Refactor playlist support for arte.tv to use API endpoint, instead of parsing HTML

yt_dlp/extractor/arte.py

@@ -309,42 +309,40 @@ class ArteTVPlaylistIE(ArteTVBaseIE):
         },
     }]
 
-    def _entries(self, playlist_data, playlist_id):
+    def _entries(self, season_ids, lang, playlist_id):
-        playlist_item_filter = lambda _, v: re.match(rf'collection_(?:videos|subcollection)_{playlist_id}', v['code'])
+        for season_id in season_ids:
-        collections = traverse_obj(playlist_data,
+            season_data = self._download_json(f'{self._API_BASE}/playlist/{lang}/{season_id}', season_id, headers={
-                                   ('data',
+                'x-validated-age': '18',
-                                    'zones',
+            })
-                                    playlist_item_filter,
+
-                                    'content',
+            collection = traverse_obj(season_data, ('data', 'attributes', 'items'))
-                                    'data',
+
-                                    ...))
+            for video in collection:
-
-        for video in collections:
                 yield {
                     '_type': 'url_transparent',
-                'url': 'https://www.arte.tv' + video['url'],
+                    'url': traverse_obj(video, ('link', 'url')),
                     'ie_key': ArteTVIE.ie_key(),
-                'id': video['id'],
+                    'id': video['providerId'],
                     'title': video.get('title'),
                     'alt_title': video.get('subtitle'),
-                'duration': int_or_none(traverse_obj(video, ('duration'))),
+                    'duration': int_or_none(traverse_obj(video, ('duration', 'seconds'))),
                     'age_limit': int_or_none(traverse_obj(video, 'ageRating')),
                 }
 
     def _real_extract(self, url):
+        _API_TOKEN = 'Nzc1Yjc1ZjJkYjk1NWFhN2I2MWEwMmRlMzAzNjI5NmU3NWU3ODg4ODJjOWMxNTMxYzEzZGRjYjg2ZGE4MmIwOA'
         lang, playlist_id = self._match_valid_url(url).group('lang', 'id')
-        webpage = self._download_webpage(url, playlist_id)
 
-        unescape_func = lambda jstring: jstring.replace('\\"', '"').replace('\\\\', '\\')
+        playlist_info = self._download_json(f'https://api.arte.tv/api/opa/v3/programs/{lang}/{playlist_id}', playlist_id,
-        json_data = self._search_json(r'\$L23.+?', webpage, 'series data',
+                                            headers={
-                                      playlist_id,
+                                            'Authorization': f'Bearer {_API_TOKEN}',
-                                      end_pattern=r'\],\[\[',
+                                            })
-                                      transform_source=unescape_func)
 
-        return self.playlist_result(self._entries(json_data, playlist_id),
+        season_ids = traverse_obj(playlist_info, ('programs', ..., 'children', (lambda _, v: v['catalogType'] == 'SEASON'), 'programId'))
+        return self.playlist_result(self._entries(season_ids, lang, playlist_id),
                                     playlist_id,
-                                    traverse_obj(json_data, ('data', 'metadata', 'title')),
+                                    traverse_obj(playlist_info, ('programs', ..., 'title')),
-                                    traverse_obj(json_data, ('data', 'metadata', 'description')))
+                                    traverse_obj(playlist_info, ('programs', ..., 'shortDescription')))
 
 
 class ArteTVCategoryIE(ArteTVBaseIE):