Merge f8703c429117a92ad63d2c94a99eee6d6f1165ae into 673277e510ebd996b62a2fcc76169bf3cce29910

* [YouTube] Extract playlist items from LOCKUP_VIEW_MODEL_...
* resolves #33073
* thx seproDev (yt-dlp/yt-dlp#11615)

Co-authored-by: sepro <sepro@sepr0.com>

youtube_dl/extractor/extractors.py

@@ -738,7 +738,7 @@ from .mtv import (
 from .muenchentv import MuenchenTVIE
 from .mwave import MwaveIE, MwaveMeetGreetIE
 from .mychannels import MyChannelsIE
-from .myspace import MySpaceIE, MySpaceAlbumIE
+from .myspace import MySpaceIE, MySpaceAlbumIE, MySpaceArtistIE
 from .myspass import MySpassIE
 from .myvi import (
     MyviIE,

youtube_dl/extractor/myspace.py

@@ -203,10 +203,35 @@ class MySpaceAlbumIE(InfoExtractor):
         entries = [
             self.url_result(t_path, ie=MySpaceIE.ie_key())
             for t_path in tracks_paths]
-        return {
+        return self.playlist_result(entries, playlist_id, self._og_search_title(webpage))
-            '_type': 'playlist',
+
-            'id': playlist_id,
+
-            'display_id': display_id,
+class MySpaceArtistIE(InfoExtractor):
-            'title': self._og_search_title(webpage),
+    IE_NAME = 'MySpace:artist'
-            'entries': entries,
+    _VALID_URL = r'https?://myspace\.com/(?P<artist>[^/]*)/music/songs'
-        }
+
+    _TEST = {
+        'url': 'https://myspace.com/studio99/music/songs/',
+        'info_dict': {
+            'title': 'Studio 99 IS CLOSED!! R.I.P.',
+        },
+        'playlist_count': 4,
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        display_id = mobj.group('artist')
+        webpage = self._download_webpage(url, display_id)
+        tracks_paths = re.findall(r'<meta itemprop="url" content ="(.*?)">', webpage)
+        if not tracks_paths:
+            raise ExtractorError(
+                '%s: No songs found, try using proxy' % display_id,
+                expected=True)
+
+        entries = [
+            self.url_result('https://myspace.com/' + display_id + t_path, ie=MySpaceIE.ie_key())
+            for t_path in tracks_paths if t_path.startswith('/music/song/')]
+        #   if we invert this if, we get album urls. but not all music are in alba.
+        #   Also the musics in alba are already here individually
+
+        return self.playlist_result(entries, playlist_title=self._og_search_title(webpage))

youtube_dl/extractor/youtube.py

@@ -27,6 +27,7 @@ from ..compat import (
 )
 from ..jsinterp import JSInterpreter
 from ..utils import (
+    bug_reports_message,
     clean_html,
     dict_get,
     error_to_compat_str,
@@ -65,6 +66,7 @@ from ..utils import (
     url_or_none,
     urlencode_postdata,
     urljoin,
+    variadic,
 )
 
 
@@ -460,6 +462,26 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
             'uploader': uploader,
         }
 
+    @staticmethod
+    def _extract_thumbnails(data, *path_list, **kw_final_key):
+        """
+        Extract thumbnails from thumbnails dict
+        @param path_list: path list to level that contains 'thumbnails' key
+        """
+        final_key = kw_final_key.get('final_key', 'thumbnails')
+
+        return traverse_obj(data, ((
+            tuple(variadic(path) + (final_key, Ellipsis)
+                  for path in path_list or [()])), {
+            'url': ('url', T(url_or_none),
+                    # Sometimes youtube gives a wrong thumbnail URL. See:
+                    # https://github.com/yt-dlp/yt-dlp/issues/233
+                    # https://github.com/ytdl-org/youtube-dl/issues/28023
+                    T(lambda u: update_url(u, query=None) if u and 'maxresdefault' in u else u)),
+            'height': ('height', T(int_or_none)),
+            'width': ('width', T(int_or_none)),
+        }, T(lambda t: t if t.get('url') else None)))
+
     def _search_results(self, query, params):
         data = {
             'context': {
@@ -3183,8 +3205,12 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
             expected_type=txt_or_none)
 
     def _grid_entries(self, grid_renderer):
-        for item in grid_renderer['items']:
+        for item in traverse_obj(grid_renderer, ('items', Ellipsis, T(dict))):
-            if not isinstance(item, dict):
+            lockup_view_model = traverse_obj(item, ('lockupViewModel', T(dict)))
+            if lockup_view_model:
+                entry = self._extract_lockup_view_model(lockup_view_model)
+                if entry:
+                    yield entry
                 continue
             renderer = self._extract_grid_item_renderer(item)
             if not isinstance(renderer, dict):
@@ -3268,6 +3294,25 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
                 continue
             yield self._extract_video(renderer)
 
+    def _extract_lockup_view_model(self, view_model):
+        content_id = view_model.get('contentId')
+        if not content_id:
+            return
+        content_type = view_model.get('contentType')
+        if content_type not in ('LOCKUP_CONTENT_TYPE_PLAYLIST', 'LOCKUP_CONTENT_TYPE_PODCAST'):
+            self.report_warning(
+                'Unsupported lockup view model content type "{0}"{1}'.format(content_type, bug_reports_message()), only_once=True)
+            return
+        return merge_dicts(self.url_result(
+            update_url_query('https://www.youtube.com/playlist', {'list': content_id}),
+            ie=YoutubeTabIE.ie_key(), video_id=content_id), {
+                'title': traverse_obj(view_model, (
+                    'metadata', 'lockupMetadataViewModel', 'title', 'content', T(compat_str))),
+                'thumbnails': self._extract_thumbnails(view_model, (
+                    'contentImage', 'collectionThumbnailViewModel', 'primaryThumbnail',
+                    'thumbnailViewModel', 'image'), final_key='sources'),
+        })
+
     def _video_entry(self, video_renderer):
         video_id = video_renderer.get('videoId')
         if video_id: