Merge 05709d989693f9ff988db9549ba0dea6ec76b218 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/rtvs.py

@@ -3,17 +3,36 @@ from __future__ import unicode_literals
 
 from .common import InfoExtractor
 
+from ..utils import (
+    extract_attributes,
+    merge_dicts,
+    mimetype2ext,
+    parse_duration,
+    parse_qs,
+    T,
+    traverse_obj,
+    unified_timestamp,
+    url_or_none,
+)
+
 
 class RTVSIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?rtvs\.sk/(?:radio|televizia)/archiv/\d+/(?P<id>\d+)'
+    _VALID_URL = r'https?://(?:www\.)?(?:rtvs|stvr)\.sk/(?:radio|televizia)/archiv(?:/\d+)?/(?P<id>\d+)'
     _TESTS = [{
         # radio archive
         'url': 'http://www.rtvs.sk/radio/archiv/11224/414872',
         'md5': '134d5d6debdeddf8a5d761cbc9edacb8',
         'info_dict': {
             'id': '414872',
+            'display_id': '135331',
             'ext': 'mp3',
-            'title': 'Ostrov pokladov 1 časť.mp3'
+            'title': 'Ostrov pokladov 1 časť.mp3',
+            'duration': 2854,
+            'thumbnail': (
+                r're:https?://www\.(?:(?P<stvr>stvr)|rtvs)\.sk'
+                '/media/a501/image/file/2/0000'
+                r'/(?(stvr)rtvs-00009383\.png|b1R8\.rtvs\.jpg)$'
+            ),
         },
         'params': {
             'skip_download': True,
@@ -26,22 +45,99 @@ class RTVSIE(InfoExtractor):
             'id': '63118',
             'ext': 'mp4',
             'title': 'Amaro Džives - Náš deň',
-            'description': 'Galavečer pri príležitosti Medzinárodného dňa Rómov.'
+            'description': r're:Cenu Romipen odovzdá .{292} Medzinárodný deň Rómov a prezentovať rómsku kultúru\.$',
+            'timestamp': 1428523500,
+            'upload_date': '20150408',
+            'thumbnail': (
+                r're:https://www\.(?:stvr|rtvs)\.sk'
+                '/media/a501/image/file/2/0031'
+                r'/L7Qm\.amaro_dzives_png\.jpg$'),
+            'duration': 4986,
         },
         'params': {
             'skip_download': True,
         }
+    }, {
+        # tv archive
+        'url': 'https://www.rtvs.sk/televizia/archiv/18083?utm_source=web&utm_medium=rozcestnik&utm_campaign=Robin',
+        'info_dict': {
+            'id': '18083',
+            'display_id': '457086',
+            'ext': 'mp4',
+            'title': 'Robin - Vodiace psy /2.časť/',
+            'description': r're:Richard so svojím psom Robinom .{49} úlohu vodiace psíky\.$',
+            'timestamp': 1711875000,
+            'upload_date': '20240331',
+            'duration': 931,
+            'thumbnail': r're:https://www\.(:rtvs|stvr)\.sk/media/a501/image/file/2/0916/robin\.jpg$',
+        }
+    }, {
+        # new domain
+        'url': 'https://www.stvr.sk/televizia/archiv/15135/512191',
+        'info_dict': {
+            'id': '512191',
+            'ext': 'mp4',
+            'title': 'Retro noviny',
+            'description': r're:Kolekciu spravodajských šotov .{62} uvádza Milan Antonič, archivár STV\.',
+            'timestamp': 1737797700,
+            'upload_date': '20250125',
+            'duration': 1686,
+            'thumbnail': 'https://www.stvr.sk/media/a501/image/file/2/0519/NSF4.retro_noviny_jpg.jpg',
+        }
     }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
         webpage = self._download_webpage(url, video_id)
+        info = self._search_json_ld(webpage, video_id, expected_type='VideoObject', default={})
+        iframe_url = info and self._search_regex(
+            r'"embedUrl"\s*:\s*"(https?://[^"]+)"', webpage, 'Iframe URL', default=None)
+        if not iframe_url:
+            iframe = self._search_regex(
+                r'''(<iframe\s[^>]*(?<!-)\bid\s*=\s*['"]?player_(?:(?!_)\w)+_\d+[^>]+>)''',
+                webpage, 'Iframe')
+            iframe = extract_attributes(iframe)
 
-        playlist_url = self._search_regex(
-            r'playlist["\']?\s*:\s*(["\'])(?P<url>(?:(?!\1).)+)\1', webpage,
-            'playlist url', group='url')
+            iframe_url = iframe['src']
 
-        data = self._download_json(
-            playlist_url, video_id, 'Downloading playlist')[0]
-        return self._parse_jwplayer_data(data, video_id=video_id)
+        webpage = self._download_webpage(iframe_url, video_id, 'Downloading iframe')
+        json_url = self._search_regex(r'var\s+url\s*=\s*"([^"]+)"\s*\+\s*ruurl', webpage, 'json URL')
+        data = self._download_json('https:{0}b=mozilla&p=win&v=97&f=0&d=1'.format(json_url), video_id)
+
+        if traverse_obj(data, 'clip', expected_type=dict):
+            data['playlist'] = [data['clip']]
+
+        formats = []
+        _require = lambda k: (lambda x: x if x.get(k) else None)
+        for item in traverse_obj(data, (
+                'playlist', 0, 'sources', Ellipsis, {
+                    'src': ('src', T(url_or_none)),
+                    'type': 'type',
+                }, T(_require('src')))):
+            item_type = item.get('type')
+            if item_type == 'application/x-mpegurl':
+                formats.extend(self._extract_m3u8_formats(
+                    item['src'], video_id, ext='mp4',
+                    entry_protocol='m3u8_native', fatal=False))
+            elif item_type == 'application/dash+xml':
+                formats.extend(self._extract_mpd_formats(
+                    item['src'], video_id, fatal=False))
+            else:
+                formats.append({
+                    'url': item['src'],
+                    'ext': mimetype2ext(item_type),
+                })
+        self._sort_formats(formats)
+
+        return merge_dicts({
+            'id': video_id,
+            'display_id': traverse_obj(parse_qs(json_url), ('id', -1)),
+            'formats': formats,
+        }, info or traverse_obj(data, ('playlist', 0, {
+            'title': 'title',
+            'description': ('description', T(lambda s: s.strip())),
+            'duration': ('length', T(parse_duration)),
+            'thumbnail': ('image', T(url_or_none)),
+            'timestamp': ('datetime_create', T(unified_timestamp)),
+        })))

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: