Merge 3033db0217e58b4c264683a49dc71660efea317a 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/beeg.py

@@ -1,18 +1,28 @@
+# coding: utf-8
+
 from __future__ import unicode_literals
 
+import itertools
+
 from .common import InfoExtractor
 from ..compat import (
     compat_str,
     compat_urlparse,
+    compat_zip as zip,
 )
 from ..utils import (
+    dict_get,
     int_or_none,
+    parse_iso8601,
+    str_or_none,
+    try_get,
     unified_timestamp,
+    urljoin,
 )
 
 
 class BeegIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?beeg\.(?:com|porn(?:/video)?)/(?P<id>\d+)'
+    _VALID_URL = r'https?://(?:www\.)?beeg\.(?:com|porn(?:/video)?)/-?(?P<id>\d+)'
     _TESTS = [{
         # api/v6 v1
         'url': 'http://beeg.com/5416503',
@@ -27,7 +37,8 @@ class BeegIE(InfoExtractor):
             'duration': 383,
             'tags': list,
             'age_limit': 18,
-        }
+        },
+        'skip': 'no longer available',
     }, {
         # api/v6 v2
         'url': 'https://beeg.com/1941093077?t=911-1391',
@@ -42,13 +53,28 @@ class BeegIE(InfoExtractor):
     }, {
         'url': 'https://beeg.porn/5416503',
         'only_matching': True,
+    }, {
+        'url': 'https://beeg.com/-01882333214',
+        'md5': 'f8fd97a58a09bc6d2ac25b6bbc98c220',
+        'info_dict': {
+            'id': '1882333214',
+            'ext': 'mp4',
+            'title': 'Be mine',
+            'description': 'Lil cam dork living in Canada with 4 kitties',
+            'timestamp': 1628218803,
+            'upload_date': '20210806',
+            'duration': 559,
+            'tags': list,
+            'age_limit': 18,
+            'thumbnail': r're:https?://thumbs\.beeg\.com/videos/1882333214/\d+\.jpe?g',
+        },
+        'params': {
+            # ignore variable HLS file data
+            'skip_download': True,
+        },
     }]
 
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-
-        webpage = self._download_webpage(url, video_id)
-
+    def _old_real_extract(self, url, video_id, webpage):
         beeg_version = self._search_regex(
             r'beeg_version\s*=\s*([\da-zA-Z_-]+)', webpage, 'beeg version',
             default='1546225636701')
@@ -67,11 +93,14 @@ class BeegIE(InfoExtractor):
         else:
             query = {'v': 1}
 
-        for api_path in ('', 'api.'):
+        def reverse_enumerate(l):
+            return zip(reversed(next(zip(*enumerate(l)), [])), l)
+
+        for rcnt, api_path in reverse_enumerate(('', 'api.')):
             video = self._download_json(
                 'https://%sbeeg.com/api/v6/%s/video/%s'
                 % (api_path, beeg_version, video_id), video_id,
-                fatal=api_path == 'api.', query=query)
+                fatal=(rcnt == 0), query=query)
             if video:
                 break
 
@@ -114,3 +143,53 @@ class BeegIE(InfoExtractor):
             'formats': formats,
             'age_limit': self._rta_search(webpage),
         }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, video_id)
+
+        formats = []
+        # leading 0s stripped from ID by site, though full ID also works
+        video = self._download_json(
+            'https://store.externulls.com/facts/file/%d' % int(video_id),
+            video_id, fatal=False)
+        if video:
+            video_data = try_get(video, lambda x: x['file'], dict) or {}
+            video_data.update(video_data.get('stuff', {}))
+            video_data.update(try_get(video, lambda x: x['fc_facts'][0], dict))
+            for format_id, key in itertools.chain(
+                    try_get(video_data, lambda x: x['resources'].items()) or [],
+                    try_get(video_data, lambda x: x['hls_resources'].items()) or []):
+                if not format_id:
+                    continue
+                formats.append({
+                    'url': urljoin('https://video.beeg.com', key),
+                    'format_id': format_id,
+                    'height': int_or_none(format_id.rsplit('_', 1)[-1]),
+                })
+            if formats:
+                self._sort_formats(formats)
+                tags = try_get(video, lambda v: [tag['tg_slug'] for tag in v['tags'] if tag.get('tg_slug')])
+                result = {
+                    'id': str_or_none(video.get('fc_file_id')) or video_id,
+                    'title': video_data.get('sf_name') or 'Beeg video %s' % video_id,
+                    'description': video_data.get('sf_story'),
+                    'timestamp': parse_iso8601(video_data.get('fc_created')),
+                    'duration': int_or_none(dict_get(video_data, ('fl_duration', 'sf_duration'))),
+                    'tags': tags,
+                    'formats': formats,
+                    'age_limit': self._rta_search(webpage) or 18,
+                    'height': int_or_none(video_data.get('fl_height')),
+                    'width': int_or_none(video_data.get('fl_width')),
+                }
+                thumb_id = try_get(video_data, lambda x: x['fc_thumbs'][0], int)
+                if thumb_id is not None:
+                    set_id = video_data.get('resources', {}).get('set_id')
+                    result['thumbnail'] = (
+                        'https://thumbs.beeg.com/videos/%s/%d.jpg' % (result['id'], thumb_id)
+                        if set_id is None else
+                        'https://thumbs-015.externulls.com/sets/%s/thumbs/%s-%0.4d.jpg' % (set_id, set_id, thumb_id))
+                return result
+
+        return self._old_real_extract(url, video_id, 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']:
-            if not isinstance(item, dict):
+        for item in traverse_obj(grid_renderer, ('items', Ellipsis, T(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: