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