Merge 4ba906565442e36afaa9ccb3c09bcb6b4f4bebcb 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/alphaporno.py

@@ -20,10 +20,9 @@ class AlphaPornoIE(InfoExtractor):
             'ext': 'mp4',
             'title': 'Sensual striptease porn with Samantha Alexandra',
             'thumbnail': r're:https?://.*\.jpg$',
-            'timestamp': 1418694611,
+            'timestamp': 1418701811,
             'upload_date': '20141216',
             'duration': 387,
-            'filesize_approx': 54120000,
             'tbr': 1145,
             'categories': list,
             'age_limit': 18,
@@ -36,42 +35,53 @@ class AlphaPornoIE(InfoExtractor):
         webpage = self._download_webpage(url, display_id)
 
         video_id = self._search_regex(
-            r"video_id\s*:\s*'([^']+)'", webpage, 'video id', default=None)
+            [r"video_id\s*:\s*'([^']+)'", r"\['video_id'\]\s*=\s*(.+?);"], webpage, 'video id')
 
-        video_url = self._search_regex(
-            r"video_url\s*:\s*'([^']+)'", webpage, 'video url')
-        ext = self._html_search_meta(
-            'encodingFormat', webpage, 'ext', default='.mp4')[1:]
+        entries = self._parse_html5_media_entries('https://alphaporno.com', webpage, video_id)
+        info = {}
+        if len(entries) > 0:
+            info = entries[0]
+            for fmt in info['formats']:
+                fmt['tbr'] = int_or_none(self._search_regex(r"br=(\d+)", fmt['url'], 'bitrate', default=None))
+        else:
+            video_url = self._search_regex(
+                r"video_url\s*:\s*'([^']+)'", webpage, 'video url')
+            ext = self._html_search_meta(
+                'encodingFormat', webpage, 'ext', default='.mp4')[1:]
+            bitrate = int_or_none(self._html_search_meta(
+                'bitrate', webpage, 'bitrate', default=None))
+            info = {
+                'url': video_url,
+                'ext': ext,
+                'tbr': bitrate,
+            }
 
         title = self._search_regex(
             [r'<meta content="([^"]+)" itemprop="description">',
-             r'class="title" itemprop="name">([^<]+)<'],
+             r'class="title" itemprop="name">([^<]+)<',
+             r'<h1>(.*?)</h1>'],
             webpage, 'title')
-        thumbnail = self._html_search_meta('thumbnail', webpage, 'thumbnail')
-        timestamp = parse_iso8601(self._html_search_meta(
-            'uploadDate', webpage, 'upload date'))
+        thumbnail = self._html_search_meta(['og:image', 'thumbnail'], webpage, 'thumbnail')
+        timestamp = parse_iso8601(self._og_search_property('video:release_date', webpage, 'upload date', fatal=False),
+                                  delimiter=' ') or parse_iso8601(self._html_search_meta('uploadDate', webpage, 'upload date'))
         duration = parse_duration(self._html_search_meta(
-            'duration', webpage, 'duration'))
+            ['duration', 'og:video:duration'], webpage, 'duration'))
         filesize_approx = parse_filesize(self._html_search_meta(
-            'contentSize', webpage, 'file size'))
-        bitrate = int_or_none(self._html_search_meta(
-            'bitrate', webpage, 'bitrate'))
+            'contentSize', webpage, 'file size', default=None))
         categories = self._html_search_meta(
             'keywords', webpage, 'categories', default='').split(',')
 
         age_limit = self._rta_search(webpage)
 
-        return {
+        info.update({
             'id': video_id,
             'display_id': display_id,
-            'url': video_url,
-            'ext': ext,
             'title': title,
             'thumbnail': thumbnail,
             'timestamp': timestamp,
             'duration': duration,
             'filesize_approx': filesize_approx,
-            'tbr': bitrate,
             'categories': categories,
             'age_limit': age_limit,
-        }
+        })
+        return info

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: