yt_dlp/extractor/ard.py

   1 import functools
   2 import re
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     OnDemandPagedList,
   7     bug_reports_message,
   8     determine_ext,
   9     int_or_none,
  10     join_nonempty,
  11     jwt_decode_hs256,
  12     make_archive_id,
  13     parse_duration,
  14     parse_iso8601,
  15     remove_start,
  16     str_or_none,
  17     unified_strdate,
  18     update_url_query,
  19     url_or_none,
  20     xpath_text,
  21 )
  22 from ..utils.traversal import traverse_obj
  23
  24
  25 class ARDMediathekBaseIE(InfoExtractor):
  26     _GEO_COUNTRIES = ['DE']
  27
  28     def _extract_media_info(self, media_info_url, webpage, video_id):
  29         media_info = self._download_json(
  30             media_info_url, video_id, 'Downloading media JSON')
  31         return self._parse_media_info(media_info, video_id, '"fsk"' in webpage)
  32
  33     def _parse_media_info(self, media_info, video_id, fsk):
  34         formats = self._extract_formats(media_info, video_id)
  35
  36         if not formats:
  37             if fsk:
  38                 self.raise_no_formats(
  39                     'This video is only available after 20:00', expected=True)
  40             elif media_info.get('_geoblocked'):
  41                 self.raise_geo_restricted(
  42                     'This video is not available due to geoblocking',
  43                     countries=self._GEO_COUNTRIES, metadata_available=True)
  44
  45         subtitles = {}
  46         subtitle_url = media_info.get('_subtitleUrl')
  47         if subtitle_url:
  48             subtitles['de'] = [{
  49                 'ext': 'ttml',
  50                 'url': subtitle_url,
  51             }, {
  52                 'ext': 'vtt',
  53                 'url': subtitle_url.replace('/ebutt/', '/webvtt/') + '.vtt',
  54             }]
  55
  56         return {
  57             'id': video_id,
  58             'duration': int_or_none(media_info.get('_duration')),
  59             'thumbnail': media_info.get('_previewImage'),
  60             'is_live': media_info.get('_isLive') is True,
  61             'formats': formats,
  62             'subtitles': subtitles,
  63         }
  64
  65     def _extract_formats(self, media_info, video_id):
  66         type_ = media_info.get('_type')
  67         media_array = media_info.get('_mediaArray', [])
  68         formats = []
  69         for num, media in enumerate(media_array):
  70             for stream in media.get('_mediaStreamArray', []):
  71                 stream_urls = stream.get('_stream')
  72                 if not stream_urls:
  73                     continue
  74                 if not isinstance(stream_urls, list):
  75                     stream_urls = [stream_urls]
  76                 quality = stream.get('_quality')
  77                 server = stream.get('_server')
  78                 for stream_url in stream_urls:
  79                     if not url_or_none(stream_url):
  80                         continue
  81                     ext = determine_ext(stream_url)
  82                     if quality != 'auto' and ext in ('f4m', 'm3u8'):
  83                         continue
  84                     if ext == 'f4m':
  85                         formats.extend(self._extract_f4m_formats(
  86                             update_url_query(stream_url, {
  87                                 'hdcore': '3.1.1',
  88                                 'plugin': 'aasp-3.1.1.69.124',
  89                             }), video_id, f4m_id='hds', fatal=False))
  90                     elif ext == 'm3u8':
  91                         formats.extend(self._extract_m3u8_formats(
  92                             stream_url, video_id, 'mp4', 'm3u8_native',
  93                             m3u8_id='hls', fatal=False))
  94                     else:
  95                         if server and server.startswith('rtmp'):
  96                             f = {
  97                                 'url': server,
  98                                 'play_path': stream_url,
  99                                 'format_id': f'a{num}-rtmp-{quality}',
 100                             }
 101                         else:
 102                             f = {
 103                                 'url': stream_url,
 104                                 'format_id': f'a{num}-{ext}-{quality}',
 105                             }
 106                         m = re.search(
 107                             r'_(?P<width>\d+)x(?P<height>\d+)\.mp4$',
 108                             stream_url)
 109                         if m:
 110                             f.update({
 111                                 'width': int(m.group('width')),
 112                                 'height': int(m.group('height')),
 113                             })
 114                         if type_ == 'audio':
 115                             f['vcodec'] = 'none'
 116                         formats.append(f)
 117         return formats
 118
 119
 120 class ARDIE(InfoExtractor):
 121     _VALID_URL = r'(?P<mainurl>https?://(?:www\.)?daserste\.de/(?:[^/?#&]+/)+(?P<id>[^/?#&]+))\.html'
 122     _TESTS = [{
 123         # available till 7.12.2023
 124         'url': 'https://www.daserste.de/information/talk/maischberger/videos/maischberger-video-424.html',
 125         'md5': '94812e6438488fb923c361a44469614b',
 126         'info_dict': {
 127             'id': 'maischberger-video-424',
 128             'display_id': 'maischberger-video-424',
 129             'ext': 'mp4',
 130             'duration': 4452.0,
 131             'title': 'maischberger am 07.12.2022',
 132             'upload_date': '20221207',
 133             'thumbnail': r're:^https?://.*\.jpg$',
 134         },
 135     }, {
 136         'url': 'https://www.daserste.de/information/politik-weltgeschehen/morgenmagazin/videosextern/dominik-kahun-aus-der-nhl-direkt-zur-weltmeisterschaft-100.html',
 137         'only_matching': True,
 138     }, {
 139         'url': 'https://www.daserste.de/information/nachrichten-wetter/tagesthemen/videosextern/tagesthemen-17736.html',
 140         'only_matching': True,
 141     }, {
 142         'url': 'https://www.daserste.de/unterhaltung/serie/in-aller-freundschaft-die-jungen-aerzte/videos/diversity-tag-sanam-afrashteh100.html',
 143         'only_matching': True,
 144     }, {
 145         'url': 'http://www.daserste.de/information/reportage-dokumentation/dokus/videos/die-story-im-ersten-mission-unter-falscher-flagge-100.html',
 146         'only_matching': True,
 147     }, {
 148         'url': 'https://www.daserste.de/unterhaltung/serie/in-aller-freundschaft-die-jungen-aerzte/Drehpause-100.html',
 149         'only_matching': True,
 150     }, {
 151         'url': 'https://www.daserste.de/unterhaltung/film/filmmittwoch-im-ersten/videos/making-ofwendezeit-video-100.html',
 152         'only_matching': True,
 153     }]
 154
 155     def _real_extract(self, url):
 156         mobj = self._match_valid_url(url)
 157         display_id = mobj.group('id')
 158
 159         player_url = mobj.group('mainurl') + '~playerXml.xml'
 160         doc = self._download_xml(player_url, display_id)
 161         video_node = doc.find('./video')
 162         upload_date = unified_strdate(xpath_text(
 163             video_node, './broadcastDate'))
 164         thumbnail = xpath_text(video_node, './/teaserImage//variant/url')
 165
 166         formats = []
 167         for a in video_node.findall('.//asset'):
 168             file_name = xpath_text(a, './fileName', default=None)
 169             if not file_name:
 170                 continue
 171             format_type = a.attrib.get('type')
 172             format_url = url_or_none(file_name)
 173             if format_url:
 174                 ext = determine_ext(file_name)
 175                 if ext == 'm3u8':
 176                     formats.extend(self._extract_m3u8_formats(
 177                         format_url, display_id, 'mp4', entry_protocol='m3u8_native',
 178                         m3u8_id=format_type or 'hls', fatal=False))
 179                     continue
 180                 elif ext == 'f4m':
 181                     formats.extend(self._extract_f4m_formats(
 182                         update_url_query(format_url, {'hdcore': '3.7.0'}),
 183                         display_id, f4m_id=format_type or 'hds', fatal=False))
 184                     continue
 185             f = {
 186                 'format_id': format_type,
 187                 'width': int_or_none(xpath_text(a, './frameWidth')),
 188                 'height': int_or_none(xpath_text(a, './frameHeight')),
 189                 'vbr': int_or_none(xpath_text(a, './bitrateVideo')),
 190                 'abr': int_or_none(xpath_text(a, './bitrateAudio')),
 191                 'vcodec': xpath_text(a, './codecVideo'),
 192                 'tbr': int_or_none(xpath_text(a, './totalBitrate')),
 193             }
 194             server_prefix = xpath_text(a, './serverPrefix', default=None)
 195             if server_prefix:
 196                 f.update({
 197                     'url': server_prefix,
 198                     'playpath': file_name,
 199                 })
 200             else:
 201                 if not format_url:
 202                     continue
 203                 f['url'] = format_url
 204             formats.append(f)
 205
 206         _SUB_FORMATS = (
 207             ('./dataTimedText', 'ttml'),
 208             ('./dataTimedTextNoOffset', 'ttml'),
 209             ('./dataTimedTextVtt', 'vtt'),
 210         )
 211
 212         subtitles = {}
 213         for subsel, subext in _SUB_FORMATS:
 214             for node in video_node.findall(subsel):
 215                 subtitles.setdefault('de', []).append({
 216                     'url': node.attrib['url'],
 217                     'ext': subext,
 218                 })
 219
 220         return {
 221             'id': xpath_text(video_node, './videoId', default=display_id),
 222             'formats': formats,
 223             'subtitles': subtitles,
 224             'display_id': display_id,
 225             'title': video_node.find('./title').text,
 226             'duration': parse_duration(video_node.find('./duration').text),
 227             'upload_date': upload_date,
 228             'thumbnail': thumbnail,
 229         }
 230
 231
 232 class ARDBetaMediathekIE(InfoExtractor):
 233     IE_NAME = 'ARDMediathek'
 234     _VALID_URL = r'''(?x)https?://
 235         (?:(?:beta|www)\.)?ardmediathek\.de/
 236         (?:[^/]+/)?
 237         (?:player|live|video)/
 238         (?:[^?#]+/)?
 239         (?P<id>[a-zA-Z0-9]+)
 240         /?(?:[?#]|$)'''
 241     _GEO_COUNTRIES = ['DE']
 242     _TOKEN_URL = 'https://sso.ardmediathek.de/sso/token'
 243
 244     _TESTS = [{
 245         'url': 'https://www.ardmediathek.de/video/filme-im-mdr/liebe-auf-vier-pfoten/mdr-fernsehen/Y3JpZDovL21kci5kZS9zZW5kdW5nLzI4MjA0MC80MjIwOTEtNDAyNTM0',
 246         'md5': 'b6e8ab03f2bcc6e1f9e6cef25fcc03c4',
 247         'info_dict': {
 248             'display_id': 'Y3JpZDovL21kci5kZS9zZW5kdW5nLzI4MjA0MC80MjIwOTEtNDAyNTM0',
 249             'id': '12939099',
 250             'title': 'Liebe auf vier Pfoten',
 251             'description': r're:^Claudia Schmitt, Anwältin in Salzburg',
 252             'duration': 5222,
 253             'thumbnail': 'https://api.ardmediathek.de/image-service/images/urn:ard:image:aee7cbf8f06de976?w=960&ch=ae4d0f2ee47d8b9b',
 254             'timestamp': 1701343800,
 255             'upload_date': '20231130',
 256             'ext': 'mp4',
 257             'episode': 'Liebe auf vier Pfoten',
 258             'series': 'Filme im MDR',
 259             'age_limit': 0,
 260             'channel': 'MDR',
 261             '_old_archive_ids': ['ardbetamediathek Y3JpZDovL21kci5kZS9zZW5kdW5nLzI4MjA0MC80MjIwOTEtNDAyNTM0'],
 262         },
 263     }, {
 264         'url': 'https://www.ardmediathek.de/mdr/video/die-robuste-roswita/Y3JpZDovL21kci5kZS9iZWl0cmFnL2Ntcy84MWMxN2MzZC0wMjkxLTRmMzUtODk4ZS0wYzhlOWQxODE2NGI/',
 265         'md5': 'a1dc75a39c61601b980648f7c9f9f71d',
 266         'info_dict': {
 267             'display_id': 'die-robuste-roswita',
 268             'id': '78566716',
 269             'title': 'Die robuste Roswita',
 270             'description': r're:^Der Mord.*totgeglaubte Ehefrau Roswita',
 271             'duration': 5316,
 272             'thumbnail': 'https://img.ardmediathek.de/standard/00/78/56/67/84/575672121/16x9/960?mandant=ard',
 273             'timestamp': 1596658200,
 274             'upload_date': '20200805',
 275             'ext': 'mp4',
 276         },
 277         'skip': 'Error',
 278     }, {
 279         'url': 'https://www.ardmediathek.de/video/tagesschau-oder-tagesschau-20-00-uhr/das-erste/Y3JpZDovL2Rhc2Vyc3RlLmRlL3RhZ2Vzc2NoYXUvZmM4ZDUxMjgtOTE0ZC00Y2MzLTgzNzAtNDZkNGNiZWJkOTll',
 280         'md5': '1e73ded21cb79bac065117e80c81dc88',
 281         'info_dict': {
 282             'id': '10049223',
 283             'ext': 'mp4',
 284             'title': 'tagesschau, 20:00 Uhr',
 285             'timestamp': 1636398000,
 286             'description': 'md5:39578c7b96c9fe50afdf5674ad985e6b',
 287             'upload_date': '20211108',
 288             'display_id': 'Y3JpZDovL2Rhc2Vyc3RlLmRlL3RhZ2Vzc2NoYXUvZmM4ZDUxMjgtOTE0ZC00Y2MzLTgzNzAtNDZkNGNiZWJkOTll',
 289             'duration': 915,
 290             'episode': 'tagesschau, 20:00 Uhr',
 291             'series': 'tagesschau',
 292             'thumbnail': 'https://api.ardmediathek.de/image-service/images/urn:ard:image:fbb21142783b0a49?w=960&ch=ee69108ae344f678',
 293             'channel': 'ARD-Aktuell',
 294             '_old_archive_ids': ['ardbetamediathek Y3JpZDovL2Rhc2Vyc3RlLmRlL3RhZ2Vzc2NoYXUvZmM4ZDUxMjgtOTE0ZC00Y2MzLTgzNzAtNDZkNGNiZWJkOTll'],
 295         },
 296     }, {
 297         'url': 'https://www.ardmediathek.de/video/7-tage/7-tage-unter-harten-jungs/hr-fernsehen/N2I2YmM5MzgtNWFlOS00ZGFlLTg2NzMtYzNjM2JlNjk4MDg3',
 298         'md5': 'c428b9effff18ff624d4f903bda26315',
 299         'info_dict': {
 300             'id': '94834686',
 301             'ext': 'mp4',
 302             'duration': 2670,
 303             'episode': '7 Tage ... unter harten Jungs',
 304             'description': 'md5:0f215470dcd2b02f59f4bd10c963f072',
 305             'upload_date': '20231005',
 306             'timestamp': 1696491171,
 307             'display_id': 'N2I2YmM5MzgtNWFlOS00ZGFlLTg2NzMtYzNjM2JlNjk4MDg3',
 308             'series': '7 Tage ...',
 309             'channel': 'HR',
 310             'thumbnail': 'https://api.ardmediathek.de/image-service/images/urn:ard:image:430c86d233afa42d?w=960&ch=fa32ba69bc87989a',
 311             'title': '7 Tage ... unter harten Jungs',
 312             '_old_archive_ids': ['ardbetamediathek N2I2YmM5MzgtNWFlOS00ZGFlLTg2NzMtYzNjM2JlNjk4MDg3'],
 313         },
 314     }, {
 315         'url': 'https://www.ardmediathek.de/video/lokalzeit-aus-duesseldorf/lokalzeit-aus-duesseldorf-oder-31-10-2024/wdr-duesseldorf/Y3JpZDovL3dkci5kZS9CZWl0cmFnLXNvcGhvcmEtOWFkMTc0ZWMtMDA5ZS00ZDEwLWFjYjctMGNmNTdhNzVmNzUz',
 316         'info_dict': {
 317             'id': '13847165',
 318             'chapters': 'count:8',
 319             'ext': 'mp4',
 320             'channel': 'WDR',
 321             'display_id': 'Y3JpZDovL3dkci5kZS9CZWl0cmFnLXNvcGhvcmEtOWFkMTc0ZWMtMDA5ZS00ZDEwLWFjYjctMGNmNTdhNzVmNzUz',
 322             'episode': 'Lokalzeit aus Düsseldorf | 31.10.2024',
 323             'series': 'Lokalzeit aus Düsseldorf',
 324             'thumbnail': 'https://api.ardmediathek.de/image-service/images/urn:ard:image:f02ec9bd9b7bd5f6?w=960&ch=612491dcd5e09b0c',
 325             'title': 'Lokalzeit aus Düsseldorf | 31.10.2024',
 326             'upload_date': '20241031',
 327             'timestamp': 1730399400,
 328             'description': 'md5:12db30b3b706314efe3778b8df1a7058',
 329             'duration': 1759,
 330             '_old_archive_ids': ['ardbetamediathek Y3JpZDovL3dkci5kZS9CZWl0cmFnLXNvcGhvcmEtOWFkMTc0ZWMtMDA5ZS00ZDEwLWFjYjctMGNmNTdhNzVmNzUz'],
 331         },
 332     }, {
 333         'url': 'https://beta.ardmediathek.de/ard/video/Y3JpZDovL2Rhc2Vyc3RlLmRlL3RhdG9ydC9mYmM4NGM1NC0xNzU4LTRmZGYtYWFhZS0wYzcyZTIxNGEyMDE',
 334         'only_matching': True,
 335     }, {
 336         'url': 'https://ardmediathek.de/ard/video/saartalk/saartalk-gesellschaftsgift-haltung-gegen-hass/sr-fernsehen/Y3JpZDovL3NyLW9ubGluZS5kZS9TVF84MTY4MA/',
 337         'only_matching': True,
 338     }, {
 339         'url': 'https://www.ardmediathek.de/ard/video/trailer/private-eyes-s01-e01/one/Y3JpZDovL3dkci5kZS9CZWl0cmFnLTE1MTgwYzczLWNiMTEtNGNkMS1iMjUyLTg5MGYzOWQxZmQ1YQ/',
 340         'only_matching': True,
 341     }, {
 342         'url': 'https://www.ardmediathek.de/ard/player/Y3JpZDovL3N3ci5kZS9hZXgvbzEwNzE5MTU/',
 343         'only_matching': True,
 344     }, {
 345         'url': 'https://www.ardmediathek.de/swr/live/Y3JpZDovL3N3ci5kZS8xMzQ4MTA0Mg',
 346         'only_matching': True,
 347     }, {
 348         'url': 'https://www.ardmediathek.de/video/coronavirus-update-ndr-info/astrazeneca-kurz-lockdown-und-pims-syndrom-81/ndr/Y3JpZDovL25kci5kZS84NzE0M2FjNi0wMWEwLTQ5ODEtOTE5NS1mOGZhNzdhOTFmOTI/',
 349         'only_matching': True,
 350     }]
 351
 352     def _extract_episode_info(self, title):
 353         patterns = [
 354             # Pattern for title like "Homo sapiens (S06/E07) - Originalversion"
 355             # from: https://www.ardmediathek.de/one/sendung/doctor-who/Y3JpZDovL3dkci5kZS9vbmUvZG9jdG9yIHdobw
 356             r'.*(?P<ep_info> \(S(?P<season_number>\d+)/E(?P<episode_number>\d+)\)).*',
 357             # E.g.: title="Fritjof aus Norwegen (2) (AD)"
 358             # from: https://www.ardmediathek.de/ard/sammlung/der-krieg-und-ich/68cMkqJdllm639Skj4c7sS/
 359             r'.*(?P<ep_info> \((?:Folge |Teil )?(?P<episode_number>\d+)(?:/\d+)?\)).*',
 360             r'.*(?P<ep_info>Folge (?P<episode_number>\d+)(?:\:| -|) )\"(?P<episode>.+)\".*',
 361             # E.g.: title="Folge 25/42: Symmetrie"
 362             # from: https://www.ardmediathek.de/ard/video/grips-mathe/folge-25-42-symmetrie/ard-alpha/Y3JpZDovL2JyLmRlL3ZpZGVvLzMyYzI0ZjczLWQ1N2MtNDAxNC05ZmZhLTFjYzRkZDA5NDU5OQ/
 363             # E.g.: title="Folge 1063 - Vertrauen"
 364             # from: https://www.ardmediathek.de/ard/sendung/die-fallers/Y3JpZDovL3N3ci5kZS8yMzAyMDQ4/
 365             r'.*(?P<ep_info>Folge (?P<episode_number>\d+)(?:/\d+)?(?:\:| -|) ).*',
 366             # As a fallback use the full title
 367             r'(?P<title>.*)',
 368         ]
 369
 370         return traverse_obj(patterns, (..., {functools.partial(re.match, string=title)}, {
 371             'season_number': ('season_number', {int_or_none}),
 372             'episode_number': ('episode_number', {int_or_none}),
 373             'episode': ((
 374                 ('episode', {str_or_none}),
 375                 ('ep_info', {lambda x: title.replace(x, '')}),
 376                 ('title', {str}),
 377             ), {str.strip}),
 378         }), get_all=False)
 379
 380     def _real_extract(self, url):
 381         display_id = self._match_id(url)
 382         query = {'embedded': 'false', 'mcV6': 'true'}
 383         headers = {}
 384
 385         if self._get_cookies(self._TOKEN_URL).get('ams'):
 386             token = self._download_json(
 387                 self._TOKEN_URL, display_id, 'Fetching token for age verification',
 388                 'Unable to fetch age verification token', fatal=False)
 389             id_token = traverse_obj(token, ('idToken', {str}))
 390             decoded_token = traverse_obj(id_token, ({jwt_decode_hs256}, {dict}))
 391             user_id = traverse_obj(decoded_token, (('user_id', 'sub'), {str}), get_all=False)
 392             if not user_id:
 393                 self.report_warning('Unable to extract token, continuing without authentication')
 394             else:
 395                 headers['x-authorization'] = f'Bearer {id_token}'
 396                 query['userId'] = user_id
 397                 if decoded_token.get('age_rating') != 18:
 398                     self.report_warning('Account is not verified as 18+; video may be unavailable')
 399
 400         page_data = self._download_json(
 401             f'https://api.ardmediathek.de/page-gateway/pages/ard/item/{display_id}',
 402             display_id, query=query, headers=headers)
 403
 404         # For user convenience we use the old contentId instead of the longer crid
 405         # Ref: https://github.com/yt-dlp/yt-dlp/issues/8731#issuecomment-1874398283
 406         old_id = traverse_obj(page_data, ('tracking', 'atiCustomVars', 'contentId', {int}))
 407         if old_id is not None:
 408             video_id = str(old_id)
 409             archive_ids = [make_archive_id(ARDBetaMediathekIE, display_id)]
 410         else:
 411             self.report_warning(f'Could not extract contentId{bug_reports_message()}')
 412             video_id = display_id
 413             archive_ids = None
 414
 415         player_data = traverse_obj(
 416             page_data, ('widgets', lambda _, v: v['type'] in ('player_ondemand', 'player_live'), {dict}), get_all=False)
 417         is_live = player_data.get('type') == 'player_live'
 418         media_data = traverse_obj(player_data, ('mediaCollection', 'embedded', {dict}))
 419
 420         if player_data.get('blockedByFsk'):
 421             self.raise_login_required('This video is only available for age verified users or after 22:00')
 422
 423         formats = []
 424         subtitles = {}
 425         for stream in traverse_obj(media_data, ('streams', ..., {dict})):
 426             kind = stream.get('kind')
 427             # Prioritize main stream over sign language and others
 428             preference = 1 if kind == 'main' else None
 429             for media in traverse_obj(stream, ('media', lambda _, v: url_or_none(v['url']))):
 430                 media_url = media['url']
 431
 432                 audio_kind = traverse_obj(media, (
 433                     'audios', 0, 'kind', {str}), default='').replace('standard', '')
 434                 lang_code = traverse_obj(media, ('audios', 0, 'languageCode', {str})) or 'deu'
 435                 lang = join_nonempty(lang_code, audio_kind)
 436                 language_preference = 10 if lang == 'deu' else -10
 437
 438                 if determine_ext(media_url) == 'm3u8':
 439                     fmts, subs = self._extract_m3u8_formats_and_subtitles(
 440                         media_url, video_id, m3u8_id=f'hls-{kind}', preference=preference, fatal=False, live=is_live)
 441                     for f in fmts:
 442                         f['language'] = lang
 443                         f['language_preference'] = language_preference
 444                     formats.extend(fmts)
 445                     self._merge_subtitles(subs, target=subtitles)
 446                 else:
 447                     formats.append({
 448                         'url': media_url,
 449                         'format_id': f'http-{kind}',
 450                         'preference': preference,
 451                         'language': lang,
