Merge 8934a232fb1e2491932fb627df3b4329f8318317 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/douyin.py

@@ -0,0 +1,97 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..compat import compat_str
+from ..utils import (
+    ExtractorError,
+    int_or_none,
+    try_get,
+    orderedSet
+)
+
+
+class DouyinVideoIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?douyin\.com/video/(?P<id>[0-9]+)'
+    _TEST = {
+        'url': 'https://www.douyin.com/video/6989098563519270181',
+        'md5': '99c5667992b8a8d46c145907f677c92b',
+        'info_dict': {
+            'id': '6989098563519270181',
+            'url': 'https://aweme.snssdk.com/aweme/v1/playwm/?video_id=v0300fg10000c3v47dbc77u9fvb20tbg&ratio=720p&line=0',
+            'title': '杨集#我的家乡 ',
+            'uploader': '🌹永恒的爱🌹',
+            'uploader_id': '104081949894',
+            'timestamp': 1627276320000.0,
+            'ext': 'mp4'
+        }
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        iteminfo = self._download_json('https://www.douyin.com/web/api/v2/aweme/iteminfo',
+                                       video_id, query={'item_ids': video_id}) or {}
+        status_code = iteminfo.get('status_code', 'status_code missing')
+        if status_code:
+            raise ExtractorError('%s (%s)' % (iteminfo.get('status_msg', 'status_msg missing'), status_code), video_id=video_id)
+
+        item_list = iteminfo.get('item_list')
+        if not item_list:
+            raise ExtractorError('The video you want to download does not exist any more',
+                                 video_id=video_id, expected=True)
+        item = item_list[0]
+
+        return {
+            'id': video_id,
+            'title': item['desc'],
+            'url': item['video']['play_addr']['url_list'][0],
+            'uploader': try_get(item, lambda x: x['author']['nickname'], compat_str),
+            'uploader_id': try_get(item, lambda x: x['author']['uid'], compat_str),
+            'duration': int_or_none(item.get('duration') or try_get(item, lambda x: x['video']['duration'], int), scale=1000),
+            'timestamp': int_or_none(item.get('create_time'), invscale=1000),
+            'width': try_get(item, lambda x: x['video']['width'], int),
+            'height': try_get(item, lambda x: x['video']['height'], int),
+            'vbr': try_get(item, lambda x: x['video']['bit_rate'], int),
+            'ext': 'mp4'
+        }
+
+
+class DouyinUserIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?douyin\.com/user/(?P<id>[a-zA-Z0-9_\-]+)'
+    _TEST = {
+        'url': 'https://www.douyin.com/user/MS4wLjABAAAAP5Q7Z-SwleIzAACYIu-LrwGbEZzN2dc5PT3hGNSTkSM',
+        'info_dict': {
+            'id': 'MS4wLjABAAAAP5Q7Z-SwleIzAACYIu-LrwGbEZzN2dc5PT3hGNSTkSM'
+        },
+        'playlist_mincount': 31
+    }
+
+    def _real_extract(self, url):
+        sec_uid = self._match_id(url)
+
+        has_more = True
+        max_cursor = ''
+
+        entries = []
+        while has_more:
+            post = self._download_json('https://www.douyin.com/web/api/v2/aweme/post',
+                                       sec_uid, query={'sec_uid': sec_uid, 'max_cursor': max_cursor, 'count': 50}) or {}
+            status_code = post.get('status_code', 'status_code missing')
+            if status_code:
+                raise ExtractorError('%s (%s)' % (post.get('status_msg', 'status_msg missing'), status_code), video_id=sec_uid)
+
+            aweme_list = post.get('aweme_list')
+            if aweme_list is None:
+                raise ExtractorError('JSON response does not contain aweme_list', video_id=sec_uid)
+
+            entries.extend([self.url_result('https://www.douyin.com/video/%s' % aweme_id,
+                                            ie=DouyinVideoIE.ie_key(), video_id=aweme_id)
+                            for aweme_id in filter(None,
+                                                   (aweme.get('aweme_id') for aweme in aweme_list
+                                                    if isinstance(aweme, dict)))])
+
+            has_more = post.get('has_more')
+            max_cursor = post.get('max_cursor')
+
+        return self.playlist_result(orderedSet(entries), sec_uid)

youtube_dl/extractor/extractors.py

@@ -305,6 +305,10 @@ from .dfb import DFBIE
 from .dhm import DHMIE
 from .digg import DiggIE
 from .dotsub import DotsubIE
+from .douyin import (
+    DouyinVideoIE,
+    DouyinUserIE
+)
 from .douyutv import (
     DouyuShowIE,
     DouyuTVIE,

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: