_ Git - youtube-dl/blob - 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 ..utils import (
   9     compat_urllib_parse,
  10     unified_strdate,
  11     ExtractorError,
  12 )
  13
  14
  15 class ProSiebenSat1IE(InfoExtractor):
  16     IE_NAME = 'prosiebensat1'
  17     IE_DESC = 'ProSiebenSat.1 Digital'
  18     _VALID_URL = r'https?://(?:www\.)?(?:(?:prosieben|prosiebenmaxx|sixx|sat1|kabeleins|ran|the-voice-of-germany)\.de|fem\.com)/(?P<id>.+)'
  19
  20     _TESTS = [
  21         {
  22             'url': 'http://www.prosieben.de/tv/circus-halligalli/videos/218-staffel-2-episode-18-jahresrueckblick-ganze-folge',
  23             'info_dict': {
  24                 'id': '2104602',
  25                 'ext': 'mp4',
  26                 'title': 'Staffel 2, Episode 18 - Jahresrückblick',
  27                 'description': 'md5:8733c81b702ea472e069bc48bb658fc1',
  28                 'upload_date': '20131231',
  29                 'duration': 5845.04,
  30             },
  31             'params': {
  32                 # rtmp download
  33                 'skip_download': True,
  34             },
  35         },
  36         {
  37             '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',
  38             'info_dict': {
  39                 'id': '2570327',
  40                 'ext': 'mp4',
  41                 'title': 'Lady-Umstyling für Audrina',
  42                 'description': 'md5:4c16d0c17a3461a0d43ea4084e96319d',
  43                 'upload_date': '20131014',
  44                 'duration': 606.76,
  45             },
  46             'params': {
  47                 # rtmp download
  48                 'skip_download': True,
  49             },
  50             'skip': 'Seems to be broken',
  51         },
  52         {
  53             'url': 'http://www.prosiebenmaxx.de/tv/experience/video/144-countdown-fuer-die-autowerkstatt-ganze-folge',
  54             'info_dict': {
  55                 'id': '2429369',
  56                 'ext': 'mp4',
  57                 'title': 'Countdown für die Autowerkstatt',
  58                 'description': 'md5:809fc051a457b5d8666013bc40698817',
  59                 'upload_date': '20140223',
  60                 'duration': 2595.04,
  61             },
  62             'params': {
  63                 # rtmp download
  64                 'skip_download': True,
  65             },
  66         },
  67         {
  68             'url': 'http://www.sixx.de/stars-style/video/sexy-laufen-in-ugg-boots-clip',
  69             'info_dict': {
  70                 'id': '2904997',
  71                 'ext': 'mp4',
  72                 'title': 'Sexy laufen in Ugg Boots',
  73                 'description': 'md5:edf42b8bd5bc4e5da4db4222c5acb7d6',
  74                 'upload_date': '20140122',
  75                 'duration': 245.32,
  76             },
  77             'params': {
  78                 # rtmp download
  79                 'skip_download': True,
  80             },
  81         },
  82         {
  83             'url': 'http://www.sat1.de/film/der-ruecktritt/video/im-interview-kai-wiesinger-clip',
  84             'info_dict': {
  85                 'id': '2906572',
  86                 'ext': 'mp4',
  87                 'title': 'Im Interview: Kai Wiesinger',
  88                 'description': 'md5:e4e5370652ec63b95023e914190b4eb9',
  89                 'upload_date': '20140203',
  90                 'duration': 522.56,
  91             },
  92             'params': {
  93                 # rtmp download
  94                 'skip_download': True,
  95             },
  96         },
  97         {
  98             'url': 'http://www.kabeleins.de/tv/rosins-restaurants/videos/jagd-auf-fertigkost-im-elsthal-teil-2-ganze-folge',
  99             'info_dict': {
 100                 'id': '2992323',
 101                 'ext': 'mp4',
 102                 'title': 'Jagd auf Fertigkost im Elsthal - Teil 2',
 103                 'description': 'md5:2669cde3febe9bce13904f701e774eb6',
 104                 'upload_date': '20141014',
 105                 'duration': 2410.44,
 106             },
 107             'params': {
 108                 # rtmp download
 109                 'skip_download': True,
 110             },
 111         },
 112         {
 113             'url': 'http://www.ran.de/fussball/bundesliga/video/schalke-toennies-moechte-raul-zurueck-ganze-folge',
 114             'info_dict': {
 115                 'id': '3004256',
 116                 'ext': 'mp4',
 117                 'title': 'Schalke: Tönnies möchte Raul zurück',
 118                 'description': 'md5:4b5b271d9bcde223b54390754c8ece3f',
 119                 'upload_date': '20140226',
 120                 'duration': 228.96,
 121             },
 122             'params': {
 123                 # rtmp download
 124                 'skip_download': True,
 125             },
 126         },
 127         {
 128             'url': 'http://www.the-voice-of-germany.de/video/31-andreas-kuemmert-rocket-man-clip',
 129             'info_dict': {
 130                 'id': '2572814',
 131                 'ext': 'mp4',
 132                 'title': 'Andreas Kümmert: Rocket Man',
 133                 'description': 'md5:6ddb02b0781c6adf778afea606652e38',
 134                 'upload_date': '20131017',
 135                 'duration': 469.88,
 136             },
 137             'params': {
 138                 # rtmp download
 139                 'skip_download': True,
 140             },
 141         },
 142         {
 143             'url': 'http://www.fem.com/wellness/videos/wellness-video-clip-kurztripps-zum-valentinstag.html',
 144             'info_dict': {
 145                 'id': '2156342',
 146                 'ext': 'mp4',
 147                 'title': 'Kurztrips zum Valentinstag',
 148                 'description': 'Romantischer Kurztrip zum Valentinstag? Wir verraten, was sich hier wirklich lohnt.',
 149                 'duration': 307.24,
 150             },
 151             'params': {
 152                 # rtmp download
 153                 'skip_download': True,
 154             },
 155         },
 156         {
 157             'url': 'http://www.prosieben.de/tv/joko-gegen-klaas/videos/playlists/episode-8-ganze-folge-playlist',
 158             'info_dict': {
 159                 'id': '439664',
 160                 'title': 'Episode 8 - Ganze Folge - Playlist',
 161                 'description': 'Das finale und härteste Duell aller Zeiten ist vorbei! Der Weltmeister für dieses Jahr steht! Alle packenden Duelle der achten Episode von "Joko gegen Klaas - das Duell um die Welt" seht ihr hier noch einmal in voller Länge!',
 162             },
 163             'playlist_count': 2,
 164         },
 165     ]
 166
 167     _CLIPID_REGEXES = [
 168         r'"clip_id"\s*:\s+"(\d+)"',
 169         r'clipid: "(\d+)"',
 170         r'clip[iI]d=(\d+)',
 171         r"'itemImageUrl'\s*:\s*'/dynamic/thumbnails/full/\d+/(\d+)",
 172     ]
 173     _TITLE_REGEXES = [
 174         r'<h2 class="subtitle" itemprop="name">\s*(.+?)</h2>',
 175         r'<header class="clearfix">\s*<h3>(.+?)</h3>',
 176         r'<!-- start video -->\s*<h1>(.+?)</h1>',
 177         r'<h1 class="att-name">\s*(.+?)</h1>',
 178     ]
 179     _DESCRIPTION_REGEXES = [
 180         r'<p itemprop="description">\s*(.+?)</p>',
 181         r'<div class="videoDecription">\s*<p><strong>Beschreibung</strong>: (.+?)</p>',
 182         r'<div class="g-plusone" data-size="medium"></div>\s*</div>\s*</header>\s*(.+?)\s*<footer>',
 183         r'<p class="att-description">\s*(.+?)\s*</p>',
 184     ]
 185     _UPLOAD_DATE_REGEXES = [
 186         r'<meta property="og:published_time" content="(.+?)">',
 187         r'<span>\s*(\d{2}\.\d{2}\.\d{4} \d{2}:\d{2}) \|\s*<span itemprop="duration"',
 188         r'<footer>\s*(\d{2}\.\d{2}\.\d{4}) \d{2}:\d{2} Uhr',
 189         r'<span style="padding-left: 4px;line-height:20px; color:#404040">(\d{2}\.\d{2}\.\d{4})</span>',
 190         r'(\d{2}\.\d{2}\.\d{4}) \| \d{2}:\d{2} Min<br/>',
 191     ]
 192     _ITEM_TYPE_REGEXES = [
 193         r"'itemType'\s*:\s*'([^']*)'",
 194     ]
 195     _ITEM_ID_REGEXES = [
 196         r"'itemId'\s*:\s*'([^']*)'",
 197     ]
 198     _PLAYLIST_CLIPS_REGEXES = [
 199         r'data-qvt=.+?<a href="([^"]+)"',
 200     ]
 201
 202     def _real_extract(self, url):
 203         video_id = self._match_id(url)
 204         webpage = self._download_webpage(url, video_id)
 205
 206         item_type = self._html_search_regex(self._ITEM_TYPE_REGEXES, webpage, 'item type', default='CLIP')
 207         if item_type == 'CLIP':
 208             return self._clip_extract(url, webpage)
 209         elif item_type == 'PLAYLIST':
 210             playlist_id = self._html_search_regex(self._ITEM_ID_REGEXES, webpage, 'playlist id')
 211
 212             for regex in self._PLAYLIST_CLIPS_REGEXES:
 213                 playlist_clips = re.findall(regex, webpage, re.DOTALL)
 214                 if playlist_clips:
 215                     title = self._html_search_regex(self._TITLE_REGEXES, webpage, 'title')
 216                     description = self._html_search_regex(self._DESCRIPTION_REGEXES, webpage, 'description', fatal=False)
 217                     root_url = re.match('(.+?//.+?)/', url).group(1)
 218
 219                     return {
 220                         '_type': 'playlist',
 221                         'id': playlist_id,
 222                         'title': title,
 223                         'description': description,
 224                         'entries': [self._clip_extract(root_url + clip_path) for clip_path in playlist_clips]
 225                     }
 226         else:
 227             raise ExtractorError('Unknown item type "%s"' % item_type)
 228
 229     def _clip_extract(self, url, webpage=None):
 230         if webpage is None:
 231             video_id = self._match_id(url)
 232             webpage = self._download_webpage(url, video_id)
 233
 234         clip_id = self._html_search_regex(self._CLIPID_REGEXES, webpage, 'clip id')
 235
 236         access_token = 'testclient'
 237         client_name = 'kolibri-1.2.5'
 238         client_location = url
 239
 240         videos_api_url = 'http://vas.sim-technik.de/vas/live/v2/videos?%s' % compat_urllib_parse.urlencode({
 241             'access_token': access_token,
 242             'client_location': client_location,
 243             'client_name': client_name,
 244             'ids': clip_id,
 245         })
 246
 247         videos = self._download_json(videos_api_url, clip_id, 'Downloading videos JSON')
 248
 249         duration = float(videos[0]['duration'])
 250         source_ids = [source['id'] for source in videos[0]['sources']]
 251         source_ids_str = ','.join(map(str, source_ids))
 252
 253         g = '01!8d8F_)r9]4s[qeuXfP%'
 254
 255         client_id = g[:2] + sha1(''.join([clip_id, g, access_token, client_location, g, client_name])
 256                                  .encode('utf-8')).hexdigest()
 257
 258         sources_api_url = 'http://vas.sim-technik.de/vas/live/v2/videos/%s/sources?%s' % (clip_id, compat_urllib_parse.urlencode({
 259             'access_token': access_token,
 260             'client_id': client_id,
 261             'client_location': client_location,
 262             'client_name': client_name,
 263         }))
 264
 265         sources = self._download_json(sources_api_url, clip_id, 'Downloading sources JSON')
 266         server_id = sources['server_id']
 267
 268         client_id = g[:2] + sha1(''.join([g, clip_id, access_token, server_id,
 269                                           client_location, source_ids_str, g, client_name])
 270                                  .encode('utf-8')).hexdigest()
 271
 272         url_api_url = 'http://vas.sim-technik.de/vas/live/v2/videos/%s/sources/url?%s' % (clip_id, compat_urllib_parse.urlencode({
 273             'access_token': access_token,
 274             'client_id': client_id,
 275             'client_location': client_location,
 276             'client_name': client_name,
 277             'server_id': server_id,
 278             'source_ids': source_ids_str,
 279         }))
 280
 281         urls = self._download_json(url_api_url, clip_id, 'Downloading urls JSON')
 282
 283         title = self._html_search_regex(self._TITLE_REGEXES, webpage, 'title')
 284         description = self._html_search_regex(self._DESCRIPTION_REGEXES, webpage, 'description', fatal=False)
 285         thumbnail = self._og_search_thumbnail(webpage)
 286
 287         upload_date = unified_strdate(self._html_search_regex(
 288             self._UPLOAD_DATE_REGEXES, webpage, 'upload date', default=None))
 289
 290         formats = []
 291
 292         urls_sources = urls['sources']
 293         if isinstance(urls_sources, dict):
 294             urls_sources = urls_sources.values()
 295
 296         def fix_bitrate(bitrate):
 297             return (bitrate // 1000) if bitrate % 1000 == 0 else bitrate
 298
 299         for source in urls_sources:
 300             protocol = source['protocol']
 301             if protocol == 'rtmp' or protocol == 'rtmpe':
 302                 mobj = re.search(r'^(?P<url>rtmpe?://[^/]+/(?P<app>[^/]+))/(?P<playpath>.+)$', source['url'])
 303                 if not mobj:
 304                     continue
 305                 formats.append({
 306                     'url': mobj.group('url'),
 307                     'app': mobj.group('app'),
 308                     'play_path': mobj.group('playpath'),
 309                     'player_url': 'http://livepassdl.conviva.com/hf/ver/2.79.0.17083/LivePassModuleMain.swf',
 310                     'page_url': 'http://www.prosieben.de',
 311                     'vbr': fix_bitrate(source['bitrate']),
 312                     'ext': 'mp4',
 313                     'format_id': '%s_%s' % (source['cdn'], source['bitrate']),
 314                 })
 315             else:
 316                 formats.append({
 317                     'url': source['url'],
 318                     'vbr': fix_bitrate(source['bitrate']),
 319                 })
 320
 321         self._sort_formats(formats)
 322
 323         return {
 324             'id': clip_id,
 325             'title': title,
 326             'description': description,
 327             'thumbnail': thumbnail,
 328             'upload_date': upload_date,
 329             'duration': duration,
 330             'formats': formats,
 331         }