 452                         'language_preference': language_preference,
 453                         **traverse_obj(media, {
 454                             'format_note': ('forcedLabel', {str}),
 455                             'width': ('maxHResolutionPx', {int_or_none}),
 456                             'height': ('maxVResolutionPx', {int_or_none}),
 457                             'vcodec': ('videoCodec', {str}),
 458                         }),
 459                     })
 460
 461         for sub in traverse_obj(media_data, ('subtitles', ..., {dict})):
 462             for sources in traverse_obj(sub, ('sources', lambda _, v: url_or_none(v['url']))):
 463                 subtitles.setdefault(sub.get('languageCode') or 'deu', []).append({
 464                     'url': sources['url'],
 465                     'ext': {'webvtt': 'vtt', 'ebutt': 'ttml'}.get(sources.get('kind')),
 466                 })
 467
 468         age_limit = traverse_obj(page_data, ('fskRating', {lambda x: remove_start(x, 'FSK')}, {int_or_none}))
 469         return {
 470             'id': video_id,
 471             'display_id': display_id,
 472             'formats': formats,
 473             'subtitles': subtitles,
 474             'is_live': is_live,
 475             'age_limit': age_limit,
 476             **traverse_obj(media_data, {
 477                 'chapters': ('pluginData', 'jumpmarks@all', 'chapterArray', lambda _, v: int_or_none(v['chapterTime']), {
 478                     'start_time': ('chapterTime', {int_or_none}),
 479                     'title': ('chapterTitle', {str}),
 480                 }),
 481             }),
 482             **traverse_obj(media_data, ('meta', {
 483                 'title': 'title',
 484                 'description': 'synopsis',
 485                 'timestamp': ('broadcastedOnDateTime', {parse_iso8601}),
 486                 'series': 'seriesTitle',
 487                 'thumbnail': ('images', 0, 'url', {url_or_none}),
 488                 'duration': ('durationSeconds', {int_or_none}),
 489                 'channel': 'clipSourceName',
 490             })),
 491             **self._extract_episode_info(page_data.get('title')),
 492             '_old_archive_ids': archive_ids,
 493         }
 494
 495
 496 class ARDMediathekCollectionIE(InfoExtractor):
 497     _VALID_URL = r'''(?x)https?://
 498         (?:(?:beta|www)\.)?ardmediathek\.de/
 499         (?:[^/?#]+/)?
 500         (?P<playlist>sendung|serie|sammlung)/
 501         (?:(?P<display_id>[^?#]+?)/)?
 502         (?P<id>[a-zA-Z0-9]+)
 503         (?:/(?P<season>\d+)(?:/(?P<version>OV|AD))?)?/?(?:[?#]|$)'''
 504     _GEO_COUNTRIES = ['DE']
 505
 506     _TESTS = [{
 507         'url': 'https://www.ardmediathek.de/serie/quiz/staffel-1-originalversion/Y3JpZDovL3dkci5kZS9vbmUvcXVpeg/1/OV',
 508         'info_dict': {
 509             'id': 'Y3JpZDovL3dkci5kZS9vbmUvcXVpeg_1_OV',
 510             'display_id': 'quiz/staffel-1-originalversion',
 511             'title': 'Staffel 1 Originalversion',
 512         },
 513         'playlist_count': 3,
 514     }, {
 515         'url': 'https://www.ardmediathek.de/serie/babylon-berlin/staffel-4-mit-audiodeskription/Y3JpZDovL2Rhc2Vyc3RlLmRlL2JhYnlsb24tYmVybGlu/4/AD',
 516         'info_dict': {
 517             'id': 'Y3JpZDovL2Rhc2Vyc3RlLmRlL2JhYnlsb24tYmVybGlu_4_AD',
 518             'display_id': 'babylon-berlin/staffel-4-mit-audiodeskription',
 519             'title': 'Staffel 4 mit Audiodeskription',
 520         },
 521         'playlist_count': 12,
 522     }, {
 523         'url': 'https://www.ardmediathek.de/serie/babylon-berlin/staffel-1/Y3JpZDovL2Rhc2Vyc3RlLmRlL2JhYnlsb24tYmVybGlu/1/',
 524         'info_dict': {
 525             'id': 'Y3JpZDovL2Rhc2Vyc3RlLmRlL2JhYnlsb24tYmVybGlu_1',
 526             'display_id': 'babylon-berlin/staffel-1',
 527             'title': 'Staffel 1',
 528         },
