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