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