[ie/ZDF] Address review findings

yt_dlp/extractor/_extractors.py

@@ -2601,7 +2601,7 @@ from .zattoo import (
 )
 from .zdf import (
     ZDFIE,
-    ZDFCollectionIE,
+    ZDFChannelIE,
     ZDFHeuteIE,
 )
 from .zee5 import (

yt_dlp/extractor/dreisat.py

@@ -1,12 +1,10 @@
 from .zdf import ZDFBaseIE
 from ..utils import (
-    NO_DEFAULT,
-    ExtractorError,
     int_or_none,
     merge_dicts,
+    parse_iso8601,
+    require,
     traverse_obj,
-    try_get,
-    unified_timestamp,
 )
 
 
@@ -57,6 +55,23 @@ class DreiSatIE(ZDFBaseIE):
             'timestamp': 1743162540,
             'upload_date': '20250328',
         },
+    }, {
+        # Video with chapters
+        'url': 'https://www.3sat.de/kultur/buchmesse/dein-buch-das-beste-von-der-leipziger-buchmesse-2025-teil-1-100.html',
+        'md5': '65e8f365799d2266c9110d83d748445e',
+        'info_dict': {
+            'id': '250330_dein_buch1_bum',
+            'ext': 'mp4',
+            'title': 'dein buch  - Das Beste von der Leipziger Buchmesse 2025 - Teil 1',
+            'description': 'md5:bae51bfc22f15563ce3acbf97d2e8844',
+            'duration': 5399.0,
+            'thumbnail': 'https://www.3sat.de/assets/buchmesse-kerkeling-100~original?cb=1743329640903',
+            'chapters': 'count:24',
+            'episode': 'dein buch  - Das Beste von der Leipziger Buchmesse 2025 - Teil 1',
+            'episode_id': 'POS_1ef236cc-b390-401e-acd0-4fb4b04315fb',
+            'timestamp': 1743327000,
+            'upload_date': '20250330',
+        },
     }, {
         # Same as https://www.zdf.de/filme/filme-sonstige/der-hauptmann-112.html
         'url': 'https://www.3sat.de/film/spielfilm/der-hauptmann-100.html',
@@ -67,53 +82,35 @@ class DreiSatIE(ZDFBaseIE):
         'only_matching': True,
     }]
 
-    def _extract_player(self, webpage, video_id, fatal=True):
+    def _real_extract(self, url):
-        return self._parse_json(
+        video_id = self._match_id(url)
-            self._search_regex(
+        webpage = self._download_webpage(url, video_id)
-                r'(?s)data-zdfplayer-jsb=(["\'])(?P<json>{.+?})\1', webpage,
+        player = self._search_json(
-                'player JSON', default='{}' if not fatal else NO_DEFAULT,
+            r'data-zdfplayer-jsb=(["\'])', webpage, 'player JSON', video_id)
-                group='json'),
-            video_id)
-
-    def _extract_regular(self, url, player, video_id):
         player_url = player['content']
         api_token = f'Bearer {player["apiToken"]}'
 
         content = self._call_api(player_url, video_id, 'video metadata', api_token)
-        return self._extract_entry(player_url, api_token, content, video_id)
 
-    def _extract_entry(self, url, api_token, content, video_id):
+        video_target = content['mainVideoContent']['http://zdf.de/rels/target']
-        title = content.get('title') or content['teaserHeadline']
+        ptmd_path = traverse_obj(video_target, (
-
-        t = content['mainVideoContent']['http://zdf.de/rels/target']
-        ptmd_path = traverse_obj(t, (
             (('streams', 'default'), None),
             ('http://zdf.de/rels/streams/ptmd', 'http://zdf.de/rels/streams/ptmd-template'),
-        ), get_all=False)
+            {str}, any, {require('ptmd path')}))
-        if not ptmd_path:
+        info = self._extract_ptmd(player_url, ptmd_path, video_id, api_token)
-            raise ExtractorError('Could not extract ptmd_path')
-
-        info = self._extract_ptmd(url, ptmd_path, video_id, api_token)
-        layouts = try_get(
-            content, lambda x: x['teaserImageRef']['layouts'], dict)
-        thumbnails = self._extract_thumbnails(layouts)
-
-        chapter_marks = t.get('streamAnchorTag') or []
-        chapter_marks.append({'anchorOffset': int_or_none(t.get('duration'))})
-        chapters = [{
-            'start_time': chap.get('anchorOffset'),
-            'end_time': next_chap.get('anchorOffset'),
-            'title': chap.get('anchorLabel'),
-        } for chap, next_chap in zip(chapter_marks, chapter_marks[1:])]
 
