yt_dlp/extractor/ard.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import json
   5 import re
   6
   7 from .common import InfoExtractor
   8 from .generic import GenericIE
   9 from ..utils import (
  10     determine_ext,
  11     ExtractorError,
  12     int_or_none,
  13     parse_duration,
  14     qualities,
  15     str_or_none,
  16     try_get,
  17     unified_strdate,
  18     unified_timestamp,
  19     update_url_query,
  20     url_or_none,
  21     xpath_text,
  22 )
  23 from ..compat import compat_etree_fromstring
  24
  25
  26 class ARDMediathekBaseIE(InfoExtractor):
  27     _GEO_COUNTRIES = ['DE']
  28
  29     def _extract_media_info(self, media_info_url, webpage, video_id):
  30         media_info = self._download_json(
  31             media_info_url, video_id, 'Downloading media JSON')
  32         return self._parse_media_info(media_info, video_id, '"fsk"' in webpage)
  33
  34     def _parse_media_info(self, media_info, video_id, fsk):
  35         formats = self._extract_formats(media_info, video_id)
  36
  37         if not formats:
  38             if fsk:
  39                 raise ExtractorError(
  40                     'This video is only available after 20:00', expected=True)
  41             elif media_info.get('_geoblocked'):
  42                 self.raise_geo_restricted(
  43                     'This video is not available due to geoblocking',
  44                     countries=self._GEO_COUNTRIES)
  45
  46         self._sort_formats(formats)
  47
  48         subtitles = {}
  49         subtitle_url = media_info.get('_subtitleUrl')
  50         if subtitle_url:
  51             subtitles['de'] = [{
  52                 'ext': 'ttml',
  53                 'url': subtitle_url,
  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 _ARD_extract_episode_info(self, title):
  66         """Try to extract season/episode data from the title."""
  67         res = {}
  68         if not title:
  69             return res
  70
  71         for pattern in [
  72             # Pattern for title like "Homo sapiens (S06/E07) - Originalversion"
  73             # from: https://www.ardmediathek.de/one/sendung/doctor-who/Y3JpZDovL3dkci5kZS9vbmUvZG9jdG9yIHdobw
  74             r'.*(?P<ep_info> \(S(?P<season_number>\d+)/E(?P<episode_number>\d+)\)).*',
  75             # E.g.: title="Fritjof aus Norwegen (2) (AD)"
  76             # from: https://www.ardmediathek.de/ard/sammlung/der-krieg-und-ich/68cMkqJdllm639Skj4c7sS/
  77             r'.*(?P<ep_info> \((?:Folge |Teil )?(?P<episode_number>\d+)(?:/\d+)?\)).*',
  78             r'.*(?P<ep_info>Folge (?P<episode_number>\d+)(?:\:| -|) )\"(?P<episode>.+)\".*',
  79             # E.g.: title="Folge 25/42: Symmetrie"
  80             # from: https://www.ardmediathek.de/ard/video/grips-mathe/folge-25-42-symmetrie/ard-alpha/Y3JpZDovL2JyLmRlL3ZpZGVvLzMyYzI0ZjczLWQ1N2MtNDAxNC05ZmZhLTFjYzRkZDA5NDU5OQ/
  81             # E.g.: title="Folge 1063 - Vertrauen"
  82             # from: https://www.ardmediathek.de/ard/sendung/die-fallers/Y3JpZDovL3N3ci5kZS8yMzAyMDQ4/
  83             r'.*(?P<ep_info>Folge (?P<episode_number>\d+)(?:/\d+)?(?:\:| -|) ).*',
  84         ]:
  85             m = re.match(pattern, title)
  86             if m:
  87                 groupdict = m.groupdict()
  88                 res['season_number'] = int_or_none(groupdict.get('season_number'))
  89                 res['episode_number'] = int_or_none(groupdict.get('episode_number'))
  90                 res['episode'] = str_or_none(groupdict.get('episode'))
  91                 # Build the episode title by removing numeric episode information:
  92                 if groupdict.get('ep_info') and not res['episode']:
  93                     res['episode'] = str_or_none(
  94                         title.replace(groupdict.get('ep_info'), ''))
  95                 if res['episode']:
  96                     res['episode'] = res['episode'].strip()
  97                 break
  98
  99         # As a fallback use the whole title as the episode name:
 100         if not res.get('episode'):
 101             res['episode'] = title.strip()
 102         return res
 103
 104     def _extract_formats(self, media_info, video_id):
 105         type_ = media_info.get('_type')
 106         media_array = media_info.get('_mediaArray', [])
 107         formats = []
 108         for num, media in enumerate(media_array):
 109             for stream in media.get('_mediaStreamArray', []):
 110                 stream_urls = stream.get('_stream')
 111                 if not stream_urls:
 112                     continue
 113                 if not isinstance(stream_urls, list):
 114                     stream_urls = [stream_urls]
 115                 quality = stream.get('_quality')
 116                 server = stream.get('_server')
 117                 for stream_url in stream_urls:
 118                     if not url_or_none(stream_url):
 119                         continue
 120                     ext = determine_ext(stream_url)
 121                     if quality != 'auto' and ext in ('f4m', 'm3u8'):
 122                         continue
 123                     if ext == 'f4m':
 124                         formats.extend(self._extract_f4m_formats(
 125                             update_url_query(stream_url, {
 126                                 'hdcore': '3.1.1',
 127                                 'plugin': 'aasp-3.1.1.69.124'
 128                             }), video_id, f4m_id='hds', fatal=False))
 129                     elif ext == 'm3u8':
 130                         formats.extend(self._extract_m3u8_formats(
 131                             stream_url, video_id, 'mp4', 'm3u8_native',
 132                             m3u8_id='hls', fatal=False))
 133                     else:
 134                         if server and server.startswith('rtmp'):
 135                             f = {
 136                                 'url': server,
 137                                 'play_path': stream_url,
 138                                 'format_id': 'a%s-rtmp-%s' % (num, quality),
 139                             }
 140                         else:
 141                             f = {
 142                                 'url': stream_url,
 143                                 'format_id': 'a%s-%s-%s' % (num, ext, quality)
 144                             }
 145                         m = re.search(
 146                             r'_(?P<width>\d+)x(?P<height>\d+)\.mp4$',
 147                             stream_url)
 148                         if m:
 149                             f.update({
 150                                 'width': int(m.group('width')),
 151                                 'height': int(m.group('height')),
 152                             })
 153                         if type_ == 'audio':
 154                             f['vcodec'] = 'none'
 155                         formats.append(f)
 156         return formats
 157
 158
 159 class ARDMediathekIE(ARDMediathekBaseIE):
 160     IE_NAME = 'ARD:mediathek'
 161     _VALID_URL = r'^https?://(?:(?:(?:www|classic)\.)?ardmediathek\.de|mediathek\.(?:daserste|rbb-online)\.de|one\.ard\.de)/(?:.*/)(?P<video_id>[0-9]+|[^0-9][^/\?]+)[^/\?]*(?:\?.*)?'
