yt_dlp/extractor/prosiebensat1.py

   1 import hashlib
   2 import re
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     ExtractorError,
   7     determine_ext,
   8     float_or_none,
   9     int_or_none,
  10     merge_dicts,
  11     unified_strdate,
  12 )
  13
  14
  15 class ProSiebenSat1BaseIE(InfoExtractor):
  16     _GEO_BYPASS = False
  17     _ACCESS_ID = None
  18     _SUPPORTED_PROTOCOLS = 'dash:clear,hls:clear,progressive:clear'
  19     _V4_BASE_URL = 'https://vas-v4.p7s1video.net/4.0/get'
  20
  21     def _extract_video_info(self, url, clip_id):
  22         client_location = url
  23
  24         video = self._download_json(
  25             'http://vas.sim-technik.de/vas/live/v2/videos',
  26             clip_id, 'Downloading videos JSON', query={
  27                 'access_token': self._TOKEN,
  28                 'client_location': client_location,
  29                 'client_name': self._CLIENT_NAME,
  30                 'ids': clip_id,
  31             })[0]
  32
  33         if not self.get_param('allow_unplayable_formats') and video.get('is_protected') is True:
  34             self.report_drm(clip_id)
  35
  36         formats = []
  37         if self._ACCESS_ID:
  38             raw_ct = self._ENCRYPTION_KEY + clip_id + self._IV + self._ACCESS_ID
  39             protocols = self._download_json(
  40                 self._V4_BASE_URL + 'protocols', clip_id,
  41                 'Downloading protocols JSON',
  42                 headers=self.geo_verification_headers(), query={
  43                     'access_id': self._ACCESS_ID,
  44                     'client_token': hashlib.sha1((raw_ct).encode()).hexdigest(),
  45                     'video_id': clip_id,
  46                 }, fatal=False, expected_status=(403,)) or {}
  47             error = protocols.get('error') or {}
  48             if error.get('title') == 'Geo check failed':
  49                 self.raise_geo_restricted(countries=['AT', 'CH', 'DE'])
  50             server_token = protocols.get('server_token')
  51             if server_token:
  52                 urls = (self._download_json(
  53                     self._V4_BASE_URL + 'urls', clip_id, 'Downloading urls JSON', query={
  54                         'access_id': self._ACCESS_ID,
  55                         'client_token': hashlib.sha1((raw_ct + server_token + self._SUPPORTED_PROTOCOLS).encode()).hexdigest(),
  56                         'protocols': self._SUPPORTED_PROTOCOLS,
  57                         'server_token': server_token,
  58                         'video_id': clip_id,
  59                     }, fatal=False) or {}).get('urls') or {}
  60                 for protocol, variant in urls.items():
  61                     source_url = variant.get('clear', {}).get('url')
  62                     if not source_url:
  63                         continue
  64                     if protocol == 'dash':
  65                         formats.extend(self._extract_mpd_formats(
  66                             source_url, clip_id, mpd_id=protocol, fatal=False))
  67                     elif protocol == 'hls':
  68                         formats.extend(self._extract_m3u8_formats(
  69                             source_url, clip_id, 'mp4', 'm3u8_native',
  70                             m3u8_id=protocol, fatal=False))
  71                     else:
  72                         formats.append({
  73                             'url': source_url,
  74                             'format_id': protocol,
  75                         })
  76         if not formats:
  77             source_ids = [str(source['id']) for source in video['sources']]
  78
  79             client_id = self._SALT[:2] + hashlib.sha1(''.join([clip_id, self._SALT, self._TOKEN, client_location, self._SALT, self._CLIENT_NAME]).encode()).hexdigest()
  80
  81             sources = self._download_json(
  82                 f'http://vas.sim-technik.de/vas/live/v2/videos/{clip_id}/sources',
  83                 clip_id, 'Downloading sources JSON', query={
  84                     'access_token': self._TOKEN,
  85                     'client_id': client_id,
  86                     'client_location': client_location,
  87                     'client_name': self._CLIENT_NAME,
  88                 })
  89             server_id = sources['server_id']
  90
  91             def fix_bitrate(bitrate):
  92                 bitrate = int_or_none(bitrate)
  93                 if not bitrate:
  94                     return None
  95                 return (bitrate // 1000) if bitrate % 1000 == 0 else bitrate
  96
  97             for source_id in source_ids:
  98                 client_id = self._SALT[:2] + hashlib.sha1(''.join([self._SALT, clip_id, self._TOKEN, server_id, client_location, source_id, self._SALT, self._CLIENT_NAME]).encode()).hexdigest()
  99                 urls = self._download_json(
 100                     f'http://vas.sim-technik.de/vas/live/v2/videos/{clip_id}/sources/url',
 101                     clip_id, 'Downloading urls JSON', fatal=False, query={
 102                         'access_token': self._TOKEN,
 103                         'client_id': client_id,
 104                         'client_location': client_location,
 105                         'client_name': self._CLIENT_NAME,
 106                         'server_id': server_id,
 107                         'source_ids': source_id,
 108                     })
 109                 if not urls:
 110                     continue
 111                 if urls.get('status_code') != 0:
 112                     raise ExtractorError('This video is unavailable', expected=True)
 113                 urls_sources = urls['sources']
 114                 if isinstance(urls_sources, dict):
 115                     urls_sources = urls_sources.values()
 116                 for source in urls_sources:
 117                     source_url = source.get('url')
 118                     if not source_url:
 119                         continue
 120                     protocol = source.get('protocol')
 121                     mimetype = source.get('mimetype')
 122                     if mimetype == 'application/f4m+xml' or 'f4mgenerator' in source_url or determine_ext(source_url) == 'f4m':
 123                         formats.extend(self._extract_f4m_formats(
 124                             source_url, clip_id, f4m_id='hds', fatal=False))
 125                     elif mimetype == 'application/x-mpegURL':
 126                         formats.extend(self._extract_m3u8_formats(
 127                             source_url, clip_id, 'mp4', 'm3u8_native',
 128                             m3u8_id='hls', fatal=False))
 129                     elif mimetype == 'application/dash+xml':
 130                         formats.extend(self._extract_mpd_formats(
 131                             source_url, clip_id, mpd_id='dash', fatal=False))
 132                     else:
 133                         tbr = fix_bitrate(source['bitrate'])
 134                         if protocol in ('rtmp', 'rtmpe'):
 135                             mobj = re.search(r'^(?P<url>rtmpe?://[^/]+)/(?P<path>.+)$', source_url)
 136                             if not mobj:
 137                                 continue
 138                             path = mobj.group('path')
 139                             mp4colon_index = path.rfind('mp4:')
 140                             app = path[:mp4colon_index]
 141                             play_path = path[mp4colon_index:]
 142                             formats.append({
 143                                 'url': '{}/{}'.format(mobj.group('url'), app),
 144                                 'app': app,
 145                                 'play_path': play_path,
 146                                 'player_url': 'http://livepassdl.conviva.com/hf/ver/2.79.0.17083/LivePassModuleMain.swf',
 147                                 'page_url': 'http://www.prosieben.de',
 148                                 'tbr': tbr,
 149                                 'ext': 'flv',
 150                                 'format_id': 'rtmp{}'.format(f'-{tbr}' if tbr else ''),
 151                             })
 152                         else:
 153                             formats.append({
 154                                 'url': source_url,
 155                                 'tbr': tbr,
 156                                 'format_id': 'http{}'.format(f'-{tbr}' if tbr else ''),
 157                             })
 158
 159         return {
 160             'duration': float_or_none(video.get('duration')),
 161             'formats': formats,
 162         }
 163
 164
 165 class ProSiebenSat1IE(ProSiebenSat1BaseIE):
 166     IE_NAME = 'prosiebensat1'
 167     IE_DESC = 'ProSiebenSat.1 Digital'
 168     _VALID_URL = r'''(?x)
 169                     https?://
 170                         (?:www\.)?
 171                         (?:
 172                             (?:beta\.)?
 173                             (?:
 174                                 prosieben(?:maxx)?|sixx|sat1(?:gold)?|kabeleins(?:doku)?|the-voice-of-germany|advopedia
 175                             )\.(?:de|at|ch)|
 176                             ran\.de|fem\.com|advopedia\.de|galileo\.tv/video
 177                         )
 178                         /(?P<id>.+)
 179                     '''
 180
 181     _TESTS = [
 182         {
 183             # Tests changes introduced in https://github.com/ytdl-org/youtube-dl/pull/6242
 184             # in response to fixing https://github.com/ytdl-org/youtube-dl/issues/6215:
 185             # - malformed f4m manifest support
 186             # - proper handling of URLs starting with `https?://` in 2.0 manifests
 187             # - recursive child f4m manifests extraction
 188             'url': 'http://www.prosieben.de/tv/circus-halligalli/videos/218-staffel-2-episode-18-jahresrueckblick-ganze-folge',
 189             'info_dict': {
 190                 'id': '2104602',
 191                 'ext': 'mp4',
 192                 'title': 'CIRCUS HALLIGALLI - Episode 18 - Staffel 2',
 193                 'description': 'md5:8733c81b702ea472e069bc48bb658fc1',
 194                 'upload_date': '20131231',
 195                 'duration': 5845.04,
 196                 'series': 'CIRCUS HALLIGALLI',
 197                 'season_number': 2,
 198                 'episode': 'Episode 18 - Staffel 2',
 199                 'episode_number': 18,
 200             },
 201         },
 202         {
 203             'url': 'http://www.prosieben.de/videokatalog/Gesellschaft/Leben/Trends/video-Lady-Umstyling-f%C3%BCr-Audrina-Rebekka-Audrina-Fergen-billig-aussehen-Battal-Modica-700544.html',
 204             'info_dict': {
 205                 'id': '2570327',
 206                 'ext': 'mp4',
 207                 'title': 'Lady-Umstyling für Audrina',
 208                 'description': 'md5:4c16d0c17a3461a0d43ea4084e96319d',
 209                 'upload_date': '20131014',
 210                 'duration': 606.76,
 211             },
 212             'params': {
 213                 # rtmp download
 214                 'skip_download': True,
 215             },
 216             'skip': 'Seems to be broken',
 217         },
 218         {
 219             'url': 'http://www.prosiebenmaxx.de/tv/experience/video/144-countdown-fuer-die-autowerkstatt-ganze-folge',
 220             'info_dict': {
 221                 'id': '2429369',
 222                 'ext': 'mp4',
 223                 'title': 'Countdown für die Autowerkstatt',
 224                 'description': 'md5:809fc051a457b5d8666013bc40698817',
 225                 'upload_date': '20140223',
 226                 'duration': 2595.04,
 227             },
 228             'params': {
 229                 # rtmp download
 230                 'skip_download': True,
 231             },
 232             'skip': 'This video is unavailable',
 233         },
 234         {
 235             'url': 'http://www.sixx.de/stars-style/video/sexy-laufen-in-ugg-boots-clip',
 236             'info_dict': {
 237                 'id': '2904997',
 238                 'ext': 'mp4',
 239                 'title': 'Sexy laufen in Ugg Boots',
 240                 'description': 'md5:edf42b8bd5bc4e5da4db4222c5acb7d6',
 241                 'upload_date': '20140122',
 242                 'duration': 245.32,
 243             },
 244             'params': {
 245                 # rtmp download
 246                 'skip_download': True,
 247             },
 248             'skip': 'This video is unavailable',
 249         },
 250         {
 251             'url': 'http://www.sat1.de/film/der-ruecktritt/video/im-interview-kai-wiesinger-clip',
 252             'info_dict': {
 253                 'id': '2906572',
 254                 'ext': 'mp4',
 255                 'title': 'Im Interview: Kai Wiesinger',
 256                 'description': 'md5:e4e5370652ec63b95023e914190b4eb9',
 257                 'upload_date': '20140203',
 258                 'duration': 522.56,
 259             },
 260             'params': {
 261                 # rtmp download
 262                 'skip_download': True,
 263             },
 264             'skip': 'This video is unavailable',
 265         },
 266         {
 267             'url': 'http://www.kabeleins.de/tv/rosins-restaurants/videos/jagd-auf-fertigkost-im-elsthal-teil-2-ganze-folge',
 268             'info_dict': {
 269                 'id': '2992323',
 270                 'ext': 'mp4',
 271                 'title': 'Jagd auf Fertigkost im Elsthal - Teil 2',
 272                 'description': 'md5:2669cde3febe9bce13904f701e774eb6',
 273                 'upload_date': '20141014',
 274                 'duration': 2410.44,
 275             },
 276             'params': {
 277                 # rtmp download
 278                 'skip_download': True,
 279             },
 280             'skip': 'This video is unavailable',
 281         },
 282         {
 283             'url': 'http://www.ran.de/fussball/bundesliga/video/schalke-toennies-moechte-raul-zurueck-ganze-folge',
 284             'info_dict': {
 285                 'id': '3004256',
 286                 'ext': 'mp4',
 287                 'title': 'Schalke: Tönnies möchte Raul zurück',
 288                 'description': 'md5:4b5b271d9bcde223b54390754c8ece3f',
 289                 'upload_date': '20140226',
 290                 'duration': 228.96,
 291             },
 292             'params': {
 293                 # rtmp download
 294                 'skip_download': True,
 295             },
 296             'skip': 'This video is unavailable',
 297         },
 298         {
 299             'url': 'http://www.the-voice-of-germany.de/video/31-andreas-kuemmert-rocket-man-clip',
 300             'info_dict': {
 301                 'id': '2572814',
 302                 'ext': 'mp4',
 303                 'title': 'The Voice of Germany - Andreas Kümmert: Rocket Man',
 304                 'description': 'md5:6ddb02b0781c6adf778afea606652e38',
 305                 'timestamp': 1382041620,
 306                 'upload_date': '20131017',
 307                 'duration': 469.88,
 308             },
 309             'params': {
 310                 'skip_download': True,
 311             },
 312         },
 313         {
 314             'url': 'http://www.fem.com/videos/beauty-lifestyle/kurztrips-zum-valentinstag',
 315             'info_dict': {
 316                 'id': '2156342',
 317                 'ext': 'mp4',
 318                 'title': 'Kurztrips zum Valentinstag',
 319                 'description': 'Romantischer Kurztrip zum Valentinstag? Nina Heinemann verrät, was sich hier wirklich lohnt.',
 320                 'duration': 307.24,
 321             },
 322             'params': {
 323                 'skip_download': True,
 324             },
 325         },
 326         {
 327             'url': 'http://www.prosieben.de/tv/joko-gegen-klaas/videos/playlists/episode-8-ganze-folge-playlist',
 328             'info_dict': {
 329                 'id': '439664',
 330                 'title': 'Episode 8 - Ganze Folge - Playlist',
 331                 'description': 'md5:63b8963e71f481782aeea877658dec84',
 332             },
 333             'playlist_count': 2,
 334             'skip': 'This video is unavailable',
 335         },
 336         {
 337             # title in <h2 class="subtitle">
 338             'url': 'http://www.prosieben.de/stars/oscar-award/videos/jetzt-erst-enthuellt-das-geheimnis-von-emma-stones-oscar-robe-clip',
 339             'info_dict': {
 340                 'id': '4895826',
 341                 'ext': 'mp4',
 342                 'title': 'Jetzt erst enthüllt: Das Geheimnis von Emma Stones Oscar-Robe',
 343                 'description': 'md5:e5ace2bc43fadf7b63adc6187e9450b9',
 344                 'upload_date': '20170302',
 345             },
 346             'params': {
 347                 'skip_download': True,
 348             },
 349             'skip': 'geo restricted to Germany',
 350         },
 351         {
 352             # geo restricted to Germany
 353             'url': 'http://www.kabeleinsdoku.de/tv/mayday-alarm-im-cockpit/video/102-notlandung-im-hudson-river-ganze-folge',
 354             'only_matching': True,
 355         },
 356         {
 357             # geo restricted to Germany
 358             'url': 'http://www.sat1gold.de/tv/edel-starck/video/11-staffel-1-episode-1-partner-wider-willen-ganze-folge',
 359             'only_matching': True,
 360         },
 361         {
 362             # geo restricted to Germany
 363             'url': 'https://www.galileo.tv/video/diese-emojis-werden-oft-missverstanden',
 364             'only_matching': True,
 365         },
 366         {
 367             'url': 'http://www.sat1gold.de/tv/edel-starck/playlist/die-gesamte-1-staffel',
 368             'only_matching': True,
 369         },
 370         {
 371             'url': 'http://www.advopedia.de/videos/lenssen-klaert-auf/lenssen-klaert-auf-folge-8-staffel-3-feiertage-und-freie-tage',
 372             'only_matching': True,
 373         },
 374     ]
 375
 376     _TOKEN = 'prosieben'
 377     _SALT = '01!8d8F_)r9]4s[qeuXfP%'
 378     _CLIENT_NAME = 'kolibri-2.0.19-splec4'
 379
 380     _ACCESS_ID = 'x_prosiebenmaxx-de'
 381     _ENCRYPTION_KEY = 'Eeyeey9oquahthainoofashoyoikosag'
 382     _IV = 'Aeluchoc6aevechuipiexeeboowedaok'
 383
 384     _CLIPID_REGEXES = [
 385         r'"clip_id"\s*:\s+"(\d+)"',
 386         r'clipid: "(\d+)"',
 387         r'clip[iI]d=(\d+)',
 388         r'clip[iI][dD]\s*=\s*["\'](\d+)',
 389         r"'itemImageUrl'\s*:\s*'/dynamic/thumbnails/full/\d+/(\d+)",
 390         r'proMamsId&quot;\s*:\s*&quot;(\d+)',
 391         r'proMamsId"\s*:\s*"(\d+)',
 392     ]
 393     _TITLE_REGEXES = [
 394         r'<h2 class="subtitle" itemprop="name">\s*(.+?)</h2>',
 395         r'<header class="clearfix">\s*<h3>(.+?)</h3>',
 396         r'<!-- start video -->\s*<h1>(.+?)</h1>',
 397         r'<h1 class="att-name">\s*(.+?)</h1>',
 398         r'<header class="module_header">\s*<h2>([^<]+)</h2>\s*</header>',
 399         r'<h2 class="video-title" itemprop="name">\s*(.+?)</h2>',
 400         r'<div[^>]+id="veeseoTitle"[^>]*>(.+?)</div>',
 401         r'<h2[^>]+class="subtitle"[^>]*>([^<]+)</h2>',
 402     ]
 403     _DESCRIPTION_REGEXES = [
 404         r'<p itemprop="description">\s*(.+?)</p>',
 405         r'<div class="videoDecription">\s*<p><strong>Beschreibung</strong>: (.+?)</p>',
 406         r'<div class="g-plusone" data-size="medium"></div>\s*</div>\s*</header>\s*(.+?)\s*<footer>',
 407         r'<p class="att-description">\s*(.+?)\s*</p>',
 408         r'<p class="video-description" itemprop="description">\s*(.+?)</p>',
 409         r'<div[^>]+id="veeseoDescription"[^>]*>(.+?)</div>',
 410     ]
 411     _UPLOAD_DATE_REGEXES = [
 412         r'<span>\s*(\d{2}\.\d{2}\.\d{4} \d{2}:\d{2}) \|\s*<span itemprop="duration"',
 413         r'<footer>\s*(\d{2}\.\d{2}\.\d{4}) \d{2}:\d{2} Uhr',
 414         r'<span style="padding-left: 4px;line-height:20px; color:#404040">(\d{2}\.\d{2}\.\d{4})</span>',
 415         r'(\d{2}\.\d{2}\.\d{4}) \| \d{2}:\d{2} Min<br/>',
 416     ]
 417     _PAGE_TYPE_REGEXES = [
 418         r'<meta name="page_type" content="([^"]+)">',
 419         r"'itemType'\s*:\s*'([^']*)'",
 420     ]
 421     _PLAYLIST_ID_REGEXES = [
 422         r'content[iI]d=(\d+)',
 423         r"'itemId'\s*:\s*'([^']*)'",
 424     ]
 425     _PLAYLIST_CLIP_REGEXES = [
 426         r'(?s)data-qvt=.+?<a href="([^"]+)"',
 427     ]
 428
 429     def _extract_clip(self, url, webpage):
 430         clip_id = self._html_search_regex(
 431             self._CLIPID_REGEXES, webpage, 'clip id')
 432         title = self._html_search_regex(
 433             self._TITLE_REGEXES, webpage, 'title',
 434             default=None) or self._og_search_title(webpage)
 435         info = self._extract_video_info(url, clip_id)
 436         description = self._html_search_regex(
 437             self._DESCRIPTION_REGEXES, webpage, 'description', default=None)
 438         if description is None:
 439             description = self._og_search_description(webpage)
 440         thumbnail = self._og_search_thumbnail(webpage)
 441         upload_date = unified_strdate(
 442             self._html_search_meta('og:published_time', webpage,
 443                                    'upload date', default=None)
 444             or self._html_search_regex(self._UPLOAD_DATE_REGEXES,
 445                                        webpage, 'upload date', default=None))
 446
 447         json_ld = self._search_json_ld(webpage, clip_id, default={})
 448
 449         return merge_dicts(info, {
 450             'id': clip_id,
 451             'title': title,
 452             'description': description,
 453             'thumbnail': thumbnail,
 454             'upload_date': upload_date,
 455         }, json_ld)
 456
 457     def _extract_playlist(self, url, webpage):
 458         playlist_id = self._html_search_regex(
 459             self._PLAYLIST_ID_REGEXES, webpage, 'playlist id')
 460         playlist = self._parse_json(
 461             self._search_regex(
 462                 r'var\s+contentResources\s*=\s*(\[.+?\]);\s*</script',
 463                 webpage, 'playlist'),
 464             playlist_id)
 465         entries = []
 466         for item in playlist:
 467             clip_id = item.get('id') or item.get('upc')
 468             if not clip_id:
 469                 continue
 470             info = self._extract_video_info(url, clip_id)
 471             info.update({
 472                 'id': clip_id,
 473                 'title': item.get('title') or item.get('teaser', {}).get('headline'),
 474                 'description': item.get('teaser', {}).get('description'),
 475                 'thumbnail': item.get('poster'),
 476                 'duration': float_or_none(item.get('duration')),
 477                 'series': item.get('tvShowTitle'),
 478                 'uploader': item.get('broadcastPublisher'),
 479             })
 480             entries.append(info)
 481         return self.playlist_result(entries, playlist_id)
 482
 483     def _real_extract(self, url):
 484         video_id = self._match_id(url)
 485         webpage = self._download_webpage(url, video_id)
 486         page_type = self._search_regex(
 487             self._PAGE_TYPE_REGEXES, webpage,
 488             'page type', default='clip').lower()
 489         if page_type == 'clip':
 490             return self._extract_clip(url, webpage)
 491         elif page_type == 'playlist':
 492             return self._extract_playlist(url, webpage)
 493         else:
 494             raise ExtractorError(
 495                 f'Unsupported page type {page_type}', expected=True)