/
+                            youtube\.com/
+                            (?:
+                               (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)
+                               \? (?:.*?[&;])*? (?:p|a|list)=
+                            |  p/
+                            )|
+                            youtu\.be/[0-9A-Za-z_-]{11}\?.*?\blist=
                         )
                         (
                             (?:PL|LL|EC|UU|FL|RD|UL)?[0-9A-Za-z-_]{10,}
@@ -1841,6 +1868,31 @@ class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
             'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
         },
         'playlist_mincout': 21,
+    }, {
+        # Playlist URL that does not actually serve a playlist
+        'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
+        'info_dict': {
+            'id': 'FqZTN594JQw',
+            'ext': 'webm',
+            'title': "Smiley's People 01 detective, Adventure Series, Action",
+            'uploader': 'STREEM',
+            'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng',
+            'uploader_url': 're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng',
+            'upload_date': '20150526',
+            'license': 'Standard YouTube License',
+            'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
+            'categories': ['People & Blogs'],
+            'tags': list,
+            'like_count': int,
+            'dislike_count': int,
+        },
+        'params': {
+            'skip_download': True,
+        },
+        'add_ie': [YoutubeIE.ie_key()],
+    }, {
+        'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
+        'only_matching': True,
     }]
 
     def _real_initialize(self):
@@ -1901,9 +1953,20 @@ class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
 
         playlist_title = self._html_search_regex(
             r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
-            page, 'title')
+            page, 'title', default=None)
+
+        has_videos = True
 
-        return self.playlist_result(self._entries(page, playlist_id), playlist_id, playlist_title)
+        if not playlist_title:
+            try:
+                # Some playlist URLs don't actually serve a playlist (e.g.
+                # https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4)
+                next(self._entries(page, playlist_id))
+            except StopIteration:
+                has_videos = False
+
+        return has_videos, self.playlist_result(
+            self._entries(page, playlist_id), playlist_id, playlist_title)
 
     def _check_download_just_video(self, url, playlist_id):
         # Check if it's a video-specific URL
@@ -1912,9 +1975,11 @@ class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
             video_id = query_dict['v'][0]
             if self._downloader.params.get('noplaylist'):
                 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
-                return self.url_result(video_id, 'Youtube', video_id=video_id)
+                return video_id, self.url_result(video_id, 'Youtube', video_id=video_id)
             else:
                 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
+                return video_id, None
+        return None, None
 
     def _real_extract(self, url):
         # Extract playlist id
@@ -1923,7 +1988,7 @@ class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
             raise ExtractorError('Invalid URL: %s' % url)
         playlist_id = mobj.group(1) or mobj.group(2)
 
-        video = self._check_download_just_video(url, playlist_id)
+        video_id, video = self._check_download_just_video(url, playlist_id)
         if video:
             return video
 
@@ -1931,7 +1996,15 @@ class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
             # Mixes require a custom extraction process
             return self._extract_mix(playlist_id)
 
-        return self._extract_playlist(playlist_id)
+        has_videos, playlist = self._extract_playlist(playlist_id)
+        if has_videos or not video_id:
+            return playlist
+
+        # Some playlist URLs don't actually serve a playlist (see
+        # https://github.com/rg3/youtube-dl/issues/10537).
+        # Fallback to plain video extraction if there is a video id
+        # along with playlist id.
+        return self.url_result(video_id, 'Youtube', video_id=video_id)
 
 
 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
@@ -2309,10 +2382,11 @@ class YoutubeWatchLaterIE(YoutubePlaylistIE):
     }]
 
     def _real_extract(self, url):
-        video = self._check_download_just_video(url, 'WL')
+        _, video = self._check_download_just_video(url, 'WL')
         if video:
             return video
-        return self._extract_playlist('WL')
+        _, playlist = self._extract_playlist('WL')
+        return playlist
 
 
 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
@@ -2343,7 +2417,7 @@ class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
 
 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
     IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
-    _VALID_URL = 'https?://www\.youtube\.com/feed/history|:ythistory'
+    _VALID_URL = r'https?://www\.youtube\.com/feed/history|:ythistory'
     _FEED_NAME = 'history'
     _PLAYLIST_TITLE = 'Youtube History'