[ie/zdf] Fix extractors (#11041)

Closes #4782, Closes #10672
Authored by: InvalidUsernameException

yt_dlp/extractor/zdf.py

@@ -5,7 +5,6 @@
     NO_DEFAULT,
     ExtractorError,
     determine_ext,
-    extract_attributes,
     float_or_none,
     int_or_none,
     join_nonempty,
@@ -25,6 +24,11 @@ class ZDFBaseIE(InfoExtractor):
     _GEO_COUNTRIES = ['DE']
     _QUALITIES = ('auto', 'low', 'med', 'high', 'veryhigh', 'hd', 'fhd', 'uhd')
 
+    def _download_v2_doc(self, document_id):
+        return self._download_json(
+            f'https://zdf-prod-futura.zdf.de/mediathekV2/document/{document_id}',
+            document_id)
+
     def _call_api(self, url, video_id, item, api_token=None, referrer=None):
         headers = {}
         if api_token:
@@ -320,9 +324,7 @@ def _extract_regular(self, url, player, video_id):
         return self._extract_entry(player['content'], player, content, video_id)
 
     def _extract_mobile(self, video_id):
-        video = self._download_json(
+        video = self._download_v2_doc(video_id)
-            f'https://zdf-cdn.live.cellular.de/mediathekV2/document/{video_id}',
-            video_id)
 
         formats = []
         formitaeten = try_get(video, lambda x: x['document']['formitaeten'], list)
@@ -374,7 +376,7 @@ def _real_extract(self, url):
 
 
 class ZDFChannelIE(ZDFBaseIE):
-    _VALID_URL = r'https?://www\.zdf\.de/(?:[^/]+/)*(?P<id>[^/?#&]+)'
+    _VALID_URL = r'https?://www\.zdf\.de/(?:[^/?#]+/)*(?P<id>[^/?#&]+)'
     _TESTS = [{
         'url': 'https://www.zdf.de/sport/das-aktuelle-sportstudio',
         'info_dict': {
@@ -387,18 +389,19 @@ class ZDFChannelIE(ZDFBaseIE):
         'info_dict': {
             'id': 'planet-e',
             'title': 'planet e.',
+            'description': 'md5:87e3b9c66a63cf1407ee443d2c4eb88e',
         },
         'playlist_mincount': 50,
     }, {
         'url': 'https://www.zdf.de/gesellschaft/aktenzeichen-xy-ungeloest',
         'info_dict': {
             'id': 'aktenzeichen-xy-ungeloest',
-            'title': 'Aktenzeichen XY... ungelöst',
+            'title': 'Aktenzeichen XY... Ungelöst',
-            'entries': "lambda x: not any('xy580-fall1-kindermoerder-gesucht-100' in e['url'] for e in x)",
+            'description': 'md5:623ede5819c400c6d04943fa8100e6e7',
         },
         'playlist_mincount': 2,
     }, {
-        'url': 'https://www.zdf.de/filme/taunuskrimi/',
+        'url': 'https://www.zdf.de/serien/taunuskrimi/',
         'only_matching': True,
     }]
 
@@ -406,36 +409,41 @@ class ZDFChannelIE(ZDFBaseIE):
     def suitable(cls, url):
         return False if ZDFIE.suitable(url) else super().suitable(url)
 
-    def _og_search_title(self, webpage, fatal=False):
+    def _extract_entry(self, entry):
-        title = super()._og_search_title(webpage, fatal=fatal)
+        return self.url_result(
-        return re.split(r'\s+[-|]\s+ZDF(?:mediathek)?$', title or '')[0] or None
+            entry['sharingUrl'], ZDFIE, **traverse_obj(entry, {
+                'id': ('basename', {str}),
+                'title': ('titel', {str}),
+                'description': ('beschreibung', {str}),
+                'duration': ('length', {float_or_none}),
+                # TODO: seasonNumber and episodeNumber can be extracted but need to also be in ZDFIE
+            }))
+
+    def _entries(self, data, document_id):
+        for entry in traverse_obj(data, (
+            'cluster', lambda _, v: v['type'] == 'teaser',
+            # If 'brandId' differs, it is a 'You might also like' video. Filter these out
+            'teaser', lambda _, v: v['type'] == 'video' and v['brandId'] == document_id and v['sharingUrl'],
+        )):
+            yield self._extract_entry(entry)
 
     def _real_extract(self, url):
         channel_id = self._match_id(url)
-
         webpage = self._download_webpage(url, channel_id)
+        document_id = self._search_regex(
+            r'docId\s*:\s*(["\'])(?P<doc_id>(?:(?!\1).)+)\1', webpage, 'document id', group='doc_id')
+        data = self._download_v2_doc(document_id)
 
-        matches = re.finditer(
+        main_video = traverse_obj(data, (
-            rf'''<div\b[^>]*?\sdata-plusbar-id\s*=\s*(["'])(?P<p_id>[\w-]+)\1[^>]*?\sdata-plusbar-url=\1(?P<url>{ZDFIE._VALID_URL})\1''',
+            'cluster', lambda _, v: v['type'] == 'teaserContent',
-            webpage)
+            'teaser', lambda _, v: v['type'] == 'video' and v['basename'] and v['sharingUrl'], any)) or {}
 
-        if self._downloader.params.get('noplaylist', False):
+        if not self._yes_playlist(channel_id, main_video.get('basename')):
-            entry = next(
+            return self._extract_entry(main_video)
-                (self.url_result(m.group('url'), ie=ZDFIE.ie_key()) for m in matches),
-                None)
-            self.to_screen('Downloading just the main video because of --no-playlist')
-            if entry:
-                return entry
-        else:
-            self.to_screen(f'Downloading playlist {channel_id} - add --no-playlist to download just the main video')
 
-        def check_video(m):
+        return self.playlist_result(
-            v_ref = self._search_regex(
+            self._entries(data, document_id), channel_id,
-                r'''(<a\b[^>]*?\shref\s*=[^>]+?\sdata-target-id\s*=\s*(["']){}\2[^>]*>)'''.format(m.group('p_id')),
+            re.split(r'\s+[-|]\s+ZDF(?:mediathek)?$', self._og_search_title(webpage) or '')[0] or None,
-                webpage, 'check id', default='')
+            join_nonempty(
-            v_ref = extract_attributes(v_ref)
+                'headline', 'text', delim='\n\n',
-            return v_ref.get('data-target-video-type') != 'novideo'
+                from_dict=traverse_obj(data, ('shortText', {dict}), default={})) or None)
-
-        return self.playlist_from_matches(
-            (m.group('url') for m in matches if check_video(m)),
-            channel_id, self._og_search_title(webpage, fatal=False))