Raphaël G. Git Repositories - youtubedl/blob - 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                             (?:beta\.)?
 133                             (?:
 134                                 prosieben(?:maxx)?|sixx|sat1(?:gold)?|kabeleins(?:doku)?|the-voice-of-germany|7tv|advopedia
 135                             )\.(?:de|at|ch)|
 136                             ran\.de|fem\.com|advopedia\.de
 137                         )
 138                         /(?P<id>.+)
 139                     '''
 140
 141     _TESTS = [
 142         {
 143             # Tests changes introduced in https://github.com/rg3/youtube-dl/pull/6242
 144             # in response to fixing https://github.com/rg3/youtube-dl/issues/6215:
 145             # - malformed f4m manifest support
 146             # - proper handling of URLs starting with `https?://` in 2.0 manifests
 147             # - recursive child f4m manifests extraction
 148             'url': 'http://www.prosieben.de/tv/circus-halligalli/videos/218-staffel-2-episode-18-jahresrueckblick-ganze-folge',
 149             'info_dict': {
 150                 'id': '2104602',
 151                 'ext': 'mp4',
 152                 'title': 'Episode 18 - Staffel 2',
 153                 'description': 'md5:8733c81b702ea472e069bc48bb658fc1',
 154                 'upload_date': '20131231',
 155                 'duration': 5845.04,
 156             },
 157         },
 158         {
 159             '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',
 160             'info_dict': {
 161                 'id': '2570327',
 162                 'ext': 'mp4',
 163                 'title': 'Lady-Umstyling für Audrina',
 164                 'description': 'md5:4c16d0c17a3461a0d43ea4084e96319d',
 165                 'upload_date': '20131014',
 166                 'duration': 606.76,
 167             },
 168             'params': {
 169                 # rtmp download
 170                 'skip_download': True,
 171             },
 172             'skip': 'Seems to be broken',
 173         },
 174         {
 175             'url': 'http://www.prosiebenmaxx.de/tv/experience/video/144-countdown-fuer-die-autowerkstatt-ganze-folge',
 176             'info_dict': {
 177                 'id': '2429369',
 178                 'ext': 'mp4',
 179                 'title': 'Countdown für die Autowerkstatt',
 180                 'description': 'md5:809fc051a457b5d8666013bc40698817',
 181                 'upload_date': '20140223',
 182                 'duration': 2595.04,
 183             },
 184             'params': {
 185                 # rtmp download
 186                 'skip_download': True,
 187             },
 188             'skip': 'This video is unavailable',
 189         },
 190         {
 191             'url': 'http://www.sixx.de/stars-style/video/sexy-laufen-in-ugg-boots-clip',
 192             'info_dict': {
 193                 'id': '2904997',
 194                 'ext': 'mp4',
 195                 'title': 'Sexy laufen in Ugg Boots',
 196                 'description': 'md5:edf42b8bd5bc4e5da4db4222c5acb7d6',
 197                 'upload_date': '20140122',
 198                 'duration': 245.32,
 199             },
 200             'params': {
 201                 # rtmp download
 202                 'skip_download': True,
 203             },
 204             'skip': 'This video is unavailable',
 205         },
 206         {
 207             'url': 'http://www.sat1.de/film/der-ruecktritt/video/im-interview-kai-wiesinger-clip',
 208             'info_dict': {
 209                 'id': '2906572',
 210                 'ext': 'mp4',
 211                 'title': 'Im Interview: Kai Wiesinger',
 212                 'description': 'md5:e4e5370652ec63b95023e914190b4eb9',
 213                 'upload_date': '20140203',
 214                 'duration': 522.56,
 215             },
 216             'params': {
 217                 # rtmp download
 218                 'skip_download': True,
 219             },
 220             'skip': 'This video is unavailable',
 221         },
 222         {
 223             'url': 'http://www.kabeleins.de/tv/rosins-restaurants/videos/jagd-auf-fertigkost-im-elsthal-teil-2-ganze-folge',
 224             'info_dict': {
 225                 'id': '2992323',
 226                 'ext': 'mp4',
 227                 'title': 'Jagd auf Fertigkost im Elsthal - Teil 2',
 228                 'description': 'md5:2669cde3febe9bce13904f701e774eb6',
 229                 'upload_date': '20141014',
 230                 'duration': 2410.44,
 231             },
 232             'params': {
 233                 # rtmp download
 234                 'skip_download': True,
 235             },
 236             'skip': 'This video is unavailable',
 237         },
 238         {
 239             'url': 'http://www.ran.de/fussball/bundesliga/video/schalke-toennies-moechte-raul-zurueck-ganze-folge',
 240             'info_dict': {
 241                 'id': '3004256',
 242                 'ext': 'mp4',
 243                 'title': 'Schalke: Tönnies möchte Raul zurück',
 244                 'description': 'md5:4b5b271d9bcde223b54390754c8ece3f',
 245                 'upload_date': '20140226',
 246                 'duration': 228.96,
 247             },
 248             'params': {
 249                 # rtmp download
 250                 'skip_download': True,
 251             },
 252             'skip': 'This video is unavailable',
 253         },
 254         {
 255             'url': 'http://www.the-voice-of-germany.de/video/31-andreas-kuemmert-rocket-man-clip',
 256             'info_dict': {
 257                 'id': '2572814',
 258                 'ext': 'mp4',
 259                 'title': 'Andreas Kümmert: Rocket Man',
 260                 'description': 'md5:6ddb02b0781c6adf778afea606652e38',
 261                 'upload_date': '20131017',
 262                 'duration': 469.88,
 263             },
 264             'params': {
 265                 'skip_download': True,
 266             },
 267         },
 268         {
 269             'url': 'http://www.fem.com/wellness/videos/wellness-video-clip-kurztripps-zum-valentinstag.html',
 270             'info_dict': {
 271                 'id': '2156342',
 272                 'ext': 'mp4',
 273                 'title': 'Kurztrips zum Valentinstag',
 274                 'description': 'Romantischer Kurztrip zum Valentinstag? Nina Heinemann verrät, was sich hier wirklich lohnt.',
 275                 'duration': 307.24,
 276             },
 277             'params': {
 278                 'skip_download': True,
 279             },
 280         },
 281         {
 282             'url': 'http://www.prosieben.de/tv/joko-gegen-klaas/videos/playlists/episode-8-ganze-folge-playlist',
 283             'info_dict': {
 284                 'id': '439664',
 285                 'title': 'Episode 8 - Ganze Folge - Playlist',
 286                 'description': 'md5:63b8963e71f481782aeea877658dec84',
 287             },
 288             'playlist_count': 2,
 289             'skip': 'This video is unavailable',
 290         },
 291         {
 292             'url': 'http://www.7tv.de/circus-halligalli/615-best-of-circus-halligalli-ganze-folge',
 293             'info_dict': {
 294                 'id': '4187506',
 295                 'ext': 'mp4',
 296                 'title': 'Best of Circus HalliGalli',
 297                 'description': 'md5:8849752efd90b9772c9db6fdf87fb9e9',
 298                 'upload_date': '20151229',
 299             },
 300             'params': {
 301                 'skip_download': True,
 302             },
 303         },
 304         {
 305             # title in <h2 class="subtitle">
 306             'url': 'http://www.prosieben.de/stars/oscar-award/videos/jetzt-erst-enthuellt-das-geheimnis-von-emma-stones-oscar-robe-clip',
 307             'info_dict': {
 308                 'id': '4895826',
 309                 'ext': 'mp4',
 310                 'title': 'Jetzt erst enthüllt: Das Geheimnis von Emma Stones Oscar-Robe',
 311                 'description': 'md5:e5ace2bc43fadf7b63adc6187e9450b9',
 312                 'upload_date': '20170302',
 313             },
 314             'params': {
 315                 'skip_download': True,
 316             },
 317             'skip': 'geo restricted to Germany',
 318         },
 319         {
 320             # geo restricted to Germany
 321             'url': 'http://www.kabeleinsdoku.de/tv/mayday-alarm-im-cockpit/video/102-notlandung-im-hudson-river-ganze-folge',
 322             'only_matching': True,
 323         },
 324         {
 325             # geo restricted to Germany
 326             'url': 'http://www.sat1gold.de/tv/edel-starck/video/11-staffel-1-episode-1-partner-wider-willen-ganze-folge',
 327             'only_matching': True,
 328         },
 329         {
 330             'url': 'http://www.sat1gold.de/tv/edel-starck/playlist/die-gesamte-1-staffel',
 331             'only_matching': True,
 332         },
 333         {
 334             'url': 'http://www.advopedia.de/videos/lenssen-klaert-auf/lenssen-klaert-auf-folge-8-staffel-3-feiertage-und-freie-tage',
 335             'only_matching': True,
 336         },
 337     ]
 338
 339     _TOKEN = 'prosieben'
 340     _SALT = '01!8d8F_)r9]4s[qeuXfP%'
 341     _CLIENT_NAME = 'kolibri-2.0.19-splec4'
 342     _CLIPID_REGEXES = [
 343         r'"clip_id"\s*:\s+"(\d+)"',
 344         r'clipid: "(\d+)"',
 345         r'clip[iI]d=(\d+)',
 346         r'clip[iI]d\s*=\s*["\'](\d+)',
 347         r"'itemImageUrl'\s*:\s*'/dynamic/thumbnails/full/\d+/(\d+)",
 348         r'proMamsId&quot;\s*:\s*&quot;(\d+)',
 349         r'proMamsId"\s*:\s*"(\d+)',
 350     ]
 351     _TITLE_REGEXES = [
 352         r'<h2 class="subtitle" itemprop="name">\s*(.+?)</h2>',
 353         r'<header class="clearfix">\s*<h3>(.+?)</h3>',
 354         r'<!-- start video -->\s*<h1>(.+?)</h1>',
 355         r'<h1 class="att-name">\s*(.+?)</h1>',
 356         r'<header class="module_header">\s*<h2>([^<]+)</h2>\s*</header>',
 357         r'<h2 class="video-title" itemprop="name">\s*(.+?)</h2>',
 358         r'<div[^>]+id="veeseoTitle"[^>]*>(.+?)</div>',
 359         r'<h2[^>]+class="subtitle"[^>]*>([^<]+)</h2>',
 360     ]
 361     _DESCRIPTION_REGEXES = [
 362         r'<p itemprop="description">\s*(.+?)</p>',
 363         r'<div class="videoDecription">\s*<p><strong>Beschreibung</strong>: (.+?)</p>',
 364         r'<div class="g-plusone" data-size="medium"></div>\s*</div>\s*</header>\s*(.+?)\s*<footer>',
 365         r'<p class="att-description">\s*(.+?)\s*</p>',
 366         r'<p class="video-description" itemprop="description">\s*(.+?)</p>',
 367         r'<div[^>]+id="veeseoDescription"[^>]*>(.+?)</div>',
 368     ]
 369     _UPLOAD_DATE_REGEXES = [
 370         r'<meta property="og:published_time" content="(.+?)">',
 371         r'<span>\s*(\d{2}\.\d{2}\.\d{4} \d{2}:\d{2}) \|\s*<span itemprop="duration"',
 372         r'<footer>\s*(\d{2}\.\d{2}\.\d{4}) \d{2}:\d{2} Uhr',
 373         r'<span style="padding-left: 4px;line-height:20px; color:#404040">(\d{2}\.\d{2}\.\d{4})</span>',
 374         r'(\d{2}\.\d{2}\.\d{4}) \| \d{2}:\d{2} Min<br/>',
 375     ]
 376     _PAGE_TYPE_REGEXES = [
 377         r'<meta name="page_type" content="([^"]+)">',
 378         r"'itemType'\s*:\s*'([^']*)'",
 379     ]
 380     _PLAYLIST_ID_REGEXES = [
 381         r'content[iI]d=(\d+)',
 382         r"'itemId'\s*:\s*'([^']*)'",
 383     ]
 384     _PLAYLIST_CLIP_REGEXES = [
 385         r'(?s)data-qvt=.+?<a href="([^"]+)"',
 386     ]
 387
 388     def _extract_clip(self, url, webpage):
 389         clip_id = self._html_search_regex(
 390             self._CLIPID_REGEXES, webpage, 'clip id')
 391         title = self._html_search_regex(
 392             self._TITLE_REGEXES, webpage, 'title',
 393             default=None) or self._og_search_title(webpage)
 394         info = self._extract_video_info(url, clip_id)
 395         description = self._html_search_regex(
 396             self._DESCRIPTION_REGEXES, webpage, 'description', default=None)
 397         if description is None:
 398             description = self._og_search_description(webpage)
 399         thumbnail = self._og_search_thumbnail(webpage)
 400         upload_date = unified_strdate(self._html_search_regex(
 401             self._UPLOAD_DATE_REGEXES, webpage, 'upload date', default=None))
 402
 403         info.update({
 404             'id': clip_id,
 405             'title': title,
 406             'description': description,
 407             'thumbnail': thumbnail,
 408             'upload_date': upload_date,
 409         })
 410         return info
 411
 412     def _extract_playlist(self, url, webpage):
 413         playlist_id = self._html_search_regex(
 414             self._PLAYLIST_ID_REGEXES, webpage, 'playlist id')
 415         playlist = self._parse_json(
 416             self._search_regex(
 417                 r'var\s+contentResources\s*=\s*(\[.+?\]);\s*</script',
 418                 webpage, 'playlist'),
 419             playlist_id)
 420         entries = []
 421         for item in playlist:
 422             clip_id = item.get('id') or item.get('upc')
 423             if not clip_id:
 424                 continue
 425             info = self._extract_video_info(url, clip_id)
 426             info.update({
 427                 'id': clip_id,
 428                 'title': item.get('title') or item.get('teaser', {}).get('headline'),
 429                 'description': item.get('teaser', {}).get('description'),
 430                 'thumbnail': item.get('poster'),
 431                 'duration': float_or_none(item.get('duration')),
 432                 'series': item.get('tvShowTitle'),
 433                 'uploader': item.get('broadcastPublisher'),
 434             })
 435             entries.append(info)
 436         return self.playlist_result(entries, playlist_id)
 437
 438     def _real_extract(self, url):
 439         video_id = self._match_id(url)
 440         webpage = self._download_webpage(url, video_id)
 441         page_type = self._search_regex(
 442             self._PAGE_TYPE_REGEXES, webpage,
 443             'page type', default='clip').lower()
 444         if page_type == 'clip':
 445             return self._extract_clip(url, webpage)
 446         elif page_type == 'playlist':
 447             return self._extract_playlist(url, webpage)
 448         else:
 449             raise ExtractorError(
 450                 'Unsupported page type %s' % page_type, expected=True)