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     ExtractorError,
  13     determine_ext,
  14     float_or_none,
  15     int_or_none,
  16     unified_strdate,
  17 )
  18
  19
  20 class ProSiebenSat1IE(InfoExtractor):
  21     IE_NAME = 'prosiebensat1'
  22     IE_DESC = 'ProSiebenSat.1 Digital'
  23     _VALID_URL = r'https?://(?:www\.)?(?:(?:prosieben|prosiebenmaxx|sixx|sat1|kabeleins|the-voice-of-germany)\.(?:de|at|ch)|ran\.de|fem\.com)/(?P<id>.+)'
  24
  25     _TESTS = [
  26         {
  27             # Tests changes introduced in https://github.com/rg3/youtube-dl/pull/6242
  28             # in response to fixing https://github.com/rg3/youtube-dl/issues/6215:
  29             # - malformed f4m manifest support
  30             # - proper handling of URLs starting with `https?://` in 2.0 manifests
  31             # - recursive child f4m manifests extraction
  32             'url': 'http://www.prosieben.de/tv/circus-halligalli/videos/218-staffel-2-episode-18-jahresrueckblick-ganze-folge',
  33             'info_dict': {
  34                 'id': '2104602',
  35                 'ext': 'mp4',
  36                 'title': 'Episode 18 - Staffel 2',
  37                 'description': 'md5:8733c81b702ea472e069bc48bb658fc1',
  38                 'upload_date': '20131231',
  39                 'duration': 5845.04,
  40             },
  41             'params': {
  42                 # rtmp download
  43                 'skip_download': True,
  44             },
  45         },
  46         {
  47             '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',
  48             'info_dict': {
  49                 'id': '2570327',
  50                 'ext': 'mp4',
  51                 'title': 'Lady-Umstyling für Audrina',
  52                 'description': 'md5:4c16d0c17a3461a0d43ea4084e96319d',
  53                 'upload_date': '20131014',
  54                 'duration': 606.76,
  55             },
  56             'params': {
  57                 # rtmp download
  58                 'skip_download': True,
  59             },
  60             'skip': 'Seems to be broken',
  61         },
  62         {
  63             'url': 'http://www.prosiebenmaxx.de/tv/experience/video/144-countdown-fuer-die-autowerkstatt-ganze-folge',
  64             'info_dict': {
  65                 'id': '2429369',
  66                 'ext': 'mp4',
  67                 'title': 'Countdown für die Autowerkstatt',
  68                 'description': 'md5:809fc051a457b5d8666013bc40698817',
  69                 'upload_date': '20140223',
  70                 'duration': 2595.04,
  71             },
  72             'params': {
  73                 # rtmp download
  74                 'skip_download': True,
  75             },
  76         },
  77         {
  78             'url': 'http://www.sixx.de/stars-style/video/sexy-laufen-in-ugg-boots-clip',
  79             'info_dict': {
  80                 'id': '2904997',
  81                 'ext': 'mp4',
  82                 'title': 'Sexy laufen in Ugg Boots',
  83                 'description': 'md5:edf42b8bd5bc4e5da4db4222c5acb7d6',
  84                 'upload_date': '20140122',
  85                 'duration': 245.32,
  86             },
  87             'params': {
  88                 # rtmp download
  89                 'skip_download': True,
  90             },
  91         },
  92         {
  93             'url': 'http://www.sat1.de/film/der-ruecktritt/video/im-interview-kai-wiesinger-clip',
  94             'info_dict': {
  95                 'id': '2906572',
  96                 'ext': 'mp4',
  97                 'title': 'Im Interview: Kai Wiesinger',
  98                 'description': 'md5:e4e5370652ec63b95023e914190b4eb9',
  99                 'upload_date': '20140203',
 100                 'duration': 522.56,
 101             },
 102             'params': {
 103                 # rtmp download
 104                 'skip_download': True,
 105             },
 106         },
 107         {
 108             'url': 'http://www.kabeleins.de/tv/rosins-restaurants/videos/jagd-auf-fertigkost-im-elsthal-teil-2-ganze-folge',
 109             'info_dict': {
 110                 'id': '2992323',
 111                 'ext': 'mp4',
 112                 'title': 'Jagd auf Fertigkost im Elsthal - Teil 2',
 113                 'description': 'md5:2669cde3febe9bce13904f701e774eb6',
 114                 'upload_date': '20141014',
 115                 'duration': 2410.44,
 116             },
 117             'params': {
 118                 # rtmp download
 119                 'skip_download': True,
 120             },
 121         },
 122         {
 123             'url': 'http://www.ran.de/fussball/bundesliga/video/schalke-toennies-moechte-raul-zurueck-ganze-folge',
 124             'info_dict': {
 125                 'id': '3004256',
 126                 'ext': 'mp4',
 127                 'title': 'Schalke: Tönnies möchte Raul zurück',
 128                 'description': 'md5:4b5b271d9bcde223b54390754c8ece3f',
 129                 'upload_date': '20140226',
 130                 'duration': 228.96,
 131             },
 132             'params': {
 133                 # rtmp download
 134                 'skip_download': True,
 135             },
 136         },
 137         {
 138             'url': 'http://www.the-voice-of-germany.de/video/31-andreas-kuemmert-rocket-man-clip',
 139             'info_dict': {
 140                 'id': '2572814',
 141                 'ext': 'mp4',
 142                 'title': 'Andreas Kümmert: Rocket Man',
 143                 'description': 'md5:6ddb02b0781c6adf778afea606652e38',
 144                 'upload_date': '20131017',
 145                 'duration': 469.88,
 146             },
 147             'params': {
 148                 # rtmp download
 149                 'skip_download': True,
 150             },
 151         },
 152         {
 153             'url': 'http://www.fem.com/wellness/videos/wellness-video-clip-kurztripps-zum-valentinstag.html',
 154             'info_dict': {
 155                 'id': '2156342',
 156                 'ext': 'mp4',
 157                 'title': 'Kurztrips zum Valentinstag',
 158                 'description': 'Romantischer Kurztrip zum Valentinstag? Wir verraten, was sich hier wirklich lohnt.',
 159                 'duration': 307.24,
 160             },
 161             'params': {
 162                 # rtmp download
 163                 'skip_download': True,
 164             },
 165         },
 166         {
 167             'url': 'http://www.prosieben.de/tv/joko-gegen-klaas/videos/playlists/episode-8-ganze-folge-playlist',
 168             'info_dict': {
 169                 'id': '439664',
 170                 'title': 'Episode 8 - Ganze Folge - Playlist',
 171                 'description': 'md5:63b8963e71f481782aeea877658dec84',
 172             },
 173             'playlist_count': 2,
 174         },
 175     ]
 176
 177     _CLIPID_REGEXES = [
 178         r'"clip_id"\s*:\s+"(\d+)"',
 179         r'clipid: "(\d+)"',
 180         r'clip[iI]d=(\d+)',
 181         r"'itemImageUrl'\s*:\s*'/dynamic/thumbnails/full/\d+/(\d+)",
 182     ]
 183     _TITLE_REGEXES = [
 184         r'<h2 class="subtitle" itemprop="name">\s*(.+?)</h2>',
 185         r'<header class="clearfix">\s*<h3>(.+?)</h3>',
 186         r'<!-- start video -->\s*<h1>(.+?)</h1>',
 187         r'<h1 class="att-name">\s*(.+?)</h1>',
 188         r'<header class="module_header">\s*<h2>([^<]+)</h2>\s*</header>',
 189     ]
 190     _DESCRIPTION_REGEXES = [
 191         r'<p itemprop="description">\s*(.+?)</p>',
 192         r'<div class="videoDecription">\s*<p><strong>Beschreibung</strong>: (.+?)</p>',
 193         r'<div class="g-plusone" data-size="medium"></div>\s*</div>\s*</header>\s*(.+?)\s*<footer>',
 194         r'<p class="att-description">\s*(.+?)\s*</p>',
 195     ]
 196     _UPLOAD_DATE_REGEXES = [
 197         r'<meta property="og:published_time" content="(.+?)">',
 198         r'<span>\s*(\d{2}\.\d{2}\.\d{4} \d{2}:\d{2}) \|\s*<span itemprop="duration"',
 199         r'<footer>\s*(\d{2}\.\d{2}\.\d{4}) \d{2}:\d{2} Uhr',
 200         r'<span style="padding-left: 4px;line-height:20px; color:#404040">(\d{2}\.\d{2}\.\d{4})</span>',
 201         r'(\d{2}\.\d{2}\.\d{4}) \| \d{2}:\d{2} Min<br/>',
 202     ]
 203     _PAGE_TYPE_REGEXES = [
 204         r'<meta name="page_type" content="([^"]+)">',
 205         r"'itemType'\s*:\s*'([^']*)'",
 206     ]
 207     _PLAYLIST_ID_REGEXES = [
 208         r'content[iI]d=(\d+)',
 209         r"'itemId'\s*:\s*'([^']*)'",
 210     ]
 211     _PLAYLIST_CLIP_REGEXES = [
 212         r'(?s)data-qvt=.+?<a href="([^"]+)"',
 213     ]
 214
 215     def _extract_clip(self, url, webpage):
 216         clip_id = self._html_search_regex(self._CLIPID_REGEXES, webpage, 'clip id')
 217
 218         access_token = 'prosieben'
 219         client_name = 'kolibri-2.0.19-splec4'
 220         client_location = url
 221
 222         videos_api_url = 'http://vas.sim-technik.de/vas/live/v2/videos?%s' % compat_urllib_parse.urlencode({
 223             'access_token': access_token,
 224             'client_location': client_location,
 225             'client_name': client_name,
 226             'ids': clip_id,
 227         })
 228
 229         video = self._download_json(videos_api_url, clip_id, 'Downloading videos JSON')[0]
 230
 231         if video.get('is_protected') is True:
 232             raise ExtractorError('This video is DRM protected.', expected=True)
 233
 234         duration = float_or_none(video.get('duration'))
 235         source_ids = [source['id'] for source in video['sources']]
 236         source_ids_str = ','.join(map(str, source_ids))
 237
 238         g = '01!8d8F_)r9]4s[qeuXfP%'
 239
 240         client_id = g[:2] + sha1(''.join([clip_id, g, access_token, client_location, g, client_name])
 241                                  .encode('utf-8')).hexdigest()
 242
 243         sources_api_url = 'http://vas.sim-technik.de/vas/live/v2/videos/%s/sources?%s' % (clip_id, compat_urllib_parse.urlencode({
 244             'access_token': access_token,
 245             'client_id': client_id,
 246             'client_location': client_location,
 247             'client_name': client_name,
 248         }))
 249
 250         sources = self._download_json(sources_api_url, clip_id, 'Downloading sources JSON')
 251         server_id = sources['server_id']
 252
 253         client_id = g[:2] + sha1(''.join([g, clip_id, access_token, server_id,
 254                                           client_location, source_ids_str, g, client_name])
 255                                  .encode('utf-8')).hexdigest()
 256
 257         url_api_url = 'http://vas.sim-technik.de/vas/live/v2/videos/%s/sources/url?%s' % (clip_id, compat_urllib_parse.urlencode({
 258             'access_token': access_token,
 259             'client_id': client_id,
 260             'client_location': client_location,
 261             'client_name': client_name,
 262             'server_id': server_id,
 263             'source_ids': source_ids_str,
 264         }))
 265
 266         urls = self._download_json(url_api_url, clip_id, 'Downloading urls JSON')
 267
 268         title = self._html_search_regex(self._TITLE_REGEXES, webpage, 'title')
 269         description = self._html_search_regex(self._DESCRIPTION_REGEXES, webpage, 'description', fatal=False)
 270         thumbnail = self._og_search_thumbnail(webpage)
 271
 272         upload_date = unified_strdate(self._html_search_regex(
 273             self._UPLOAD_DATE_REGEXES, webpage, 'upload date', default=None))
 274
 275         formats = []
 276
 277         urls_sources = urls['sources']
 278         if isinstance(urls_sources, dict):
 279             urls_sources = urls_sources.values()
 280
 281         def fix_bitrate(bitrate):
 282             bitrate = int_or_none(bitrate)
 283             if not bitrate:
 284                 return None
 285             return (bitrate // 1000) if bitrate % 1000 == 0 else bitrate
 286
 287         for source in urls_sources:
 288             protocol = source['protocol']
 289             source_url = source['url']
 290             if protocol == 'rtmp' or protocol == 'rtmpe':
 291                 mobj = re.search(r'^(?P<url>rtmpe?://[^/]+)/(?P<path>.+)$', source_url)
 292                 if not mobj:
 293                     continue
 294                 path = mobj.group('path')
 295                 mp4colon_index = path.rfind('mp4:')
 296                 app = path[:mp4colon_index]
 297                 play_path = path[mp4colon_index:]
 298                 formats.append({
 299                     'url': '%s/%s' % (mobj.group('url'), app),
 300                     'app': app,
 301                     'play_path': play_path,
 302                     'player_url': 'http://livepassdl.conviva.com/hf/ver/2.79.0.17083/LivePassModuleMain.swf',
 303                     'page_url': 'http://www.prosieben.de',
 304                     'vbr': fix_bitrate(source['bitrate']),
 305                     'ext': 'mp4',
 306                     'format_id': '%s_%s' % (source['cdn'], source['bitrate']),
 307                 })
 308             elif 'f4mgenerator' in source_url or determine_ext(source_url) == 'f4m':
 309                 formats.extend(self._extract_f4m_formats(source_url, clip_id))
 310             else:
 311                 formats.append({
 312                     'url': source_url,
 313                     'vbr': fix_bitrate(source['bitrate']),
 314                 })
 315
 316         self._sort_formats(formats)
 317
 318         return {
 319             'id': clip_id,
 320             'title': title,
 321             'description': description,
 322             'thumbnail': thumbnail,
 323             'upload_date': upload_date,
 324             'duration': duration,
 325             'formats': formats,
 326         }
 327
 328     def _extract_playlist(self, url, webpage):
 329         playlist_id = self._html_search_regex(
 330             self._PLAYLIST_ID_REGEXES, webpage, 'playlist id')
 331         for regex in self._PLAYLIST_CLIP_REGEXES:
 332             playlist_clips = re.findall(regex, webpage)
 333             if playlist_clips:
 334                 title = self._html_search_regex(
 335                     self._TITLE_REGEXES, webpage, 'title')
 336                 description = self._html_search_regex(
 337                     self._DESCRIPTION_REGEXES, webpage, 'description', fatal=False)
 338                 entries = [
 339                     self.url_result(
 340                         re.match('(.+?//.+?)/', url).group(1) + clip_path,
 341                         'ProSiebenSat1')
 342                     for clip_path in playlist_clips]
 343                 return self.playlist_result(entries, playlist_id, title, description)
 344
 345     def _real_extract(self, url):
 346         video_id = self._match_id(url)
 347         webpage = self._download_webpage(url, video_id)
 348         page_type = self._search_regex(
 349             self._PAGE_TYPE_REGEXES, webpage,
 350             'page type', default='clip').lower()
 351         if page_type == 'clip':
 352             return self._extract_clip(url, webpage)
 353         elif page_type == 'playlist':
 354             return self._extract_playlist(url, webpage)