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 )
  12
  13
  14 class ProSiebenSat1IE(InfoExtractor):
  15     IE_NAME = 'prosiebensat1'
  16     IE_DESC = 'ProSiebenSat.1 Digital'
  17     _VALID_URL = r'https?://(?:www\.)?(?:(?:prosieben|prosiebenmaxx|sixx|sat1|kabeleins|ran|the-voice-of-germany)\.de|fem\.com)/(?P<id>.+)'
  18
  19     _TESTS = [
  20         {
  21             'url': 'http://www.prosieben.de/tv/circus-halligalli/videos/218-staffel-2-episode-18-jahresrueckblick-ganze-folge',
  22             'info_dict': {
  23                 'id': '2104602',
  24                 'ext': 'mp4',
  25                 'title': 'Staffel 2, Episode 18 - Jahresrückblick',
  26                 'description': 'md5:8733c81b702ea472e069bc48bb658fc1',
  27                 'upload_date': '20131231',
  28                 'duration': 5845.04,
  29             },
  30             'params': {
  31                 # rtmp download
  32                 'skip_download': True,
  33             },
  34         },
  35         {
  36             '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',
  37             'info_dict': {
  38                 'id': '2570327',
  39                 'ext': 'mp4',
  40                 'title': 'Lady-Umstyling für Audrina',
  41                 'description': 'md5:4c16d0c17a3461a0d43ea4084e96319d',
  42                 'upload_date': '20131014',
  43                 'duration': 606.76,
  44             },
  45             'params': {
  46                 # rtmp download
  47                 'skip_download': True,
  48             },
  49             'skip': 'Seems to be broken',
  50         },
  51         {
  52             'url': 'http://www.prosiebenmaxx.de/tv/experience/video/144-countdown-fuer-die-autowerkstatt-ganze-folge',
  53             'info_dict': {
  54                 'id': '2429369',
  55                 'ext': 'mp4',
  56                 'title': 'Countdown für die Autowerkstatt',
  57                 'description': 'md5:809fc051a457b5d8666013bc40698817',
  58                 'upload_date': '20140223',
  59                 'duration': 2595.04,
  60             },
  61             'params': {
  62                 # rtmp download
  63                 'skip_download': True,
  64             },
  65         },
  66         {
  67             'url': 'http://www.sixx.de/stars-style/video/sexy-laufen-in-ugg-boots-clip',
  68             'info_dict': {
  69                 'id': '2904997',
  70                 'ext': 'mp4',
  71                 'title': 'Sexy laufen in Ugg Boots',
  72                 'description': 'md5:edf42b8bd5bc4e5da4db4222c5acb7d6',
  73                 'upload_date': '20140122',
  74                 'duration': 245.32,
  75             },
  76             'params': {
  77                 # rtmp download
  78                 'skip_download': True,
  79             },
  80         },
  81         {
  82             'url': 'http://www.sat1.de/film/der-ruecktritt/video/im-interview-kai-wiesinger-clip',
  83             'info_dict': {
  84                 'id': '2906572',
  85                 'ext': 'mp4',
  86                 'title': 'Im Interview: Kai Wiesinger',
  87                 'description': 'md5:e4e5370652ec63b95023e914190b4eb9',
  88                 'upload_date': '20140203',
  89                 'duration': 522.56,
  90             },
  91             'params': {
  92                 # rtmp download
  93                 'skip_download': True,
  94             },
  95         },
  96         {
  97             'url': 'http://www.kabeleins.de/tv/rosins-restaurants/videos/jagd-auf-fertigkost-im-elsthal-teil-2-ganze-folge',
  98             'info_dict': {
  99                 'id': '2992323',
 100                 'ext': 'mp4',
 101                 'title': 'Jagd auf Fertigkost im Elsthal - Teil 2',
 102                 'description': 'md5:2669cde3febe9bce13904f701e774eb6',
 103                 'upload_date': '20141014',
 104                 'duration': 2410.44,
 105             },
 106             'params': {
 107                 # rtmp download
 108                 'skip_download': True,
 109             },
 110         },
 111         {
 112             'url': 'http://www.ran.de/fussball/bundesliga/video/schalke-toennies-moechte-raul-zurueck-ganze-folge',
 113             'info_dict': {
 114                 'id': '3004256',
 115                 'ext': 'mp4',
 116                 'title': 'Schalke: Tönnies möchte Raul zurück',
 117                 'description': 'md5:4b5b271d9bcde223b54390754c8ece3f',
 118                 'upload_date': '20140226',
 119                 'duration': 228.96,
 120             },
 121             'params': {
 122                 # rtmp download
 123                 'skip_download': True,
 124             },
 125         },
 126         {
 127             'url': 'http://www.the-voice-of-germany.de/video/31-andreas-kuemmert-rocket-man-clip',
 128             'info_dict': {
 129                 'id': '2572814',
 130                 'ext': 'mp4',
 131                 'title': 'Andreas Kümmert: Rocket Man',
 132                 'description': 'md5:6ddb02b0781c6adf778afea606652e38',
 133                 'upload_date': '20131017',
 134                 'duration': 469.88,
 135             },
 136             'params': {
 137                 # rtmp download
 138                 'skip_download': True,
 139             },
 140         },
 141         {
 142             'url': 'http://www.fem.com/wellness/videos/wellness-video-clip-kurztripps-zum-valentinstag.html',
 143             'info_dict': {
 144                 'id': '2156342',
 145                 'ext': 'mp4',
 146                 'title': 'Kurztrips zum Valentinstag',
 147                 'description': 'Romantischer Kurztrip zum Valentinstag? Wir verraten, was sich hier wirklich lohnt.',
 148                 'duration': 307.24,
 149             },
 150             'params': {
 151                 # rtmp download
 152                 'skip_download': True,
 153             },
 154         },
 155         {
 156             'url': 'http://www.prosieben.de/tv/joko-gegen-klaas/videos/playlists/episode-8-ganze-folge-playlist',
 157             'info_dict': {
 158                 'id': '439664',
 159                 'title': 'Episode 8 - Ganze Folge - Playlist',
 160                 'description': 'md5:63b8963e71f481782aeea877658dec84',
 161             },
 162             'playlist_count': 2,
 163         },
 164     ]
 165
 166     _CLIPID_REGEXES = [
 167         r'"clip_id"\s*:\s+"(\d+)"',
 168         r'clipid: "(\d+)"',
 169         r'clip[iI]d=(\d+)',
 170         r"'itemImageUrl'\s*:\s*'/dynamic/thumbnails/full/\d+/(\d+)",
 171     ]
 172     _TITLE_REGEXES = [
 173         r'<h2 class="subtitle" itemprop="name">\s*(.+?)</h2>',
 174         r'<header class="clearfix">\s*<h3>(.+?)</h3>',
 175         r'<!-- start video -->\s*<h1>(.+?)</h1>',
 176         r'<h1 class="att-name">\s*(.+?)</h1>',
 177     ]
 178     _DESCRIPTION_REGEXES = [
 179         r'<p itemprop="description">\s*(.+?)</p>',
 180         r'<div class="videoDecription">\s*<p><strong>Beschreibung</strong>: (.+?)</p>',
 181         r'<div class="g-plusone" data-size="medium"></div>\s*</div>\s*</header>\s*(.+?)\s*<footer>',
 182         r'<p class="att-description">\s*(.+?)\s*</p>',
 183     ]
 184     _UPLOAD_DATE_REGEXES = [
 185         r'<meta property="og:published_time" content="(.+?)">',
 186         r'<span>\s*(\d{2}\.\d{2}\.\d{4} \d{2}:\d{2}) \|\s*<span itemprop="duration"',
 187         r'<footer>\s*(\d{2}\.\d{2}\.\d{4}) \d{2}:\d{2} Uhr',
 188         r'<span style="padding-left: 4px;line-height:20px; color:#404040">(\d{2}\.\d{2}\.\d{4})</span>',
 189         r'(\d{2}\.\d{2}\.\d{4}) \| \d{2}:\d{2} Min<br/>',
 190     ]
 191     _PAGE_TYPE_REGEXES = [
 192         r'<meta name="page_type" content="([^"]+)">',
 193         r"'itemType'\s*:\s*'([^']*)'",
 194     ]
 195     _PLAYLIST_ID_REGEXES = [
 196         r'content[iI]d=(\d+)',
 197         r"'itemId'\s*:\s*'([^']*)'",
 198     ]
 199     _PLAYLIST_CLIP_REGEXES = [
 200         r'(?s)data-qvt=.+?<a href="([^"]+)"',
 201     ]
 202
 203     def _extract_clip(self, url, webpage):
 204         clip_id = self._html_search_regex(self._CLIPID_REGEXES, webpage, 'clip id')
 205
 206         access_token = 'testclient'
 207         client_name = 'kolibri-1.2.5'
 208         client_location = url
 209
 210         videos_api_url = 'http://vas.sim-technik.de/vas/live/v2/videos?%s' % compat_urllib_parse.urlencode({
 211             'access_token': access_token,
 212             'client_location': client_location,
 213             'client_name': client_name,
 214             'ids': clip_id,
 215         })
 216
 217         videos = self._download_json(videos_api_url, clip_id, 'Downloading videos JSON')
 218
 219         duration = float(videos[0]['duration'])
 220         source_ids = [source['id'] for source in videos[0]['sources']]
 221         source_ids_str = ','.join(map(str, source_ids))
 222
 223         g = '01!8d8F_)r9]4s[qeuXfP%'
 224
 225         client_id = g[:2] + sha1(''.join([clip_id, g, access_token, client_location, g, client_name])
 226                                  .encode('utf-8')).hexdigest()
 227
 228         sources_api_url = 'http://vas.sim-technik.de/vas/live/v2/videos/%s/sources?%s' % (clip_id, compat_urllib_parse.urlencode({
 229             'access_token': access_token,
 230             'client_id': client_id,
 231             'client_location': client_location,
 232             'client_name': client_name,
 233         }))
 234
 235         sources = self._download_json(sources_api_url, clip_id, 'Downloading sources JSON')
 236         server_id = sources['server_id']
 237
 238         client_id = g[:2] + sha1(''.join([g, clip_id, access_token, server_id,
 239                                           client_location, source_ids_str, g, client_name])
 240                                  .encode('utf-8')).hexdigest()
 241
 242         url_api_url = 'http://vas.sim-technik.de/vas/live/v2/videos/%s/sources/url?%s' % (clip_id, compat_urllib_parse.urlencode({
 243             'access_token': access_token,
 244             'client_id': client_id,
 245             'client_location': client_location,
 246             'client_name': client_name,
 247             'server_id': server_id,
 248             'source_ids': source_ids_str,
 249         }))
 250
 251         urls = self._download_json(url_api_url, clip_id, 'Downloading urls JSON')
 252
 253         title = self._html_search_regex(self._TITLE_REGEXES, webpage, 'title')
 254         description = self._html_search_regex(self._DESCRIPTION_REGEXES, webpage, 'description', fatal=False)
 255         thumbnail = self._og_search_thumbnail(webpage)
 256
 257         upload_date = unified_strdate(self._html_search_regex(
 258             self._UPLOAD_DATE_REGEXES, webpage, 'upload date', default=None))
 259
 260         formats = []
 261
 262         urls_sources = urls['sources']
 263         if isinstance(urls_sources, dict):
 264             urls_sources = urls_sources.values()
 265
 266         def fix_bitrate(bitrate):
 267             return (bitrate // 1000) if bitrate % 1000 == 0 else bitrate
 268
 269         for source in urls_sources:
 270             protocol = source['protocol']
 271             if protocol == 'rtmp' or protocol == 'rtmpe':
 272                 mobj = re.search(r'^(?P<url>rtmpe?://[^/]+/(?P<app>[^/]+))/(?P<playpath>.+)$', source['url'])
 273                 if not mobj:
 274                     continue
 275                 formats.append({
 276                     'url': mobj.group('url'),
 277                     'app': mobj.group('app'),
 278                     'play_path': mobj.group('playpath'),
 279                     'player_url': 'http://livepassdl.conviva.com/hf/ver/2.79.0.17083/LivePassModuleMain.swf',
 280                     'page_url': 'http://www.prosieben.de',
 281                     'vbr': fix_bitrate(source['bitrate']),
 282                     'ext': 'mp4',
 283                     'format_id': '%s_%s' % (source['cdn'], source['bitrate']),
 284                 })
 285             else:
 286                 formats.append({
 287                     'url': source['url'],
 288                     'vbr': fix_bitrate(source['bitrate']),
 289                 })
 290
 291         self._sort_formats(formats)
 292
 293         return {
 294             'id': clip_id,
 295             'title': title,
 296             'description': description,
 297             'thumbnail': thumbnail,
 298             'upload_date': upload_date,
 299             'duration': duration,
 300             'formats': formats,
 301         }
 302
 303     def _extract_playlist(self, url, webpage):
 304         playlist_id = self._html_search_regex(
 305             self._PLAYLIST_ID_REGEXES, webpage, 'playlist id')
 306         for regex in self._PLAYLIST_CLIP_REGEXES:
 307             playlist_clips = re.findall(regex, webpage)
 308             if playlist_clips:
 309                 title = self._html_search_regex(
 310                     self._TITLE_REGEXES, webpage, 'title')
 311                 description = self._html_search_regex(
 312                     self._DESCRIPTION_REGEXES, webpage, 'description', fatal=False)
 313                 entries = [
 314                     self.url_result(
 315                         re.match('(.+?//.+?)/', url).group(1) + clip_path,
 316                         'ProSiebenSat1')
 317                     for clip_path in playlist_clips]
 318                 return self.playlist_result(entries, playlist_id, title, description)
 319
 320     def _real_extract(self, url):
 321         video_id = self._match_id(url)
 322         webpage = self._download_webpage(url, video_id)
 323         page_type = self._search_regex(
 324             self._PAGE_TYPE_REGEXES, webpage,
 325             'page type', default='clip').lower()
 326         if page_type == 'clip':
 327             return self._extract_clip(url, webpage)
 328         elif page_type == 'playlist':
 329             return self._extract_playlist(url, webpage)