 162
 163     _TESTS = [{
 164         # available till 26.07.2022
 165         'url': 'http://www.ardmediathek.de/tv/S%C3%9CDLICHT/Was-ist-die-Kunst-der-Zukunft-liebe-Ann/BR-Fernsehen/Video?bcastId=34633636&documentId=44726822',
 166         'info_dict': {
 167             'id': '44726822',
 168             'ext': 'mp4',
 169             'title': 'Was ist die Kunst der Zukunft, liebe Anna McCarthy?',
 170             'description': 'md5:4ada28b3e3b5df01647310e41f3a62f5',
 171             'duration': 1740,
 172         },
 173         'params': {
 174             # m3u8 download
 175             'skip_download': True,
 176         }
 177     }, {
 178         'url': 'https://one.ard.de/tv/Mord-mit-Aussicht/Mord-mit-Aussicht-6-39-T%C3%B6dliche-Nach/ONE/Video?bcastId=46384294&documentId=55586872',
 179         'only_matching': True,
 180     }, {
 181         # audio
 182         'url': 'http://www.ardmediathek.de/tv/WDR-H%C3%B6rspiel-Speicher/Tod-eines-Fu%C3%9Fballers/WDR-3/Audio-Podcast?documentId=28488308&bcastId=23074086',
 183         'only_matching': True,
 184     }, {
 185         'url': 'http://mediathek.daserste.de/sendungen_a-z/328454_anne-will/22429276_vertrauen-ist-gut-spionieren-ist-besser-geht',
 186         'only_matching': True,
 187     }, {
 188         # audio
 189         'url': 'http://mediathek.rbb-online.de/radio/Hörspiel/Vor-dem-Fest/kulturradio/Audio?documentId=30796318&topRessort=radio&bcastId=9839158',
 190         'only_matching': True,
 191     }, {
 192         'url': 'https://classic.ardmediathek.de/tv/Panda-Gorilla-Co/Panda-Gorilla-Co-Folge-274/Das-Erste/Video?bcastId=16355486&documentId=58234698',
 193         'only_matching': True,
 194     }]
 195
 196     @classmethod
 197     def suitable(cls, url):
 198         return False if ARDBetaMediathekIE.suitable(url) else super(ARDMediathekIE, cls).suitable(url)
 199
 200     def _real_extract(self, url):
 201         # determine video id from url
 202         m = re.match(self._VALID_URL, url)
 203
 204         document_id = None
 205
 206         numid = re.search(r'documentId=([0-9]+)', url)
 207         if numid:
 208             document_id = video_id = numid.group(1)
 209         else:
 210             video_id = m.group('video_id')
 211
 212         webpage = self._download_webpage(url, video_id)
 213
 214         ERRORS = (
 215             ('>Leider liegt eine Störung vor.', 'Video %s is unavailable'),
 216             ('>Der gewünschte Beitrag ist nicht mehr verfügbar.<',
 217              'Video %s is no longer available'),
 218         )
 219
 220         for pattern, message in ERRORS:
 221             if pattern in webpage:
 222                 raise ExtractorError(message % video_id, expected=True)
 223
 224         if re.search(r'[\?&]rss($|[=&])', url):
 225             doc = compat_etree_fromstring(webpage.encode('utf-8'))
 226             if doc.tag == 'rss':
 227                 return GenericIE()._extract_rss(url, video_id, doc)
 228
 229         title = self._og_search_title(webpage, default=None) or self._html_search_regex(
 230             [r'<h1(?:\s+class="boxTopHeadline")?>(.*?)</h1>',
 231              r'<meta name="dcterms\.title" content="(.*?)"/>',
 232              r'<h4 class="headline">(.*?)</h4>',
 233              r'<title[^>]*>(.*?)</title>'],
 234             webpage, 'title')
 235         description = self._og_search_description(webpage, default=None) or self._html_search_meta(
 236             'dcterms.abstract', webpage, 'description', default=None)
 237         if description is None:
 238             description = self._html_search_meta(
 239                 'description', webpage, 'meta description', default=None)
 240         if description is None:
 241             description = self._html_search_regex(
 242                 r'<p\s+class="teasertext">(.+?)</p>',
 243                 webpage, 'teaser text', default=None)
 244
 245         # Thumbnail is sometimes not present.
 246         # It is in the mobile version, but that seems to use a different URL
 247         # structure altogether.
