Merge a66ef4baaceb9f3104dfb031e96269ffc72ae70c 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

@@ -499,6 +499,7 @@ from .hrti import (
     HRTiIE,
     HRTiPlaylistIE,
 )
+from .hse24 import HSE24IE
 from .huajiao import HuajiaoIE
 from .huffpost import HuffPostIE
 from .hungama import (

youtube_dl/extractor/hse24.py

@@ -0,0 +1,95 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from time import mktime, strptime
+
+
+class HSE24IE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?hse24\.de/dpl/(?:p/product|c/tv-shows)?/(?P<id>[0-9]+)'
+    _GEO_COUNTRIES = ['DE']
+    _TESTS = [{
+        'url': 'https://www.hse24.de/dpl/c/tv-shows/476357',
+        'info_dict': {
+            'id': '476357',
+            'ext': 'mp4',
+            'title': 'Jürgen Hirsch Schuhe zum Wohlfühlen',
+            'timestamp': 1604926800,
+            'upload_date': '20201109',
+            'channel': 'HSE24EXTRA',
+            'uploader': 'Daniela Elger'
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        'url': 'https://www.hse24.de/dpl/p/product/408630',
+        'info_dict': {
+            'id': '408630',
+            'ext': 'mp4',
+            'title': 'Hose im Ponte-Mix',
+            'timestamp': 1603058400,
+            'upload_date': '20201018',
+            'channel': 'HSE24',
+            'uploader': 'Judith Williams'
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+
+        json_str = self._html_search_regex(r'window\.__REDUX_DATA__ = ({.*});?', webpage, 'json_str')
+        json_str = json_str.replace('\n', '')
+        json_data = self._parse_json(json_str, video_id)
+
+        formats = []
+        if 'tvShow' in json_data:  # /dpl/c/tv-shows
+            show = json_data['tvShow']['tvShow']
+            title, date, hour, uploader = show['title'], show['date'], show['hour'], show['presenter']
+            channel = self._search_regex(r'tvShow \| ([A-Z0-9]+)_', show['actionFieldText'], video_id)
+            timestamp = mktime(strptime(' '.join((date, hour)), '%Y-%m-%d %H'))
+
+            video = json_data['tvShow']['tvShowVideo']
+            thumbnail = video['poster']
+            sources = video['sources']
+            for src in sources:
+                if src['mimetype'] == 'application/x-mpegURL':
+                    formats.extend(self._extract_m3u8_formats(src['url'], video_id, ext='mp4'))
+        elif 'productDetail' in json_data:  # /dpl/p/product
+            product = json_data['productDetail']['product']
+            title, uploader = product['name']['short'], product['brand']['brandName']
+            channel = 'HSE24'
+            default_time = '2001-01-01T01:01:01+0100'
+            time = product['variants'][0]['price']['validFrom'] if len(product['variants']) > 0 else default_time
+            # python2 compatibility (no %z directive support)
+            time = strptime(time[:-6], '%Y-%m-%dT%H:%M:%S')
+            if time[-6] == '+':
+                time += strptime(time[-6:], '%H:%M')
+            elif time[-6] == '-':
+                time -= strptime(time[-6:], '%H:%M')
+            timestamp = mktime(time)
+
+            for video in json_data['productContent']['productContent']['videos']:
+                thumbnail = video['poster']
+                sources = video['sources']
+                for src in sources:
+                    if src['mimetype'] == 'application/x-mpegURL':
+                        formats.extend(self._extract_m3u8_formats(src['url'], video_id, ext='mp4'))
+                    # elif src['mimetype'] == 'video/mp4':
+                    #     formats.append({'url': src['url']})
+
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': title,
+            'formats': formats,
+            'timestamp': int(timestamp),
+            'thumbnail': thumbnail,
+            'channel': channel,
+            'uploader': uploader
+        }

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: