youtube_dl/extractor/prosiebensat1.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from hashlib import sha1
   7 from .common import InfoExtractor
   8 from ..compat import compat_str
   9 from ..utils import (
  10     ExtractorError,
  11     determine_ext,
  12     float_or_none,
  13     int_or_none,
  14     unified_strdate,
  15 )
  16
  17
  18 class ProSiebenSat1BaseIE(InfoExtractor):
  19     def _extract_video_info(self, url, clip_id):
  20         client_location = url
  21
  22         video = self._download_json(
  23             'http://vas.sim-technik.de/vas/live/v2/videos',
  24             clip_id, 'Downloading videos JSON', query={
  25                 'access_token': self._TOKEN,
  26                 'client_location': client_location,
  27                 'client_name': self._CLIENT_NAME,
  28                 'ids': clip_id,
  29             })[0]
  30
  31         if video.get('is_protected') is True:
  32             raise ExtractorError('This video is DRM protected.', expected=True)
  33
  34         duration = float_or_none(video.get('duration'))
  35         source_ids = [compat_str(source['id']) for source in video['sources']]
  36
  37         client_id = self._SALT[:2] + sha1(''.join([clip_id, self._SALT, self._TOKEN, client_location, self._SALT, self._CLIENT_NAME]).encode('utf-8')).hexdigest()
  38
  39         sources = self._download_json(
  40             'http://vas.sim-technik.de/vas/live/v2/videos/%s/sources' % clip_id,
  41             clip_id, 'Downloading sources JSON', query={
  42                 'access_token': self._TOKEN,
  43                 'client_id': client_id,
  44                 'client_location': client_location,
  45                 'client_name': self._CLIENT_NAME,
  46             })
  47         server_id = sources['server_id']
  48
  49         def fix_bitrate(bitrate):
  50             bitrate = int_or_none(bitrate)
  51             if not bitrate:
  52                 return None
  53             return (bitrate // 1000) if bitrate % 1000 == 0 else bitrate
  54
  55         formats = []
  56         for source_id in source_ids:
  57             client_id = self._SALT[:2] + sha1(''.join([self._SALT, clip_id, self._TOKEN, server_id, client_location, source_id, self._SALT, self._CLIENT_NAME]).encode('utf-8')).hexdigest()
  58             urls = self._download_json(
  59                 'http://vas.sim-technik.de/vas/live/v2/videos/%s/sources/url' % clip_id,
  60                 clip_id, 'Downloading urls JSON', fatal=False, query={
  61                     'access_token': self._TOKEN,
  62                     'client_id': client_id,
  63                     'client_location': client_location,
  64                     'client_name': self._CLIENT_NAME,
  65                     'server_id': server_id,
  66                     'source_ids': source_id,
  67                 })
  68             if not urls:
  69                 continue
  70             if urls.get('status_code') != 0:
  71                 raise ExtractorError('This video is unavailable', expected=True)
  72             urls_sources = urls['sources']
  73             if isinstance(urls_sources, dict):
  74                 urls_sources = urls_sources.values()
  75             for source in urls_sources:
  76                 source_url = source.get('url')
  77                 if not source_url:
  78                     continue
  79                 protocol = source.get('protocol')
  80                 mimetype = source.get('mimetype')
  81                 if mimetype == 'application/f4m+xml' or 'f4mgenerator' in source_url or determine_ext(source_url) == 'f4m':
  82                     formats.extend(self._extract_f4m_formats(
  83                         source_url, clip_id, f4m_id='hds', fatal=False))
  84                 elif mimetype == 'application/x-mpegURL':
  85                     formats.extend(self._extract_m3u8_formats(
  86                         source_url, clip_id, 'mp4', 'm3u8_native',
  87                         m3u8_id='hls', fatal=False))
  88                 elif mimetype == 'application/dash+xml':
  89                     formats.extend(self._extract_mpd_formats(
  90                         source_url, clip_id, mpd_id='dash', fatal=False))
  91                 else:
  92                     tbr = fix_bitrate(source['bitrate'])
  93                     if protocol in ('rtmp', 'rtmpe'):
  94                         mobj = re.search(r'^(?P<url>rtmpe?://[^/]+)/(?P<path>.+)$', source_url)
  95                         if not mobj:
  96                             continue
  97                         path = mobj.group('path')
  98                         mp4colon_index = path.rfind('mp4:')
  99                         app = path[:mp4colon_index]
 100                         play_path = path[mp4colon_index:]
 101                         formats.append({
 102                             'url': '%s/%s' % (mobj.group('url'), app),
 103                             'app': app,
 104                             'play_path': play_path,
 105                             'player_url': 'http://livepassdl.conviva.com/hf/ver/2.79.0.17083/LivePassModuleMain.swf',
 106                             'page_url': 'http://www.prosieben.de',
 107                             'tbr': tbr,
 108                             'ext': 'flv',
 109                             'format_id': 'rtmp%s' % ('-%d' % tbr if tbr else ''),
 110                         })
 111                     else:
 112                         formats.append({
 113                             'url': source_url,
 114                             'tbr': tbr,
 115                             'format_id': 'http%s' % ('-%d' % tbr if tbr else ''),
 116                         })
 117         self._sort_formats(formats)
 118
 119         return {
 120             'duration': duration,
 121             'formats': formats,
 122         }
 123
 124
 125 class ProSiebenSat1IE(ProSiebenSat1BaseIE):
 126     IE_NAME = 'prosiebensat1'
 127     IE_DESC = 'ProSiebenSat.1 Digital'
 128     _VALID_URL = r'''(?x)
 129                     https?://
 130                         (?:www\.)?
 131                         (?:
 132                             (?:
 133                                 prosieben(?:maxx)?|sixx|sat1(?:gold)?|kabeleins(?:doku)?|the-voice-of-germany|7tv|advopedia
 134                             )\.(?:de|at|ch)|
 135                             ran\.de|fem\.com|advopedia\.de
 136                         )
 137                         /(?P<id>.+)
 138                     '''
 139
 140     _TESTS = [
 141         {
 142             # Tests changes introduced in https://github.com/rg3/youtube-dl/pull/6242
 143             # in response to fixing https://github.com/rg3/youtube-dl/issues/6215:
 144             # - malformed f4m manifest support
 145             # - proper handling of URLs starting with `https?://` in 2.0 manifests
 146             # - recursive child f4m manifests extraction
 147             'url': 'http://www.prosieben.de/tv/circus-halligalli/videos/218-staffel-2-episode-18-jahresrueckblick-ganze-folge',
 148             'info_dict': {
 149                 'id': '2104602',
 150                 'ext': 'mp4',
 151                 'title': 'Episode 18 - Staffel 2',
 152                 'description': 'md5:8733c81b702ea472e069bc48bb658fc1',
 153                 'upload_date': '20131231',
 154                 'duration': 5845.04,
 155             },
 156         },
 157         {
 158             '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',
 159             'info_dict': {
 160                 'id': '2570327',
 161                 'ext': 'mp4',
 162                 'title': 'Lady-Umstyling für Audrina',
 163                 'description': 'md5:4c16d0c17a3461a0d43ea4084e96319d',
 164                 'upload_date': '20131014',
 165                 'duration': 606.76,
 166             },
 167             'params': {
 168                 # rtmp download
 169                 'skip_download': True,
 170             },
 171             'skip': 'Seems to be broken',
 172         },
 173         {
 174             'url': 'http://www.prosiebenmaxx.de/tv/experience/video/144-countdown-fuer-die-autowerkstatt-ganze-folge',
 175             'info_dict': {
 176                 'id': '2429369',
 177                 'ext': 'mp4',
 178                 'title': 'Countdown für die Autowerkstatt',
 179                 'description': 'md5:809fc051a457b5d8666013bc40698817',
 180                 'upload_date': '20140223',
 181                 'duration': 2595.04,
 182             },
 183             'params': {
 184                 # rtmp download
 185                 'skip_download': True,
 186             },
 187             'skip': 'This video is unavailable',
 188         },
 189         {
 190             'url': 'http://www.sixx.de/stars-style/video/sexy-laufen-in-ugg-boots-clip',
 191             'info_dict': {
 192                 'id': '2904997',
 193                 'ext': 'mp4',
 194                 'title': 'Sexy laufen in Ugg Boots',
 195                 'description': 'md5:edf42b8bd5bc4e5da4db4222c5acb7d6',
 196                 'upload_date': '20140122',
 197                 'duration': 245.32,
 198             },
 199             'params': {
 200                 # rtmp download
 201                 'skip_download': True,
 202             },
 203             'skip': 'This video is unavailable',
 204         },
 205         {
 206             'url': 'http://www.sat1.de/film/der-ruecktritt/video/im-interview-kai-wiesinger-clip',
 207             'info_dict': {
 208                 'id': '2906572',
 209                 'ext': 'mp4',
 210                 'title': 'Im Interview: Kai Wiesinger',
 211                 'description': 'md5:e4e5370652ec63b95023e914190b4eb9',
 212                 'upload_date': '20140203',
 213                 'duration': 522.56,
 214             },
 215             'params': {
 216                 # rtmp download
 217                 'skip_download': True,
 218             },
 219             'skip': 'This video is unavailable',
 220         },
 221         {
 222             'url': 'http://www.kabeleins.de/tv/rosins-restaurants/videos/jagd-auf-fertigkost-im-elsthal-teil-2-ganze-folge',
 223             'info_dict': {
 224                 'id': '2992323',
 225                 'ext': 'mp4',
 226                 'title': 'Jagd auf Fertigkost im Elsthal - Teil 2',
 227                 'description': 'md5:2669cde3febe9bce13904f701e774eb6',
 228                 'upload_date': '20141014',
 229                 'duration': 2410.44,
 230             },
 231             'params': {
 232                 # rtmp download
 233                 'skip_download': True,
 234             },
 235             'skip': 'This video is unavailable',
 236         },
 237         {
 238             'url': 'http://www.ran.de/fussball/bundesliga/video/schalke-toennies-moechte-raul-zurueck-ganze-folge',
 239             'info_dict': {
 240                 'id': '3004256',
 241                 'ext': 'mp4',
 242                 'title': 'Schalke: Tönnies möchte Raul zurück',
 243                 'description': 'md5:4b5b271d9bcde223b54390754c8ece3f',
 244                 'upload_date': '20140226',
 245                 'duration': 228.96,
 246             },
 247             'params': {
 248                 # rtmp download
 249                 'skip_download': True,
 250             },
 251             'skip': 'This video is unavailable',
 252         },
 253         {
 254             'url': 'http://www.the-voice-of-germany.de/video/31-andreas-kuemmert-rocket-man-clip',
 255             'info_dict': {
 256                 'id': '2572814',
 257                 'ext': 'mp4',
 258                 'title': 'Andreas Kümmert: Rocket Man',
 259                 'description': 'md5:6ddb02b0781c6adf778afea606652e38',
 260                 'upload_date': '20131017',
 261                 'duration': 469.88,
 262             },
 263             'params': {
 264                 'skip_download': True,
 265             },
 266         },
 267         {
 268             'url': 'http://www.fem.com/wellness/videos/wellness-video-clip-kurztripps-zum-valentinstag.html',
 269             'info_dict': {
 270                 'id': '2156342',
 271                 'ext': 'mp4',
 272                 'title': 'Kurztrips zum Valentinstag',
 273                 'description': 'Romantischer Kurztrip zum Valentinstag? Nina Heinemann verrät, was sich hier wirklich lohnt.',
 274                 'duration': 307.24,
 275             },
 276             'params': {
 277                 'skip_download': True,
 278             },
 279         },
 280         {
 281             'url': 'http://www.prosieben.de/tv/joko-gegen-klaas/videos/playlists/episode-8-ganze-folge-playlist',
 282             'info_dict': {
 283                 'id': '439664',
 284                 'title': 'Episode 8 - Ganze Folge - Playlist',
 285                 'description': 'md5:63b8963e71f481782aeea877658dec84',
 286             },
 287             'playlist_count': 2,
 288             'skip': 'This video is unavailable',
 289         },
 290         {
 291             'url': 'http://www.7tv.de/circus-halligalli/615-best-of-circus-halligalli-ganze-folge',
 292             'info_dict': {
 293                 'id': '4187506',
 294                 'ext': 'mp4',
 295                 'title': 'Best of Circus HalliGalli',
 296                 'description': 'md5:8849752efd90b9772c9db6fdf87fb9e9',
 297                 'upload_date': '20151229',
 298             },
 299             'params': {
 300                 'skip_download': True,
 301             },
 302         },
 303         {
 304             # title in <h2 class="subtitle">
 305             'url': 'http://www.prosieben.de/stars/oscar-award/videos/jetzt-erst-enthuellt-das-geheimnis-von-emma-stones-oscar-robe-clip',
 306             'info_dict': {
 307                 'id': '4895826',
 308                 'ext': 'mp4',
 309                 'title': 'Jetzt erst enthüllt: Das Geheimnis von Emma Stones Oscar-Robe',
 310                 'description': 'md5:e5ace2bc43fadf7b63adc6187e9450b9',
 311                 'upload_date': '20170302',
 312             },
 313             'params': {
 314                 'skip_download': True,
 315             },
 316             'skip': 'geo restricted to Germany',
 317         },
 318         {
 319             # geo restricted to Germany
 320             'url': 'http://www.kabeleinsdoku.de/tv/mayday-alarm-im-cockpit/video/102-notlandung-im-hudson-river-ganze-folge',
 321             'only_matching': True,
 322         },
 323         {
 324             # geo restricted to Germany
 325             'url': 'http://www.sat1gold.de/tv/edel-starck/video/11-staffel-1-episode-1-partner-wider-willen-ganze-folge',
 326             'only_matching': True,
 327         },
 328         {
 329             'url': 'http://www.sat1gold.de/tv/edel-starck/playlist/die-gesamte-1-staffel',
 330             'only_matching': True,
 331         },
 332         {
 333             'url': 'http://www.advopedia.de/videos/lenssen-klaert-auf/lenssen-klaert-auf-folge-8-staffel-3-feiertage-und-freie-tage',
 334             'only_matching': True,
 335         },
 336     ]
 337
 338     _TOKEN = 'prosieben'
 339     _SALT = '01!8d8F_)r9]4s[qeuXfP%'
 340     _CLIENT_NAME = 'kolibri-2.0.19-splec4'
 341     _CLIPID_REGEXES = [
 342         r'"clip_id"\s*:\s+"(\d+)"',
 343         r'clipid: "(\d+)"',
 344         r'clip[iI]d=(\d+)',
 345         r'clip[iI]d\s*=\s*["\'](\d+)',
 346         r"'itemImageUrl'\s*:\s*'/dynamic/thumbnails/full/\d+/(\d+)",
 347     ]
 348     _TITLE_REGEXES = [
 349         r'<h2 class="subtitle" itemprop="name">\s*(.+?)</h2>',
 350         r'<header class="clearfix">\s*<h3>(.+?)</h3>',
 351         r'<!-- start video -->\s*<h1>(.+?)</h1>',
 352         r'<h1 class="att-name">\s*(.+?)</h1>',
 353         r'<header class="module_header">\s*<h2>([^<]+)</h2>\s*</header>',
 354         r'<h2 class="video-title" itemprop="name">\s*(.+?)</h2>',
 355         r'<div[^>]+id="veeseoTitle"[^>]*>(.+?)</div>',
 356         r'<h2[^>]+class="subtitle"[^>]*>([^<]+)</h2>',
 357     ]
 358     _DESCRIPTION_REGEXES = [
 359         r'<p itemprop="description">\s*(.+?)</p>',
 360         r'<div class="videoDecription">\s*<p><strong>Beschreibung</strong>: (.+?)</p>',
 361         r'<div class="g-plusone" data-size="medium"></div>\s*</div>\s*</header>\s*(.+?)\s*<footer>',
 362         r'<p class="att-description">\s*(.+?)\s*</p>',
 363         r'<p class="video-description" itemprop="description">\s*(.+?)</p>',
 364         r'<div[^>]+id="veeseoDescription"[^>]*>(.+?)</div>',
 365     ]
 366     _UPLOAD_DATE_REGEXES = [
 367         r'<meta property="og:published_time" content="(.+?)">',
 368         r'<span>\s*(\d{2}\.\d{2}\.\d{4} \d{2}:\d{2}) \|\s*<span itemprop="duration"',
 369         r'<footer>\s*(\d{2}\.\d{2}\.\d{4}) \d{2}:\d{2} Uhr',
 370         r'<span style="padding-left: 4px;line-height:20px; color:#404040">(\d{2}\.\d{2}\.\d{4})</span>',
 371         r'(\d{2}\.\d{2}\.\d{4}) \| \d{2}:\d{2} Min<br/>',
 372     ]
 373     _PAGE_TYPE_REGEXES = [
 374         r'<meta name="page_type" content="([^"]+)">',
 375         r"'itemType'\s*:\s*'([^']*)'",
 376     ]
 377     _PLAYLIST_ID_REGEXES = [
 378         r'content[iI]d=(\d+)',
 379         r"'itemId'\s*:\s*'([^']*)'",
 380     ]
 381     _PLAYLIST_CLIP_REGEXES = [
 382         r'(?s)data-qvt=.+?<a href="([^"]+)"',
 383     ]
 384
 385     def _extract_clip(self, url, webpage):
 386         clip_id = self._html_search_regex(
 387             self._CLIPID_REGEXES, webpage, 'clip id')
 388         title = self._html_search_regex(
 389             self._TITLE_REGEXES, webpage, 'title',
 390             default=None) or self._og_search_title(webpage)
 391         info = self._extract_video_info(url, clip_id)
 392         description = self._html_search_regex(
 393             self._DESCRIPTION_REGEXES, webpage, 'description', default=None)
 394         if description is None:
 395             description = self._og_search_description(webpage)
 396         thumbnail = self._og_search_thumbnail(webpage)
 397         upload_date = unified_strdate(self._html_search_regex(
 398             self._UPLOAD_DATE_REGEXES, webpage, 'upload date', default=None))
 399
 400         info.update({
 401             'id': clip_id,
 402             'title': title,
 403             'description': description,
 404             'thumbnail': thumbnail,
 405             'upload_date': upload_date,
 406         })
 407         return info
 408
 409     def _extract_playlist(self, url, webpage):
 410         playlist_id = self._html_search_regex(
 411             self._PLAYLIST_ID_REGEXES, webpage, 'playlist id')
 412         playlist = self._parse_json(
 413             self._search_regex(
 414                 r'var\s+contentResources\s*=\s*(\[.+?\]);\s*</script',
 415                 webpage, 'playlist'),
 416             playlist_id)
 417         entries = []
 418         for item in playlist:
 419             clip_id = item.get('id') or item.get('upc')
 420             if not clip_id:
 421                 continue
 422             info = self._extract_video_info(url, clip_id)
 423             info.update({
 424                 'id': clip_id,
 425                 'title': item.get('title') or item.get('teaser', {}).get('headline'),
 426                 'description': item.get('teaser', {}).get('description'),
 427                 'thumbnail': item.get('poster'),
 428                 'duration': float_or_none(item.get('duration')),
 429                 'series': item.get('tvShowTitle'),
 430                 'uploader': item.get('broadcastPublisher'),
 431             })
 432             entries.append(info)
 433         return self.playlist_result(entries, playlist_id)
 434
 435     def _real_extract(self, url):
 436         video_id = self._match_id(url)
 437         webpage = self._download_webpage(url, video_id)
 438         page_type = self._search_regex(
 439             self._PAGE_TYPE_REGEXES, webpage,
 440             'page type', default='clip').lower()
 441         if page_type == 'clip':
 442             return self._extract_clip(url, webpage)
 443         elif page_type == 'playlist':
 444             return self._extract_playlist(url, webpage)
 445         else:
 446             raise ExtractorError(
 447                 'Unsupported page type %s' % page_type, expected=True)