 248         thumbnail = self._og_search_thumbnail(webpage, default=None)
 249
 250         media_streams = re.findall(r'''(?x)
 251             mediaCollection\.addMediaStream\([0-9]+,\s*[0-9]+,\s*"[^"]*",\s*
 252             "([^"]+)"''', webpage)
 253
 254         if media_streams:
 255             QUALITIES = qualities(['lo', 'hi', 'hq'])
 256             formats = []
 257             for furl in set(media_streams):
 258                 if furl.endswith('.f4m'):
 259                     fid = 'f4m'
 260                 else:
 261                     fid_m = re.match(r'.*\.([^.]+)\.[^.]+$', furl)
 262                     fid = fid_m.group(1) if fid_m else None
 263                 formats.append({
 264                     'quality': QUALITIES(fid),
 265                     'format_id': fid,
 266                     'url': furl,
 267                 })
 268             self._sort_formats(formats)
 269             info = {
 270                 'formats': formats,
 271             }
 272         else:  # request JSON file
 273             if not document_id:
 274                 video_id = self._search_regex(
 275                     r'/play/(?:config|media)/(\d+)', webpage, 'media id')
 276             info = self._extract_media_info(
 277                 'http://www.ardmediathek.de/play/media/%s' % video_id,
 278                 webpage, video_id)
 279
 280         info.update({
 281             'id': video_id,
 282             'title': self._live_title(title) if info.get('is_live') else title,
 283             'description': description,
 284             'thumbnail': thumbnail,
 285         })
 286         info.update(self._ARD_extract_episode_info(info['title']))
 287
 288         return info
 289
 290
 291 class ARDIE(InfoExtractor):
 292     _VALID_URL = r'(?P<mainurl>https?://(?:www\.)?daserste\.de/[^?#]+/videos(?:extern)?/(?P<display_id>[^/?#]+)-(?:video-?)?(?P<id>[0-9]+))\.html'
 293     _TESTS = [{
 294         # available till 7.01.2022
 295         'url': 'https://www.daserste.de/information/talk/maischberger/videos/maischberger-die-woche-video100.html',
 296         'md5': '867d8aa39eeaf6d76407c5ad1bb0d4c1',
 297         'info_dict': {
 298             'display_id': 'maischberger-die-woche',
 299             'id': '100',
 300             'ext': 'mp4',
 301             'duration': 3687.0,
 302             'title': 'maischberger. die woche vom 7. Januar 2021',
 303             'upload_date': '20210107',
 304             'thumbnail': r're:^https?://.*\.jpg$',
 305         },
 306     }, {
 307         'url': 'https://www.daserste.de/information/reportage-dokumentation/erlebnis-erde/videosextern/woelfe-und-herdenschutzhunde-ungleiche-brueder-102.html',
 308         'only_matching': True,
 309     }, {
 310         'url': 'http://www.daserste.de/information/reportage-dokumentation/dokus/videos/die-story-im-ersten-mission-unter-falscher-flagge-100.html',
 311         'only_matching': True,
 312     }]
 313
 314     def _real_extract(self, url):
 315         mobj = re.match(self._VALID_URL, url)
 316         display_id = mobj.group('display_id')
 317
 318         player_url = mobj.group('mainurl') + '~playerXml.xml'
 319         doc = self._download_xml(player_url, display_id)
 320         video_node = doc.find('./video')
 321         upload_date = unified_strdate(xpath_text(
 322             video_node, './broadcastDate'))
 323         thumbnail = xpath_text(video_node, './/teaserImage//variant/url')
 324
 325         formats = []
 326         for a in video_node.findall('.//asset'):
 327             file_name = xpath_text(a, './fileName', default=None)
 328             if not file_name:
