Merge 47217e7b547c08a5e20c94aa9e31828183bdb5ca 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/twentythreevideo.py

@@ -4,11 +4,12 @@ import re
 
 from .common import InfoExtractor
 from ..utils import int_or_none
+from ..compat import compat_urllib_parse_urlencode
 
 
 class TwentyThreeVideoIE(InfoExtractor):
     IE_NAME = '23video'
-    _VALID_URL = r'https?://(?P<domain>[^.]+\.(?:twentythree\.net|23video\.com|filmweb\.no))/v\.ihtml/player\.html\?(?P<query>.*?\bphoto(?:_|%5f)id=(?P<id>\d+).*)'
+    _VALID_URL = r'https?://(?P<domain>[^.]+\.(?:twentythree\.net|kglteater\.dk|23video\.com|filmweb\.no))/v\.ihtml/player\.html\?(?P<query>.*?\bphoto(?:_|%5f)id=(?P<id>\d+).*)'
     _TESTS = [{
         'url': 'https://video.twentythree.net/v.ihtml/player.html?showDescriptions=0&source=site&photo%5fid=20448876&autoPlay=1',
         'md5': '75fcf216303eb1dae9920d651f85ced4',
@@ -23,18 +24,93 @@ class TwentyThreeVideoIE(InfoExtractor):
         }
     }, {
         'url': 'https://bonnier-publications-danmark.23video.com/v.ihtml/player.html?token=f0dc46476e06e13afd5a1f84a29e31e8&source=embed&photo%5fid=36137620',
-        'only_matching': True,
+        'md5': '772a91f83d129ee5f015b12bea61a78b',
+        'info_dict': {
+            'id': '36137620',
+            'ext': 'mp4',
+            'title': 'Photoshop Elements 2019 - Photo Text',
+            'timestamp': 1538664032,
+            'upload_date': '20181004',
+            'uploader_id': '10801356',
+            'uploader': 'Kristoffer Engbo',
+        }
+    }, {
+        'url': 'https://video.kglteater.dk/v.ihtml/player.html?source=share&photo%5fid=65098499',
+        'md5': '4e20a33ce86b13ca114ee44a0a8d8efb',
+        'info_dict': {
+            'id': '65098499',
+            'ext': 'mp4',
+            'title': 'Askepot',
+            'timestamp': 1605173942,
+            'upload_date': '20201112',
+            'uploader_id': '62151179',
+            'uploader': 'jbny',
+        }
+    }, {
+        'url': 'https://video.kglteater.dk/v.ihtml/player.html?showDescriptions=0&source=site&photo%5fid=52486482&autoPlay=1',
+        'md5': 'c39ffb965079fb4395788e6814ec3cdc',
+        'info_dict': {
+            'id': '52486482',
+            'ext': 'mp4',
+            'title': 'N\xf8ddekn\xe6kkeren 2019',
+            'timestamp': 1558953133,
+            'upload_date': '20190527',
+            'uploader_id': '7450690',
+            'uploader': 'Tejs Holm',
+        }
     }]
 
     def _real_extract(self, url):
         domain, query, photo_id = re.match(self._VALID_URL, url).groups()
         base_url = 'https://%s' % domain
-        photo_data = self._download_json(
-            base_url + '/api/photo/list?' + query, photo_id, query={
-                'format': 'json',
-            }, transform_source=lambda s: self._search_regex(r'(?s)({.+})', s, 'photo data'))['photo']
-        title = photo_data['title']
 
+        def is_geo_blocked():
+            # /api/player/settings
+            playersettings_0 = {'player_id': 0, 'parameters': 'showDescriptions=0&source=site&photo%5fid=' + photo_id + '&autoPlay=1', '_li': 0, '_bot': 0}
+            playersettings_0_param = '/api/player/settings?' + compat_urllib_parse_urlencode(playersettings_0)
+
+            # /api/live/list
+            livelist_1 = {'include_actions_p': 1, 'showDescriptions': 0, 'source': 'site', 'photo_id': photo_id, 'autoPlay': 1, 'upcoming_p': 1, 'ordering': 'streaming', 'player_id': 0}
+            livelist_1_param = '/api/live/list?' + compat_urllib_parse_urlencode(livelist_1)
+
+            # /api/photo/list
+            photolist_2 = {'size': 10, 'include_actions_p': 1, 'showDescriptions': 0, 'source': 'site', 'photo_id': photo_id, 'autoPlay': 1, 'player_id': 0}
+            photolist_2_param = '/api/photo/list?' + compat_urllib_parse_urlencode(photolist_2)
+
+            new_query = {'format': 'json', 'callback': 'test', 'playersettings_0': playersettings_0_param, 'livelist_1': livelist_1_param, 'photolist_2': photolist_2_param}
+            photolist_result = self._download_json(
+                base_url + '/api/concatenate',
+                photo_id,
+                query=new_query,
+                transform_source=lambda s: self._search_regex(r'(?s)({.+})', s, 'photolist_2')
+            )["photolist_2"]
+
+            if "photos" in photolist_result:
+                for photo in photolist_result['photos']:
+                    if photo['photo_id'] == photo_id:
+                        return photo['protection_method'] == 'geoblocking'
+
+            return False
+
+        def get_video_query():
+            if is_geo_blocked():
+                token = self._download_json(
+                    base_url + '/api/protection/verify',
+                    photo_id,
+                    query={'protection_method': 'geoblocking', 'object_id': photo_id, 'object_type': 'photo', 'format': 'json', 'callback': 'visualplatform_1'},
+                    transform_source=lambda s: self._search_regex(r'(?s)({.+})', s, 'protectedtoken'))['protectedtoken']['protected_token']
+
+                return {'format': 'json', 'token': token}
+            else:
+                return {'format': 'json'}
+
+        video_query = get_video_query()
+        photo_data = self._download_json(
+            base_url + '/api/photo/list?' + query, photo_id,
+            query=video_query,
+            transform_source=lambda s: self._search_regex(r'(?s)({.+})', s, 'photo data'))['photo']
+
+        title = photo_data['title']
         formats = []
 
         audio_path = photo_data.get('audio_download')

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: