Raphaël G. Git Repositories - youtubedl/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 ..compat import (
   9     compat_urllib_parse,
  10 )
  11 from ..utils import (
  12     unified_strdate,
  13     int_or_none,
  14 )
  15
  16
  17 class ProSiebenSat1IE(InfoExtractor):
  18     IE_NAME = 'prosiebensat1'
  19     IE_DESC = 'ProSiebenSat.1 Digital'
  20     _VALID_URL = r'https?://(?:www\.)?(?:(?:prosieben|prosiebenmaxx|sixx|sat1|kabeleins|the-voice-of-germany)\.(?:de|at)|ran\.de|fem\.com)/(?P<id>.+)'
  21
  22     _TESTS = [
  23         {
  24             'url': 'http://www.prosieben.de/tv/circus-halligalli/videos/218-staffel-2-episode-18-jahresrueckblick-ganze-folge',
  25             'info_dict': {
  26                 'id': '2104602',
  27                 'ext': 'mp4',
  28                 'title': 'Episode 18 - Staffel 2',
  29                 'description': 'md5:8733c81b702ea472e069bc48bb658fc1',
  30                 'upload_date': '20131231',
  31                 'duration': 5845.04,
  32             },
  33             'params': {
  34                 # rtmp download
  35                 'skip_download': True,
  36             },
  37         },
  38         {
  39             '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',
  40             'info_dict': {
  41                 'id': '2570327',
  42                 'ext': 'mp4',
  43                 'title': 'Lady-Umstyling für Audrina',
  44                 'description': 'md5:4c16d0c17a3461a0d43ea4084e96319d',
  45                 'upload_date': '20131014',
  46                 'duration': 606.76,
  47             },
  48             'params': {
  49                 # rtmp download
  50                 'skip_download': True,
  51             },
  52             'skip': 'Seems to be broken',
  53         },
  54         {
  55             'url': 'http://www.prosiebenmaxx.de/tv/experience/video/144-countdown-fuer-die-autowerkstatt-ganze-folge',
  56             'info_dict': {
  57                 'id': '2429369',
  58                 'ext': 'mp4',
  59                 'title': 'Countdown für die Autowerkstatt',
  60                 'description': 'md5:809fc051a457b5d8666013bc40698817',
  61                 'upload_date': '20140223',
  62                 'duration': 2595.04,
  63             },
  64             'params': {
  65                 # rtmp download
  66                 'skip_download': True,
  67             },
  68         },
  69         {
  70             'url': 'http://www.sixx.de/stars-style/video/sexy-laufen-in-ugg-boots-clip',
  71             'info_dict': {
  72                 'id': '2904997',
  73                 'ext': 'mp4',
  74                 'title': 'Sexy laufen in Ugg Boots',
  75                 'description': 'md5:edf42b8bd5bc4e5da4db4222c5acb7d6',
  76                 'upload_date': '20140122',
  77                 'duration': 245.32,
  78             },
  79             'params': {
  80                 # rtmp download
  81                 'skip_download': True,
  82             },
  83         },
  84         {
  85             'url': 'http://www.sat1.de/film/der-ruecktritt/video/im-interview-kai-wiesinger-clip',
  86             'info_dict': {
  87                 'id': '2906572',
  88                 'ext': 'mp4',
  89                 'title': 'Im Interview: Kai Wiesinger',
  90                 'description': 'md5:e4e5370652ec63b95023e914190b4eb9',
  91                 'upload_date': '20140203',
  92                 'duration': 522.56,
  93             },
  94             'params': {
  95                 # rtmp download
  96                 'skip_download': True,
  97             },
  98         },
  99         {
 100             'url': 'http://www.kabeleins.de/tv/rosins-restaurants/videos/jagd-auf-fertigkost-im-elsthal-teil-2-ganze-folge',
 101             'info_dict': {
 102                 'id': '2992323',
 103                 'ext': 'mp4',
 104                 'title': 'Jagd auf Fertigkost im Elsthal - Teil 2',
 105                 'description': 'md5:2669cde3febe9bce13904f701e774eb6',
 106                 'upload_date': '20141014',
 107                 'duration': 2410.44,
 108             },
 109             'params': {
 110                 # rtmp download
 111                 'skip_download': True,
 112             },
 113         },
 114         {
 115             'url': 'http://www.ran.de/fussball/bundesliga/video/schalke-toennies-moechte-raul-zurueck-ganze-folge',
 116             'info_dict': {
 117                 'id': '3004256',
 118                 'ext': 'mp4',
 119                 'title': 'Schalke: Tönnies möchte Raul zurück',
 120                 'description': 'md5:4b5b271d9bcde223b54390754c8ece3f',
 121                 'upload_date': '20140226',
 122                 'duration': 228.96,
 123             },
