Raphaël G. Git Repositories - youtubedl/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 compat_urllib_parse_urlencode
   9 from ..utils import (
  10     ExtractorError,
  11     determine_ext,
  12     float_or_none,
  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|7tv)\.(?:de|at|ch)|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': 'flv',
  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': 'flv',
 140                 'title': 'Andreas Kümmert: Rocket Man',
 141                 'description': 'md5:6ddb02b0781c6adf778afea606652e38',
 142                 'upload_date': '20131017',
 143                 'duration': 469.88,
 144             },
 145             'params': {
 146                 'skip_download': True,
 147             },
 148         },
 149         {
 150             'url': 'http://www.fem.com/wellness/videos/wellness-video-clip-kurztripps-zum-valentinstag.html',
 151             'info_dict': {
 152                 'id': '2156342',
 153                 'ext': 'flv',
 154                 'title': 'Kurztrips zum Valentinstag',
 155                 'description': 'Romantischer Kurztrip zum Valentinstag? Nina Heinemann verrät, was sich hier wirklich lohnt.',
 156                 'duration': 307.24,
 157             },
 158             'params': {
 159                 'skip_download': True,
 160             },
 161         },
 162         {
 163             'url': 'http://www.prosieben.de/tv/joko-gegen-klaas/videos/playlists/episode-8-ganze-folge-playlist',
 164             'info_dict': {
 165                 'id': '439664',
 166                 'title': 'Episode 8 - Ganze Folge - Playlist',
 167                 'description': 'md5:63b8963e71f481782aeea877658dec84',
 168             },
 169             'playlist_count': 2,
 170         },
 171         {
 172             'url': 'http://www.7tv.de/circus-halligalli/615-best-of-circus-halligalli-ganze-folge',
 173             'info_dict': {
 174                 'id': '4187506',
 175                 'ext': 'flv',
 176                 'title': 'Best of Circus HalliGalli',
 177                 'description': 'md5:8849752efd90b9772c9db6fdf87fb9e9',
 178                 'upload_date': '20151229',
 179             },
 180             'params': {
 181                 'skip_download': True,
 182             },
 183         },
 184     ]
 185
 186     _CLIPID_REGEXES = [
 187         r'"clip_id"\s*:\s+"(\d+)"',
 188         r'clipid: "(\d+)"',
 189         r'clip[iI]d=(\d+)',
 190         r'clip[iI]d\s*=\s*["\'](\d+)',
 191         r"'itemImageUrl'\s*:\s*'/dynamic/thumbnails/full/\d+/(\d+)",
 192     ]
 193     _TITLE_REGEXES = [
 194         r'<h2 class="subtitle" itemprop="name">\s*(.+?)</h2>',
 195         r'<header class="clearfix">\s*<h3>(.+?)</h3>',
 196         r'<!-- start video -->\s*<h1>(.+?)</h1>',
 197         r'<h1 class="att-name">\s*(.+?)</h1>',
 198         r'<header class="module_header">\s*<h2>([^<]+)</h2>\s*</header>',
 199         r'<h2 class="video-title" itemprop="name">\s*(.+?)</h2>',
 200         r'<div[^>]+id="veeseoTitle"[^>]*>(.+?)</div>',
 201     ]
 202     _DESCRIPTION_REGEXES = [
 203         r'<p itemprop="description">\s*(.+?)</p>',
 204         r'<div class="videoDecription">\s*<p><strong>Beschreibung</strong>: (.+?)</p>',
 205         r'<div class="g-plusone" data-size="medium"></div>\s*</div>\s*</header>\s*(.+?)\s*<footer>',
 206         r'<p class="att-description">\s*(.+?)\s*</p>',
 207         r'<p class="video-description" itemprop="description">\s*(.+?)</p>',
 208         r'<div[^>]+id="veeseoDescription"[^>]*>(.+?)</div>',
 209     ]
 210     _UPLOAD_DATE_REGEXES = [
 211         r'<meta property="og:published_time" content="(.+?)">',
 212         r'<span>\s*(\d{2}\.\d{2}\.\d{4} \d{2}:\d{2}) \|\s*<span itemprop="duration"',
 213         r'<footer>\s*(\d{2}\.\d{2}\.\d{4}) \d{2}:\d{2} Uhr',
 214         r'<span style="padding-left: 4px;line-height:20px; color:#404040">(\d{2}\.\d{2}\.\d{4})</span>',
 215         r'(\d{2}\.\d{2}\.\d{4}) \| \d{2}:\d{2} Min<br/>',
 216     ]
 217     _PAGE_TYPE_REGEXES = [
 218         r'<meta name="page_type" content="([^"]+)">',
 219         r"'itemType'\s*:\s*'([^']*)'",
 220     ]
 221     _PLAYLIST_ID_REGEXES = [
 222         r'content[iI]d=(\d+)',
 223         r"'itemId'\s*:\s*'([^']*)'",
 224     ]
 225     _PLAYLIST_CLIP_REGEXES = [
 226         r'(?s)data-qvt=.+?<a href="([^"]+)"',
 227     ]
 228
 229     def _extract_clip(self, url, webpage):
 230         clip_id = self._html_search_regex(self._CLIPID_REGEXES, webpage, 'clip id')
 231
 232         access_token = 'prosieben'
 233         client_name = 'kolibri-2.0.19-splec4'
 234         client_location = url
 235
 236         videos_api_url = 'http://vas.sim-technik.de/vas/live/v2/videos?%s' % compat_urllib_parse_urlencode({
 237             'access_token': access_token,
 238             'client_location': client_location,
 239             'client_name': client_name,
 240             'ids': clip_id,
 241         })
 242
 243         video = self._download_json(videos_api_url, clip_id, 'Downloading videos JSON')[0]