         return merge_dicts(info, {
-            'title': title,
+            **traverse_obj(content, {
-            'description': content.get('leadParagraph') or content.get('teasertext'),
+                'title': (('title', 'teaserHeadline'), {str}, any),
-            'duration': int_or_none(t.get('duration')),
+                'episode': (('title', 'teaserHeadline'), {str}, any),
-            'timestamp': unified_timestamp(content.get('editorialDate')),
+                'description': (('leadParagraph', 'teasertext'), {str}, any),
-            'thumbnails': thumbnails,
+                'timestamp': ('editorialDate', {parse_iso8601}),
-            'chapters': chapters or None,
+            }),
-            'episode': title,
+            **traverse_obj(video_target, {
+                'duration': ('duration', {int_or_none}),
+                'chapters': ('streamAnchorTag', {self._extract_chapters}),
+            }),
+            'thumbnails': self._extract_thumbnails(traverse_obj(content, ('teaserImageRef', 'layouts', {dict}))),
             **traverse_obj(content, ('programmeItem', 0, 'http://zdf.de/rels/target', {
                 'series_id': ('http://zdf.de/rels/cmdm/series', 'seriesUuid', {str}),
                 'series': ('http://zdf.de/rels/cmdm/series', 'seriesTitle', {str}),
@@ -124,10 +121,3 @@ class DreiSatIE(ZDFBaseIE):
                 'episode_id': ('contentId', {str}),
             })),
         })
-
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-
-        webpage = self._download_webpage(url, video_id)
-        player = self._extract_player(webpage, url)
-        return self._extract_regular(url, player, video_id)

yt_dlp/extractor/phoenix.py

@@ -10,7 +10,7 @@ from ..utils import (
 
 class PhoenixIE(ZDFBaseIE):
     IE_NAME = 'phoenix.de'
-    _VALID_URL = r'https?://(?:www\.)?phoenix\.de/(?:[^/]+/)*[^/?#&]*-a-(?P<id>\d+)\.html'
+    _VALID_URL = r'https?://(?:www\.)?phoenix\.de/(?:[^/?#]+/)*[^/?#&]*-a-(?P<id>\d+)\.html'
     _TESTS = [{
         'url': 'https://www.phoenix.de/sendungen/dokumentationen/spitzbergen-a-893349.html',
         'md5': '768bbcd67a51fbfb22263ecdbc49146d',

yt_dlp/extractor/zdf.py

@@ -34,6 +34,12 @@ class ZDFBaseIE(InfoExtractor):
             url, video_id, note=f'Downloading {item}',
             errnote=f'Failed to download {item}', headers=headers)
 
+    def _extract_chapters(self, data):
+        return traverse_obj(data, (lambda _, v: 'anchorOffset' in v, {
+            'start_time': ('anchorOffset', {float_or_none}),
+            'title': ('anchorLabel', {str}),
+        }), default=None)
+
     @staticmethod
     def _extract_subtitles(src):
         seen_urls = set()
@@ -506,21 +512,12 @@ query VideoByCanonical($canonical: String!) {
                 'season_number': ('episodeInfo', 'seasonNumber', {int_or_none}),
                 'series': ('smartCollection', 'title', {str}),
                 'series_id': ('smartCollection', 'canonical', {str}),
-                'chapters': ('currentMedia', 'nodes', 0, {self._extract_chapters}),
+                'chapters': ('currentMedia', 'nodes', 0, 'streamAnchorTags', 'nodes', {self._extract_chapters}),
             })),
         }
 
-    def _extract_chapters(self, data):
-        chapter_marks = traverse_obj(data, ('streamAnchorTags', 'nodes'))
-        chapter_marks.append({'anchorOffset': float_or_none(data.get('duration'))})
-        return [{
-            'start_time': chap.get('anchorOffset'),
-            'end_time': next_chap.get('anchorOffset'),
-            'title': chap.get('anchorLabel'),
-        } for chap, next_chap in zip(chapter_marks, chapter_marks[1:])] or None
 
-
+class ZDFChannelIE(ZDFBaseIE):
-class ZDFCollectionIE(ZDFBaseIE):
     _VALID_URL = r'https?://www\.zdf\.de/(?:[^/?#]+/)*(?P<id>[^/?#]+)'
     _TESTS = [{
         # Playlist, legacy URL before website redesign in 2025-03