 124             'params': {
 125                 # rtmp download
 126                 'skip_download': True,
 127             },
 128         },
 129         {
 130             'url': 'http://www.the-voice-of-germany.de/video/31-andreas-kuemmert-rocket-man-clip',
 131             'info_dict': {
 132                 'id': '2572814',
 133                 'ext': 'mp4',
 134                 'title': 'Andreas Kümmert: Rocket Man',
 135                 'description': 'md5:6ddb02b0781c6adf778afea606652e38',
 136                 'upload_date': '20131017',
 137                 'duration': 469.88,
 138             },
 139             'params': {
 140                 # rtmp download
 141                 'skip_download': True,
 142             },
 143         },
 144         {
 145             'url': 'http://www.fem.com/wellness/videos/wellness-video-clip-kurztripps-zum-valentinstag.html',
 146             'info_dict': {
 147                 'id': '2156342',
 148                 'ext': 'mp4',
 149                 'title': 'Kurztrips zum Valentinstag',
 150                 'description': 'Romantischer Kurztrip zum Valentinstag? Wir verraten, was sich hier wirklich lohnt.',
 151                 'duration': 307.24,
 152             },
 153             'params': {
 154                 # rtmp download
 155                 'skip_download': True,
 156             },
 157         },
 158         {
 159             'url': 'http://www.prosieben.de/tv/joko-gegen-klaas/videos/playlists/episode-8-ganze-folge-playlist',
 160             'info_dict': {
 161                 'id': '439664',
 162                 'title': 'Episode 8 - Ganze Folge - Playlist',
 163                 'description': 'md5:63b8963e71f481782aeea877658dec84',
 164             },
 165             'playlist_count': 2,
 166         },
 167     ]
 168
 169     _CLIPID_REGEXES = [
 170         r'"clip_id"\s*:\s+"(\d+)"',
 171         r'clipid: "(\d+)"',
 172         r'clip[iI]d=(\d+)',
 173         r"'itemImageUrl'\s*:\s*'/dynamic/thumbnails/full/\d+/(\d+)",
 174     ]
 175     _TITLE_REGEXES = [
 176         r'<h2 class="subtitle" itemprop="name">\s*(.+?)</h2>',
 177         r'<header class="clearfix">\s*<h3>(.+?)</h3>',
 178         r'<!-- start video -->\s*<h1>(.+?)</h1>',
 179         r'<h1 class="att-name">\s*(.+?)</h1>',
 180     ]
 181     _DESCRIPTION_REGEXES = [
 182         r'<p itemprop="description">\s*(.+?)</p>',
 183         r'<div class="videoDecription">\s*<p><strong>Beschreibung</strong>: (.+?)</p>',
 184         r'<div class="g-plusone" data-size="medium"></div>\s*</div>\s*</header>\s*(.+?)\s*<footer>',
 185         r'<p class="att-description">\s*(.+?)\s*</p>',
 186     ]
 187     _UPLOAD_DATE_REGEXES = [
 188         r'<meta property="og:published_time" content="(.+?)">',
 189         r'<span>\s*(\d{2}\.\d{2}\.\d{4} \d{2}:\d{2}) \|\s*<span itemprop="duration"',
 190         r'<footer>\s*(\d{2}\.\d{2}\.\d{4}) \d{2}:\d{2} Uhr',
 191         r'<span style="padding-left: 4px;line-height:20px; color:#404040">(\d{2}\.\d{2}\.\d{4})</span>',
 192         r'(\d{2}\.\d{2}\.\d{4}) \| \d{2}:\d{2} Min<br/>',
 193     ]
 194     _PAGE_TYPE_REGEXES = [
 195         r'<meta name="page_type" content="([^"]+)">',
 196         r"'itemType'\s*:\s*'([^']*)'",
 197     ]
 198     _PLAYLIST_ID_REGEXES = [
 199         r'content[iI]d=(\d+)',
 200         r"'itemId'\s*:\s*'([^']*)'",
 201     ]
 202     _PLAYLIST_CLIP_REGEXES = [
 203         r'(?s)data-qvt=.+?<a href="([^"]+)"',
 204     ]
 205
 206     def _extract_clip(self, url, webpage):
 207         clip_id = self._html_search_regex(self._CLIPID_REGEXES, webpage, 'clip id')
 208
 209         access_token = 'testclient'
 210         client_name = 'kolibri-1.2.5'
 211         client_location = url
 212
 213         videos_api_url = 'http://vas.sim-technik.de/vas/live/v2/videos?%s' % compat_urllib_parse.urlencode({
 214             'access_token': access_token,
 215             'client_location': client_location,
 216             'client_name': client_name,
 217             'ids': clip_id,
 218         })
 219
 220         videos = self._download_json(videos_api_url, clip_id, 'Downloading videos JSON')
 221
 222         duration = float(videos[0]['duration'])
 223         source_ids = [source['id'] for source in videos[0]['sources']]
 224         source_ids_str = ','.join(map(str, source_ids))
 225
 226         g = '01!8d8F_)r9]4s[qeuXfP%'
 227
 228         client_id = g[:2] + sha1(''.join([clip_id, g, access_token, client_location, g, client_name])
 229                                  .encode('utf-8')).hexdigest()
 230
 231         sources_api_url = 'http://vas.sim-technik.de/vas/live/v2/videos/%s/sources?%s' % (clip_id, compat_urllib_parse.urlencode({
 232             'access_token': access_token,
 233             'client_id': client_id,
 234             'client_location': client_location,
 235             'client_name': client_name,
 236         }))
 237
 238         sources = self._download_json(sources_api_url, clip_id, 'Downloading sources JSON')
 239         server_id = sources['server_id']
 240
 241         client_id = g[:2] + sha1(''.join([g, clip_id, access_token, server_id,
 242                                           client_location, source_ids_str, g, client_name])
 243                                  .encode('utf-8')).hexdigest()
 244
 245         url_api_url = 'http://vas.sim-technik.de/vas/live/v2/videos/%s/sources/url?%s' % (clip_id, compat_urllib_parse.urlencode({
 246             'access_token': access_token,
 247             'client_id': client_id,
 248             'client_location': client_location,
 249             'client_name': client_name,
 250             'server_id': server_id,
 251             'source_ids': source_ids_str,
 252         }))
 253
 254         urls = self._download_json(url_api_url, clip_id, 'Downloading urls JSON')
 255
 256         title = self._html_search_regex(self._TITLE_REGEXES, webpage, 'title')
 257         description = self._html_search_regex(self._DESCRIPTION_REGEXES, webpage, 'description', fatal=False)
 258         thumbnail = self._og_search_thumbnail(webpage)
 259
 260         upload_date = unified_strdate(self._html_search_regex(
 261             self._UPLOAD_DATE_REGEXES, webpage, 'upload date', default=None))
 262
 263         formats = []
 264
 265         urls_sources = urls['sources']
 266         if isinstance(urls_sources, dict):
 267             urls_sources = urls_sources.values()
 268
 269         def fix_bitrate(bitrate):
 270             bitrate = int_or_none(bitrate)
 271             if not bitrate:
 272                 return None
 273             return (bitrate // 1000) if bitrate % 1000 == 0 else bitrate
 274
 275         for source in urls_sources:
 276             protocol = source['protocol']
 277             if protocol == 'rtmp' or protocol == 'rtmpe':
 278                 mobj = re.search(r'^(?P<url>rtmpe?://[^/]+/(?P<app>[^/]+))/(?P<playpath>.+)$', source['url'])
 279                 if not mobj:
 280                     continue
 281                 formats.append({
 282                     'url': mobj.group('url'),
 283                     'app': mobj.group('app'),
 284                     'play_path': mobj.group('playpath'),
 285                     'player_url': 'http://livepassdl.conviva.com/hf/ver/2.79.0.17083/LivePassModuleMain.swf',
 286                     'page_url': 'http://www.prosieben.de',
 287                     'vbr': fix_bitrate(source['bitrate']),
 288                     'ext': 'mp4',
 289                     'format_id': '%s_%s' % (source['cdn'], source['bitrate']),
 290                 })
 291             else:
 292                 formats.append({
 293                     'url': source['url'],
 294                     'vbr': fix_bitrate(source['bitrate']),
 295                 })
 296
 297         self._sort_formats(formats)
 298
 299         return {
 300             'id': clip_id,
 301             'title': title,
 302             'description': description,
 303             'thumbnail': thumbnail,
 304             'upload_date': upload_date,
 305             'duration': duration,
 306             'formats': formats,
 307         }
 308
 309     def _extract_playlist(self, url, webpage):
 310         playlist_id = self._html_search_regex(
 311             self._PLAYLIST_ID_REGEXES, webpage, 'playlist id')
 312         for regex in self._PLAYLIST_CLIP_REGEXES:
 313             playlist_clips = re.findall(regex, webpage)
 314             if playlist_clips:
 315                 title = self._html_search_regex(
 316                     self._TITLE_REGEXES, webpage, 'title')
 317                 description = self._html_search_regex(
 318                     self._DESCRIPTION_REGEXES, webpage, 'description', fatal=False)
 319                 entries = [
 320                     self.url_result(
 321                         re.match('(.+?//.+?)/', url).group(1) + clip_path,
 322                         'ProSiebenSat1')
 323                     for clip_path in playlist_clips]
 324                 return self.playlist_result(entries, playlist_id, title, description)
 325
 326     def _real_extract(self, url):
 327         video_id = self._match_id(url)
 328         webpage = self._download_webpage(url, video_id)
 329         page_type = self._search_regex(
 330             self._PAGE_TYPE_REGEXES, webpage,
 331             'page type', default='clip').lower()
 332         if page_type == 'clip':
 333             return self._extract_clip(url, webpage)
 334         elif page_type == 'playlist':
 335             return self._extract_playlist(url, webpage)