yt_dlp/extractor/prosiebensat1.py

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