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