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     clean_html,
  12     RegexNotFoundError,
  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': '20140225',
  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': '20140225',
 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': 'md5:8ba6301e70351ae0bedf8da00f7ba528',
 150                 'upload_date': '20130206',
 151                 'duration': 307.24,
 152             },
 153             'params': {
 154                 # rtmp download
 155                 'skip_download': True,
 156             },
 157         },
 158     ]
 159
 160     _CLIPID_REGEXES = [
 161         r'"clip_id"\s*:\s+"(\d+)"',
 162         r'clipid: "(\d+)"',
 163         r'clipId=(\d+)',
 164     ]
 165     _TITLE_REGEXES = [
 166         r'<h2 class="subtitle" itemprop="name">\s*(.+?)</h2>',
 167         r'<header class="clearfix">\s*<h3>(.+?)</h3>',
 168         r'<!-- start video -->\s*<h1>(.+?)</h1>',
 169         r'<div class="ep-femvideos-pi4-video-txt">\s*<h2>(.+?)</h2>',
 170     ]
 171     _DESCRIPTION_REGEXES = [
 172         r'<p itemprop="description">\s*(.+?)</p>',
 173         r'<div class="videoDecription">\s*<p><strong>Beschreibung</strong>: (.+?)</p>',
 174         r'<div class="g-plusone" data-size="medium"></div>\s*</div>\s*</header>\s*(.+?)\s*<footer>',
 175         r'<p>(.+?)</p>\s*<div class="ep-femvideos-pi4-video-footer">',
 176     ]
 177     _UPLOAD_DATE_REGEXES = [
 178         r'<meta property="og:published_time" content="(.+?)">',
 179         r'<span>\s*(\d{2}\.\d{2}\.\d{4} \d{2}:\d{2}) \|\s*<span itemprop="duration"',
 180         r'<footer>\s*(\d{2}\.\d{2}\.\d{4}) \d{2}:\d{2} Uhr',
 181         r'<span style="padding-left: 4px;line-height:20px; color:#404040">(\d{2}\.\d{2}\.\d{4})</span>',
 182         r'(\d{2}\.\d{2}\.\d{4}) \| \d{2}:\d{2} Min<br/>',
 183     ]
 184
 185     def _real_extract(self, url):
 186         mobj = re.match(self._VALID_URL, url)
 187         video_id = mobj.group('id')
 188
 189         page = self._download_webpage(url, video_id, 'Downloading page')
 190
 191         def extract(patterns, name, page, fatal=False):
 192             for pattern in patterns:
 193                 mobj = re.search(pattern, page)
 194                 if mobj:
 195                     return clean_html(mobj.group(1))
 196             if fatal:
 197                 raise RegexNotFoundError(u'Unable to extract %s' % name)
 198             return None
 199
 200         clip_id = extract(self._CLIPID_REGEXES, 'clip id', page, fatal=True)
 201
 202         access_token = 'testclient'
 203         client_name = 'kolibri-1.2.5'
 204         client_location = url
 205
 206         videos_api_url = 'http://vas.sim-technik.de/vas/live/v2/videos?%s' % compat_urllib_parse.urlencode({
 207             'access_token': access_token,
 208             'client_location': client_location,
 209             'client_name': client_name,
 210             'ids': clip_id,
 211         })
 212
 213         videos = self._download_json(videos_api_url, clip_id, 'Downloading videos JSON')
 214
 215         duration = float(videos[0]['duration'])
 216         source_ids = [source['id'] for source in videos[0]['sources']]
 217         source_ids_str = ','.join(map(str, source_ids))
 218
 219         g = '01!8d8F_)r9]4s[qeuXfP%'
 220
 221         client_id = g[:2] + sha1(''.join([clip_id, g, access_token, client_location, g, client_name])
 222                                  .encode('utf-8')).hexdigest()
 223
 224         sources_api_url = 'http://vas.sim-technik.de/vas/live/v2/videos/%s/sources?%s' % (clip_id, compat_urllib_parse.urlencode({
 225             'access_token': access_token,
 226             'client_id': client_id,
 227             'client_location': client_location,
 228             'client_name': client_name,
 229         }))
 230
 231         sources = self._download_json(sources_api_url, clip_id, 'Downloading sources JSON')
 232         server_id = sources['server_id']
 233
 234         client_id = g[:2] + sha1(''.join([g, clip_id, access_token, server_id,
 235                                           client_location, source_ids_str, g, client_name])
 236                                  .encode('utf-8')).hexdigest()
 237
 238         url_api_url = 'http://vas.sim-technik.de/vas/live/v2/videos/%s/sources/url?%s' % (clip_id, compat_urllib_parse.urlencode({
 239             'access_token': access_token,
 240             'client_id': client_id,
 241             'client_location': client_location,
 242             'client_name': client_name,
 243             'server_id': server_id,
 244             'source_ids': source_ids_str,
 245         }))
 246
 247         urls = self._download_json(url_api_url, clip_id, 'Downloading urls JSON')
 248
 249         title = extract(self._TITLE_REGEXES, 'title', page, fatal=True)
 250         description = extract(self._DESCRIPTION_REGEXES, 'description', page)
 251         thumbnail = self._og_search_thumbnail(page)
 252
 253         upload_date = extract(self._UPLOAD_DATE_REGEXES, 'upload date', page)
 254         if upload_date:
 255             upload_date = unified_strdate(upload_date)
 256
 257         formats = []
 258
 259         urls_sources = urls['sources']
 260         if isinstance(urls_sources, dict):
 261             urls_sources = urls_sources.values()
 262
 263         def fix_bitrate(bitrate):
 264             return bitrate / 1000 if bitrate % 1000 == 0 else bitrate
 265
 266         for source in urls_sources:
 267             protocol = source['protocol']
 268             if protocol == 'rtmp' or protocol == 'rtmpe':
 269                 mobj = re.search(r'^(?P<url>rtmpe?://[^/]+/(?P<app>[^/]+))/(?P<playpath>.+)$', source['url'])
 270                 if not mobj:
 271                     continue
 272                 formats.append({
 273                     'url': mobj.group('url'),
 274                     'app': mobj.group('app'),
 275                     'play_path': mobj.group('playpath'),
 276                     'player_url': 'http://livepassdl.conviva.com/hf/ver/2.79.0.17083/LivePassModuleMain.swf',
 277                     'page_url': 'http://www.prosieben.de',
 278                     'vbr': fix_bitrate(source['bitrate']),
 279                     'ext': 'mp4',
 280                     'format_id': '%s_%s' % (source['cdn'], source['bitrate']),
 281                 })
 282             else:
 283                 formats.append({
 284                     'url': source['url'],
 285                     'vbr': fix_bitrate(source['bitrate']),
 286                 })
 287
 288         self._sort_formats(formats)
 289
 290         return {
 291             'id': clip_id,
 292             'title': title,
 293             'description': description,
 294             'thumbnail': thumbnail,
 295             'upload_date': upload_date,
 296             'duration': duration,
 297             'formats': formats,
 298         }