 329                 continue
 330             format_type = a.attrib.get('type')
 331             format_url = url_or_none(file_name)
 332             if format_url:
 333                 ext = determine_ext(file_name)
 334                 if ext == 'm3u8':
 335                     formats.extend(self._extract_m3u8_formats(
 336                         format_url, display_id, 'mp4', entry_protocol='m3u8_native',
 337                         m3u8_id=format_type or 'hls', fatal=False))
 338                     continue
 339                 elif ext == 'f4m':
 340                     formats.extend(self._extract_f4m_formats(
 341                         update_url_query(format_url, {'hdcore': '3.7.0'}),
 342                         display_id, f4m_id=format_type or 'hds', fatal=False))
 343                     continue
 344             f = {
 345                 'format_id': format_type,
 346                 'width': int_or_none(xpath_text(a, './frameWidth')),
 347                 'height': int_or_none(xpath_text(a, './frameHeight')),
 348                 'vbr': int_or_none(xpath_text(a, './bitrateVideo')),
 349                 'abr': int_or_none(xpath_text(a, './bitrateAudio')),
 350                 'vcodec': xpath_text(a, './codecVideo'),
 351                 'tbr': int_or_none(xpath_text(a, './totalBitrate')),
 352             }
 353             server_prefix = xpath_text(a, './serverPrefix', default=None)
 354             if server_prefix:
 355                 f.update({
 356                     'url': server_prefix,
 357                     'playpath': file_name,
 358                 })
 359             else:
 360                 if not format_url:
 361                     continue
 362                 f['url'] = format_url
 363             formats.append(f)
 364         self._sort_formats(formats)
 365
 366         return {
 367             'id': mobj.group('id'),
 368             'formats': formats,
 369             'display_id': display_id,
 370             'title': video_node.find('./title').text,
 371             'duration': parse_duration(video_node.find('./duration').text),
 372             'upload_date': upload_date,
 373             'thumbnail': thumbnail,
 374         }
 375
 376
 377 class ARDBetaMediathekIE(ARDMediathekBaseIE):
 378     _VALID_URL = r'https://(?:(?:beta|www)\.)?ardmediathek\.de/(?P<client>[^/]+)/(?P<mode>player|live|video|sendung|sammlung)/(?P<display_id>(?:[^/]+/)*)(?P<video_id>[a-zA-Z0-9]+)'
 379     _TESTS = [{
 380         'url': 'https://www.ardmediathek.de/mdr/video/die-robuste-roswita/Y3JpZDovL21kci5kZS9iZWl0cmFnL2Ntcy84MWMxN2MzZC0wMjkxLTRmMzUtODk4ZS0wYzhlOWQxODE2NGI/',
 381         'md5': 'a1dc75a39c61601b980648f7c9f9f71d',
 382         'info_dict': {
 383             'display_id': 'die-robuste-roswita',
 384             'id': '78566716',
 385             'title': 'Die robuste Roswita',
 386             'description': r're:^Der Mord.*totgeglaubte Ehefrau Roswita',
 387             'duration': 5316,
 388             'thumbnail': 'https://img.ardmediathek.de/standard/00/78/56/67/84/575672121/16x9/960?mandant=ard',
 389             'timestamp': 1596658200,
 390             'upload_date': '20200805',
 391             'ext': 'mp4',
 392         },
 393     }, {
 394         'url': 'https://beta.ardmediathek.de/ard/video/Y3JpZDovL2Rhc2Vyc3RlLmRlL3RhdG9ydC9mYmM4NGM1NC0xNzU4LTRmZGYtYWFhZS0wYzcyZTIxNGEyMDE',
 395         'only_matching': True,
 396     }, {
 397         'url': 'https://ardmediathek.de/ard/video/saartalk/saartalk-gesellschaftsgift-haltung-gegen-hass/sr-fernsehen/Y3JpZDovL3NyLW9ubGluZS5kZS9TVF84MTY4MA/',
 398         'only_matching': True,
 399     }, {
 400         'url': 'https://www.ardmediathek.de/ard/video/trailer/private-eyes-s01-e01/one/Y3JpZDovL3dkci5kZS9CZWl0cmFnLTE1MTgwYzczLWNiMTEtNGNkMS1iMjUyLTg5MGYzOWQxZmQ1YQ/',
 401         'only_matching': True,
 402     }, {
 403         'url': 'https://www.ardmediathek.de/ard/player/Y3JpZDovL3N3ci5kZS9hZXgvbzEwNzE5MTU/',
 404         'only_matching': True,
 405     }, {
 406         'url': 'https://www.ardmediathek.de/swr/live/Y3JpZDovL3N3ci5kZS8xMzQ4MTA0Mg',
 407         'only_matching': True,
 408     }, {
 409         # playlist of type 'sendung'
 410         'url': 'https://www.ardmediathek.de/ard/sendung/doctor-who/Y3JpZDovL3dkci5kZS9vbmUvZG9jdG9yIHdobw/',
 411         'only_matching': True,
 412     }, {
 413         # playlist of type 'sammlung'
 414         'url': 'https://www.ardmediathek.de/ard/sammlung/team-muenster/5JpTzLSbWUAK8184IOvEir/',
 415         'only_matching': True,
 416     }]
 417
 418     def _ARD_load_playlist_snipped(self, playlist_id, display_id, client, mode, pageNumber):
 419         """ Query the ARD server for playlist information
 420         and returns the data in "raw" format """
 421         if mode == 'sendung':
 422             graphQL = json.dumps({
 423                 'query': '''{
 424                     showPage(
 425                         client: "%s"
 426                         showId: "%s"
 427                         pageNumber: %d
 428                     ) {
 429                         pagination {
 430                             pageSize
 431                             totalElements
 432                         }
 433                         teasers {        # Array
 434                             mediumTitle
 435                             links { target { id href title } }
 436                             type
 437                         }
 438                     }}''' % (client, playlist_id, pageNumber),
 439             }).encode()
 440         else:  # mode == 'sammlung'
 441             graphQL = json.dumps({
 442                 'query': '''{
 443                     morePage(
 444                         client: "%s"
 445                         compilationId: "%s"
 446                         pageNumber: %d
 447                     ) {
 448                         widget {
 449                             pagination {
 450                                 pageSize
 451                                 totalElements
 452                             }
 453                             teasers {        # Array
 454                                 mediumTitle
 455                                 links { target { id href title } }
 456                                 type
 457                             }
 458                         }
 459                     }}''' % (client, playlist_id, pageNumber),
 460             }).encode()
 461         # Ressources for ARD graphQL debugging:
 462         # https://api-test.ardmediathek.de/public-gateway
 463         show_page = self._download_json(
 464             'https://api.ardmediathek.de/public-gateway',
 465             '[Playlist] %s' % display_id,
 466             data=graphQL,
 467             headers={'Content-Type': 'application/json'})['data']
 468         # align the structure of the returned data:
 469         if mode == 'sendung':
 470             show_page = show_page['showPage']
 471         else:  # mode == 'sammlung'
 472             show_page = show_page['morePage']['widget']
 473         return show_page
 474
 475     def _ARD_extract_playlist(self, url, playlist_id, display_id, client, mode):
 476         """ Collects all playlist entries and returns them as info dict.
