Merge fe02a15b1c35aad2463dfc6ff160e50cd8c2a04c 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

@@ -736,6 +736,7 @@ from .mtv import (
     MTVJapanIE,
 )
 from .muenchentv import MuenchenTVIE
+from .mujrozhlas import MujRozhlasIE
 from .mwave import MwaveIE, MwaveMeetGreetIE
 from .mychannels import MyChannelsIE
 from .myspace import MySpaceIE, MySpaceAlbumIE

youtube_dl/extractor/mujrozhlas.py

@@ -0,0 +1,95 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    clean_html,
+    ExtractorError,
+    js_to_json,
+)
+
+
+class MujRozhlasIE(InfoExtractor):
+    IE_NAME = 'mujRozhlas'
+    IE_DESC = 'https://www.mujrozhlas.cz/'
+    _VALID_URL = r'https?://www\.mujrozhlas\.cz/(?:[^/]+/)*(?P<id>[^/?#&]+)'
+    _TESTS = [
+        {
+            'url': 'https://www.mujrozhlas.cz/vinohradska-12/zadne-dalsi-pusy-konec-prchala-spol-muze-znamenat-babisuv-odchod-z-politiky-tipuje',
+            'md5': '34ecaa47f64079a63d6b80498c280e9d',
+            'info_dict': {
+                'id': '0c54ba72-93dd-3a29-b567-910d3d8c71a8',
+                'ext': 'mp3',
+                'title': 'Žádné další pusy. Konec Prchala a spol. může znamenat Babišův odchod z politiky, tipuje politolog',
+                'description': 'md5:ec0610bdb1f591061dbd224d2dd9c19e',
+            },
+        },
+        {
+            'url': 'https://www.mujrozhlas.cz/kazki/princi-ta-zliy-drakon',
+            'md5': 'cbad6f68db6dc4d6d798d69b5d258aa5',
+            'info_dict': {
+                'id': 'ec5f53b2-3910-448e-8e7f-d6d1a19f4926',
+                'ext': 'm4a',
+                'title': 'Принці та злий дракон',
+                'description': 'md5:b21701e09c2b509c4451194af7ac271b',
+            },
+            'params': {
+                'format': 'hls-128',
+            },
+        },
+    ]
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+        webpage = self._download_webpage(url, display_id)
+        player_data = self._search_regex(
+            r'\bvar dl = ({[^\n]+});',
+            webpage, 'player data', default=None)
+        if not player_data:
+            raise ExtractorError('Could not find player data')
+
+        player_data = self._parse_json(player_data, display_id, js_to_json)
+        audio_id = player_data['contentId']
+        bundle = player_data['siteEntityBundle']
+        if bundle not in ('episode', 'serialPart'):
+            raise ExtractorError('Unsupported entity: {0}'.format(bundle))
+
+        url = 'https://api.mujrozhlas.cz/episodes/{0}'.format(player_data['contentId'])
+        webpage = self._download_webpage(url, audio_id)
+        attr = self._parse_json(webpage, audio_id)['data']['attributes']
+
+        formats = []
+        for link in attr['audioLinks']:
+            variant = link['variant']
+            if variant == 'hls':
+                formats.extend(self._extract_m3u8_formats(
+                    link['url'], audio_id, 'm4a', entry_protocol='m3u8_native',
+                    m3u8_id='hls', fatal=False))
+            elif variant == 'dash':
+                formats.extend(self._extract_mpd_formats(
+                    link['url'], audio_id, mpd_id='dash', fatal=False))
+            elif variant == 'mp3':
+                url = link['url']
+                m = re.search(
+                    r'(?P<proto>[^:]+):(?:.*/)*(?P<id>[^.]+)\.(?P<ext>[^/.]+)$',
+                    url)
+                bitrate = link['bitrate']
+                formats.append({
+                    'url': link['url'],
+                    'protocol': m.group('proto'),
+                    'ext': m.group('ext'),
+                    'format_id': '-'.join(('mp3', str(bitrate))),
+                    'vcodec': 'none',
+                    'abr': bitrate,
+                    'tbr': bitrate,
+                })
+        self._sort_formats(formats)
+
+        return {
+            'id': audio_id,
+            'title': attr['title'],
+            'description': clean_html(attr['description']),
+            'formats': formats,
+        }

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: