youtube_dl/extractor/prosiebensat1.py

   1 # encoding: 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                 else:
  89                     tbr = fix_bitrate(source['bitrate'])
  90                     if protocol in ('rtmp', 'rtmpe'):
  91                         mobj = re.search(r'^(?P<url>rtmpe?://[^/]+)/(?P<path>.+)$', source_url)
  92                         if not mobj:
  93                             continue
  94                         path = mobj.group('path')
  95                         mp4colon_index = path.rfind('mp4:')
  96                         app = path[:mp4colon_index]
  97                         play_path = path[mp4colon_index:]
  98                         formats.append({
  99                             'url': '%s/%s' % (mobj.group('url'), app),
 100                             'app': app,
 101                             'play_path': play_path,
 102                             'player_url': 'http://livepassdl.conviva.com/hf/ver/2.79.0.17083/LivePassModuleMain.swf',
 103                             'page_url': 'http://www.prosieben.de',
 104                             'tbr': tbr,
 105                             'ext': 'flv',
 106                             'format_id': 'rtmp%s' % ('-%d' % tbr if tbr else ''),
 107                         })
 108                     else:
 109                         formats.append({
 110                             'url': source_url,
 111                             'tbr': tbr,
 112                             'format_id': 'http%s' % ('-%d' % tbr if tbr else ''),
 113                         })
 114         self._sort_formats(formats)
 115
 116         return {
 117             'duration': duration,
 118             'formats': formats,
 119         }
 120
 121
 122 class ProSiebenSat1IE(ProSiebenSat1BaseIE):
 123     IE_NAME = 'prosiebensat1'
 124     IE_DESC = 'ProSiebenSat.1 Digital'
 125     _VALID_URL = r'https?://(?:www\.)?(?:(?:prosieben|prosiebenmaxx|sixx|sat1|kabeleins|the-voice-of-germany|7tv)\.(?:de|at|ch)|ran\.de|fem\.com)/(?P<id>.+)'
 126
 127     _TESTS = [
 128         {
 129             # Tests changes introduced in https://github.com/rg3/youtube-dl/pull/6242
 130             # in response to fixing https://github.com/rg3/youtube-dl/issues/6215:
 131             # - malformed f4m manifest support
 132             # - proper handling of URLs starting with `https?://` in 2.0 manifests
 133             # - recursive child f4m manifests extraction
 134             'url': 'http://www.prosieben.de/tv/circus-halligalli/videos/218-staffel-2-episode-18-jahresrueckblick-ganze-folge',
 135             'info_dict': {
 136                 'id': '2104602',
 137                 'ext': 'flv',
 138                 'title': 'Episode 18 - Staffel 2',
 139                 'description': 'md5:8733c81b702ea472e069bc48bb658fc1',
 140                 'upload_date': '20131231',
 141                 'duration': 5845.04,
 142             },
 143             'params': {
 144                 # rtmp download
 145                 'skip_download': True,
 146             },
 147         },
 148         {
 149             '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',
 150             'info_dict': {
 151                 'id': '2570327',
 152                 'ext': 'mp4',
 153                 'title': 'Lady-Umstyling für Audrina',
 154                 'description': 'md5:4c16d0c17a3461a0d43ea4084e96319d',
 155                 'upload_date': '20131014',
 156                 'duration': 606.76,
 157             },
 158             'params': {
 159                 # rtmp download
 160                 'skip_download': True,
 161             },
 162             'skip': 'Seems to be broken',
 163         },
 164         {
 165             'url': 'http://www.prosiebenmaxx.de/tv/experience/video/144-countdown-fuer-die-autowerkstatt-ganze-folge',
 166             'info_dict': {
 167                 'id': '2429369',
 168                 'ext': 'mp4',
 169                 'title': 'Countdown für die Autowerkstatt',
 170                 'description': 'md5:809fc051a457b5d8666013bc40698817',
 171                 'upload_date': '20140223',
 172                 'duration': 2595.04,
 173             },
 174             'params': {
 175                 # rtmp download
 176                 'skip_download': True,
 177             },
 178             'skip': 'This video is unavailable',
 179         },
 180         {
 181             'url': 'http://www.sixx.de/stars-style/video/sexy-laufen-in-ugg-boots-clip',
 182             'info_dict': {
 183                 'id': '2904997',
 184                 'ext': 'mp4',
 185                 'title': 'Sexy laufen in Ugg Boots',
 186                 'description': 'md5:edf42b8bd5bc4e5da4db4222c5acb7d6',
 187                 'upload_date': '20140122',
 188                 'duration': 245.32,
 189             },
 190             'params': {
 191                 # rtmp download
 192                 'skip_download': True,
 193             },
 194             'skip': 'This video is unavailable',
 195         },
 196         {
 197             'url': 'http://www.sat1.de/film/der-ruecktritt/video/im-interview-kai-wiesinger-clip',
 198             'info_dict': {
 199                 'id': '2906572',
 200                 'ext': 'mp4',
 201                 'title': 'Im Interview: Kai Wiesinger',
 202                 'description': 'md5:e4e5370652ec63b95023e914190b4eb9',
 203                 'upload_date': '20140203',
 204                 'duration': 522.56,
 205             },
 206             'params': {
 207                 # rtmp download
 208                 'skip_download': True,
 209             },
 210             'skip': 'This video is unavailable',
 211         },
 212         {
 213             'url': 'http://www.kabeleins.de/tv/rosins-restaurants/videos/jagd-auf-fertigkost-im-elsthal-teil-2-ganze-folge',
 214             'info_dict': {
 215                 'id': '2992323',
 216                 'ext': 'mp4',
 217                 'title': 'Jagd auf Fertigkost im Elsthal - Teil 2',
 218                 'description': 'md5:2669cde3febe9bce13904f701e774eb6',
 219                 'upload_date': '20141014',
 220                 'duration': 2410.44,
 221             },
 222             'params': {
 223                 # rtmp download
 224                 'skip_download': True,
 225             },
 226             'skip': 'This video is unavailable',
 227         },
 228         {
 229             'url': 'http://www.ran.de/fussball/bundesliga/video/schalke-toennies-moechte-raul-zurueck-ganze-folge',
 230             'info_dict': {
 231                 'id': '3004256',
 232                 'ext': 'mp4',
 233                 'title': 'Schalke: Tönnies möchte Raul zurück',
 234                 'description': 'md5:4b5b271d9bcde223b54390754c8ece3f',
 235                 'upload_date': '20140226',
 236                 'duration': 228.96,
 237             },
 238             'params': {
 239                 # rtmp download
 240                 'skip_download': True,
 241             },
 242             'skip': 'This video is unavailable',
 243         },
 244         {
 245             'url': 'http://www.the-voice-of-germany.de/video/31-andreas-kuemmert-rocket-man-clip',
 246             'info_dict': {
 247                 'id': '2572814',
 248                 'ext': 'flv',
 249                 'title': 'Andreas Kümmert: Rocket Man',
 250                 'description': 'md5:6ddb02b0781c6adf778afea606652e38',
 251                 'upload_date': '20131017',
 252                 'duration': 469.88,
 253             },
 254             'params': {
 255                 'skip_download': True,
 256             },
 257         },
 258         {
 259             'url': 'http://www.fem.com/wellness/videos/wellness-video-clip-kurztripps-zum-valentinstag.html',
 260             'info_dict': {
 261                 'id': '2156342',
 262                 'ext': 'flv',
 263                 'title': 'Kurztrips zum Valentinstag',
 264                 'description': 'Romantischer Kurztrip zum Valentinstag? Nina Heinemann verrät, was sich hier wirklich lohnt.',
 265                 'duration': 307.24,
 266             },
 267             'params': {
 268                 'skip_download': True,
 269             },
 270         },
 271         {
 272             'url': 'http://www.prosieben.de/tv/joko-gegen-klaas/videos/playlists/episode-8-ganze-folge-playlist',
 273             'info_dict': {
 274                 'id': '439664',
 275                 'title': 'Episode 8 - Ganze Folge - Playlist',
 276                 'description': 'md5:63b8963e71f481782aeea877658dec84',
 277             },
 278             'playlist_count': 2,
 279         },
 280         {
 281             'url': 'http://www.7tv.de/circus-halligalli/615-best-of-circus-halligalli-ganze-folge',
 282             'info_dict': {
 283                 'id': '4187506',
 284                 'ext': 'flv',
 285                 'title': 'Best of Circus HalliGalli',
 286                 'description': 'md5:8849752efd90b9772c9db6fdf87fb9e9',
 287                 'upload_date': '20151229',
 288             },
 289             'params': {
 290                 'skip_download': True,
 291             },
 292         },
 293     ]
 294
 295     _TOKEN = 'prosieben'
 296     _SALT = '01!8d8F_)r9]4s[qeuXfP%'
 297     _CLIENT_NAME = 'kolibri-2.0.19-splec4'
 298     _CLIPID_REGEXES = [
 299         r'"clip_id"\s*:\s+"(\d+)"',
 300         r'clipid: "(\d+)"',
 301         r'clip[iI]d=(\d+)',
 302         r'clip[iI]d\s*=\s*["\'](\d+)',
 303         r"'itemImageUrl'\s*:\s*'/dynamic/thumbnails/full/\d+/(\d+)",
 304     ]
 305     _TITLE_REGEXES = [
 306         r'<h2 class="subtitle" itemprop="name">\s*(.+?)</h2>',
 307         r'<header class="clearfix">\s*<h3>(.+?)</h3>',
 308         r'<!-- start video -->\s*<h1>(.+?)</h1>',
 309         r'<h1 class="att-name">\s*(.+?)</h1>',
 310         r'<header class="module_header">\s*<h2>([^<]+)</h2>\s*</header>',
 311         r'<h2 class="video-title" itemprop="name">\s*(.+?)</h2>',
 312         r'<div[^>]+id="veeseoTitle"[^>]*>(.+?)</div>',
 313     ]
 314     _DESCRIPTION_REGEXES = [
 315         r'<p itemprop="description">\s*(.+?)</p>',
 316         r'<div class="videoDecription">\s*<p><strong>Beschreibung</strong>: (.+?)</p>',
 317         r'<div class="g-plusone" data-size="medium"></div>\s*</div>\s*</header>\s*(.+?)\s*<footer>',
 318         r'<p class="att-description">\s*(.+?)\s*</p>',
 319         r'<p class="video-description" itemprop="description">\s*(.+?)</p>',
 320         r'<div[^>]+id="veeseoDescription"[^>]*>(.+?)</div>',
 321     ]
 322     _UPLOAD_DATE_REGEXES = [
 323         r'<meta property="og:published_time" content="(.+?)">',
 324         r'<span>\s*(\d{2}\.\d{2}\.\d{4} \d{2}:\d{2}) \|\s*<span itemprop="duration"',
 325         r'<footer>\s*(\d{2}\.\d{2}\.\d{4}) \d{2}:\d{2} Uhr',
 326         r'<span style="padding-left: 4px;line-height:20px; color:#404040">(\d{2}\.\d{2}\.\d{4})</span>',
 327         r'(\d{2}\.\d{2}\.\d{4}) \| \d{2}:\d{2} Min<br/>',
 328     ]
 329     _PAGE_TYPE_REGEXES = [
 330         r'<meta name="page_type" content="([^"]+)">',
 331         r"'itemType'\s*:\s*'([^']*)'",
 332     ]
 333     _PLAYLIST_ID_REGEXES = [
 334         r'content[iI]d=(\d+)',
 335         r"'itemId'\s*:\s*'([^']*)'",
 336     ]
 337     _PLAYLIST_CLIP_REGEXES = [
 338         r'(?s)data-qvt=.+?<a href="([^"]+)"',
 339     ]
 340
 341     def _extract_clip(self, url, webpage):
 342         clip_id = self._html_search_regex(
 343             self._CLIPID_REGEXES, webpage, 'clip id')
 344         title = self._html_search_regex(self._TITLE_REGEXES, webpage, 'title')
 345         info = self._extract_video_info(url, clip_id)
 346         description = self._html_search_regex(
 347             self._DESCRIPTION_REGEXES, webpage, 'description', fatal=False)
 348         thumbnail = self._og_search_thumbnail(webpage)
 349         upload_date = unified_strdate(self._html_search_regex(
 350             self._UPLOAD_DATE_REGEXES, webpage, 'upload date', default=None))
 351
 352         info.update({
 353             'id': clip_id,
 354             'title': title,
 355             'description': description,
 356             'thumbnail': thumbnail,
 357             'upload_date': upload_date,
 358         })
 359         return info
 360
 361     def _extract_playlist(self, url, webpage):
 362         playlist_id = self._html_search_regex(
 363             self._PLAYLIST_ID_REGEXES, webpage, 'playlist id')
 364         for regex in self._PLAYLIST_CLIP_REGEXES:
 365             playlist_clips = re.findall(regex, webpage)
 366             if playlist_clips:
 367                 title = self._html_search_regex(
 368                     self._TITLE_REGEXES, webpage, 'title')
 369                 description = self._html_search_regex(
 370                     self._DESCRIPTION_REGEXES, webpage, 'description', fatal=False)
 371                 entries = [
 372                     self.url_result(
 373                         re.match('(.+?//.+?)/', url).group(1) + clip_path,
 374                         'ProSiebenSat1')
 375                     for clip_path in playlist_clips]
 376                 return self.playlist_result(entries, playlist_id, title, description)
 377
 378     def _real_extract(self, url):
 379         video_id = self._match_id(url)
 380         webpage = self._download_webpage(url, video_id)
 381         page_type = self._search_regex(
 382             self._PAGE_TYPE_REGEXES, webpage,
 383             'page type', default='clip').lower()
 384         if page_type == 'clip':
 385             return self._extract_clip(url, webpage)
 386         elif page_type == 'playlist':
 387             return self._extract_playlist(url, webpage)