 477         Supports playlists of mode 'sendung' and 'sammlung', and also nested
 478         playlists. """
 479         entries = []
 480         pageNumber = 0
 481         while True:  # iterate by pageNumber
 482             show_page = self._ARD_load_playlist_snipped(
 483                 playlist_id, display_id, client, mode, pageNumber)
 484             for teaser in show_page['teasers']:  # process playlist items
 485                 if '/compilation/' in teaser['links']['target']['href']:
 486                     # alternativ cond.: teaser['type'] == "compilation"
 487                     # => This is an nested compilation, e.g. like:
 488                     # https://www.ardmediathek.de/ard/sammlung/die-kirche-bleibt-im-dorf/5eOHzt8XB2sqeFXbIoJlg2/
 489                     link_mode = 'sammlung'
 490                 else:
 491                     link_mode = 'video'
 492
 493                 item_url = 'https://www.ardmediathek.de/%s/%s/%s/%s/%s' % (
 494                     client, link_mode, display_id,
 495                     # perform HTLM quoting of episode title similar to ARD:
 496                     re.sub('^-|-$', '',  # remove '-' from begin/end
 497                            re.sub('[^a-zA-Z0-9]+', '-',  # replace special chars by -
 498                                   teaser['links']['target']['title'].lower()
 499                                   .replace('ä', 'ae').replace('ö', 'oe')
 500                                   .replace('ü', 'ue').replace('ß', 'ss'))),
 501                     teaser['links']['target']['id'])
 502                 entries.append(self.url_result(
 503                     item_url,
 504                     ie=ARDBetaMediathekIE.ie_key()))
 505
 506             if (show_page['pagination']['pageSize'] * (pageNumber + 1)
 507                >= show_page['pagination']['totalElements']):
 508                 # we've processed enough pages to get all playlist entries
 509                 break
 510             pageNumber = pageNumber + 1
 511
 512         return self.playlist_result(entries, playlist_title=display_id)
 513
 514     def _real_extract(self, url):
 515         mobj = re.match(self._VALID_URL, url)
 516         video_id = mobj.group('video_id')
 517         display_id = mobj.group('display_id')
 518         if display_id:
 519             display_id = display_id.rstrip('/')
 520         if not display_id:
 521             display_id = video_id
 522
 523         if mobj.group('mode') in ('sendung', 'sammlung'):
 524             # this is a playlist-URL
 525             return self._ARD_extract_playlist(
 526                 url, video_id, display_id,
 527                 mobj.group('client'),
 528                 mobj.group('mode'))
 529
 530         player_page = self._download_json(
 531             'https://api.ardmediathek.de/public-gateway',
 532             display_id, data=json.dumps({
 533                 'query': '''{
 534   playerPage(client:"%s", clipId: "%s") {
 535     blockedByFsk
 536     broadcastedOn
 537     maturityContentRating
 538     mediaCollection {
 539       _duration
 540       _geoblocked
 541       _isLive
 542       _mediaArray {
 543         _mediaStreamArray {
 544           _quality
 545           _server
 546           _stream
 547         }
 548       }
 549       _previewImage
 550       _subtitleUrl
 551       _type
 552     }
 553     show {
 554       title
 555     }
 556     synopsis
 557     title
 558     tracking {
 559       atiCustomVars {
 560         contentId
 561       }
 562     }
 563   }
 564 }''' % (mobj.group('client'), video_id),
 565             }).encode(), headers={
 566                 'Content-Type': 'application/json'
 567             })['data']['playerPage']
 568         title = player_page['title']
 569         content_id = str_or_none(try_get(
 570             player_page, lambda x: x['tracking']['atiCustomVars']['contentId']))
 571         media_collection = player_page.get('mediaCollection') or {}
 572         if not media_collection and content_id:
 573             media_collection = self._download_json(
 574                 'https://www.ardmediathek.de/play/media/' + content_id,
 575                 content_id, fatal=False) or {}
 576         info = self._parse_media_info(
 577             media_collection, content_id or video_id,
 578             player_page.get('blockedByFsk'))
 579         age_limit = None
 580         description = player_page.get('synopsis')
 581         maturity_content_rating = player_page.get('maturityContentRating')
 582         if maturity_content_rating:
 583             age_limit = int_or_none(maturity_content_rating.lstrip('FSK'))
 584         if not age_limit and description:
 585             age_limit = int_or_none(self._search_regex(
 586                 r'\(FSK\s*(\d+)\)\s*$', description, 'age limit', default=None))
 587         info.update({
 588             'age_limit': age_limit,
 589             'display_id': display_id,
 590             'title': title,
 591             'description': description,
 592             'timestamp': unified_timestamp(player_page.get('broadcastedOn')),
 593             'series': try_get(player_page, lambda x: x['show']['title']),
 594         })
 595         info.update(self._ARD_extract_episode_info(info['title']))
 596         return info