 529         'playlist_count': 8,
 530     }, {
 531         'url': 'https://www.ardmediathek.de/sendung/tatort/Y3JpZDovL2Rhc2Vyc3RlLmRlL3RhdG9ydA',
 532         'info_dict': {
 533             'id': 'Y3JpZDovL2Rhc2Vyc3RlLmRlL3RhdG9ydA',
 534             'display_id': 'tatort',
 535             'title': 'Tatort',
 536         },
 537         'playlist_mincount': 500,
 538     }, {
 539         'url': 'https://www.ardmediathek.de/sammlung/die-kirche-bleibt-im-dorf/5eOHzt8XB2sqeFXbIoJlg2',
 540         'info_dict': {
 541             'id': '5eOHzt8XB2sqeFXbIoJlg2',
 542             'display_id': 'die-kirche-bleibt-im-dorf',
 543             'title': 'Die Kirche bleibt im Dorf',
 544             'description': 'Die Kirche bleibt im Dorf',
 545         },
 546         'playlist_count': 4,
 547     }, {
 548         # playlist of type 'sendung'
 549         'url': 'https://www.ardmediathek.de/ard/sendung/doctor-who/Y3JpZDovL3dkci5kZS9vbmUvZG9jdG9yIHdobw/',
 550         'only_matching': True,
 551     }, {
 552         # playlist of type 'serie'
 553         'url': 'https://www.ardmediathek.de/serie/nachtstreife/staffel-1/Y3JpZDovL3N3ci5kZS9zZGIvc3RJZC8xMjQy/1',
 554         'only_matching': True,
 555     }, {
 556         # playlist of type 'sammlung'
 557         'url': 'https://www.ardmediathek.de/ard/sammlung/team-muenster/5JpTzLSbWUAK8184IOvEir/',
 558         'only_matching': True,
 559     }]
 560
 561     _PAGE_SIZE = 100
 562
 563     def _real_extract(self, url):
 564         playlist_id, display_id, playlist_type, season_number, version = self._match_valid_url(url).group(
 565             'id', 'display_id', 'playlist', 'season', 'version')
 566
 567         def call_api(page_num):
 568             api_path = 'compilations/ard' if playlist_type == 'sammlung' else 'widgets/ard/asset'
 569             return self._download_json(
 570                 f'https://api.ardmediathek.de/page-gateway/{api_path}/{playlist_id}', playlist_id,
 571                 f'Downloading playlist page {page_num}', query={
 572                     'pageNumber': page_num,
 573                     'pageSize': self._PAGE_SIZE,
 574                     **({
 575                         'seasoned': 'true',
 576                         'seasonNumber': season_number,
 577                         'withOriginalversion': 'true' if version == 'OV' else 'false',
 578                         'withAudiodescription': 'true' if version == 'AD' else 'false',
 579                     } if season_number else {}),
 580                 })
 581
 582         def fetch_page(page_num):
 583             for item in traverse_obj(call_api(page_num), ('teasers', ..., {dict})):
 584                 item_id = traverse_obj(item, ('links', 'target', ('urlId', 'id')), 'id', get_all=False)
 585                 if not item_id or item_id == playlist_id:
 586                     continue
 587                 item_mode = 'sammlung' if item.get('type') == 'compilation' else 'video'
 588                 yield self.url_result(
 589                     f'https://www.ardmediathek.de/{item_mode}/{item_id}',
 590                     ie=(ARDMediathekCollectionIE if item_mode == 'sammlung' else ARDBetaMediathekIE),
 591                     **traverse_obj(item, {
 592                         'id': ('id', {str}),
 593                         'title': ('longTitle', {str}),
 594                         'duration': ('duration', {int_or_none}),
 595                         'timestamp': ('broadcastedOn', {parse_iso8601}),
 596                     }))
 597
 598         page_data = call_api(0)
 599         full_id = join_nonempty(playlist_id, season_number, version, delim='_')
 600
 601         return self.playlist_result(
 602             OnDemandPagedList(fetch_page, self._PAGE_SIZE), full_id, display_id=display_id,
 603             title=page_data.get('title'), description=page_data.get('synopsis'))