 244
 245         if video.get('is_protected') is True:
 246             raise ExtractorError('This video is DRM protected.', expected=True)
 247
 248         duration = float_or_none(video.get('duration'))
 249         source_ids = [source['id'] for source in video['sources']]
 250         source_ids_str = ','.join(map(str, source_ids))
 251
 252         g = '01!8d8F_)r9]4s[qeuXfP%'
 253
 254         client_id = g[:2] + sha1(''.join([clip_id, g, access_token, client_location, g, client_name])
 255                                  .encode('utf-8')).hexdigest()
 256
 257         sources_api_url = 'http://vas.sim-technik.de/vas/live/v2/videos/%s/sources?%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         }))
 263
 264         sources = self._download_json(sources_api_url, clip_id, 'Downloading sources JSON')
 265         server_id = sources['server_id']
 266
 267         client_id = g[:2] + sha1(''.join([g, clip_id, access_token, server_id,
 268                                           client_location, source_ids_str, g, client_name])
 269                                  .encode('utf-8')).hexdigest()
 270
 271         url_api_url = 'http://vas.sim-technik.de/vas/live/v2/videos/%s/sources/url?%s' % (clip_id, compat_urllib_parse_urlencode({
 272             'access_token': access_token,
 273             'client_id': client_id,
 274             'client_location': client_location,
 275             'client_name': client_name,
 276             'server_id': server_id,
 277             'source_ids': source_ids_str,
 278         }))
 279
 280         urls = self._download_json(url_api_url, clip_id, 'Downloading urls JSON')
 281
 282         title = self._html_search_regex(self._TITLE_REGEXES, webpage, 'title')
 283         description = self._html_search_regex(self._DESCRIPTION_REGEXES, webpage, 'description', fatal=False)
 284         thumbnail = self._og_search_thumbnail(webpage)
 285
 286         upload_date = unified_strdate(self._html_search_regex(
 287             self._UPLOAD_DATE_REGEXES, webpage, 'upload date', default=None))
 288
 289         formats = []
 290
 291         urls_sources = urls['sources']
 292         if isinstance(urls_sources, dict):
 293             urls_sources = urls_sources.values()
 294
 295         def fix_bitrate(bitrate):
 296             bitrate = int_or_none(bitrate)
 297             if not bitrate:
 298                 return None
 299             return (bitrate // 1000) if bitrate % 1000 == 0 else bitrate
 300
 301         for source in urls_sources:
 302             protocol = source['protocol']
 303             source_url = source['url']
 304             if protocol == 'rtmp' or protocol == 'rtmpe':
 305                 mobj = re.search(r'^(?P<url>rtmpe?://[^/]+)/(?P<path>.+)$', source_url)
 306                 if not mobj:
 307                     continue
 308                 path = mobj.group('path')
 309                 mp4colon_index = path.rfind('mp4:')
 310                 app = path[:mp4colon_index]
 311                 play_path = path[mp4colon_index:]
 312                 formats.append({
 313                     'url': '%s/%s' % (mobj.group('url'), app),
 314                     'app': app,
 315                     'play_path': play_path,
 316                     'player_url': 'http://livepassdl.conviva.com/hf/ver/2.79.0.17083/LivePassModuleMain.swf',
 317                     'page_url': 'http://www.prosieben.de',
 318                     'vbr': fix_bitrate(source['bitrate']),
 319                     'ext': 'mp4',
 320                     'format_id': '%s_%s' % (source['cdn'], source['bitrate']),
 321                 })
 322             elif 'f4mgenerator' in source_url or determine_ext(source_url) == 'f4m':
 323                 formats.extend(self._extract_f4m_formats(source_url, clip_id))
 324             else:
 325                 formats.append({
 326                     'url': source_url,
 327                     'vbr': fix_bitrate(source['bitrate']),
 328                 })
 329
 330         self._sort_formats(formats)
 331
 332         return {
 333             'id': clip_id,
 334             'title': title,
 335             'description': description,
 336             'thumbnail': thumbnail,
 337             'upload_date': upload_date,
 338             'duration': duration,
 339             'formats': formats,
 340         }
 341
 342     def _extract_playlist(self, url, webpage):
 343         playlist_id = self._html_search_regex(
 344             self._PLAYLIST_ID_REGEXES, webpage, 'playlist id')
 345         for regex in self._PLAYLIST_CLIP_REGEXES:
 346             playlist_clips = re.findall(regex, webpage)
 347             if playlist_clips:
 348                 title = self._html_search_regex(
 349                     self._TITLE_REGEXES, webpage, 'title')
 350                 description = self._html_search_regex(
 351                     self._DESCRIPTION_REGEXES, webpage, 'description', fatal=False)
 352                 entries = [
 353                     self.url_result(
 354                         re.match('(.+?//.+?)/', url).group(1) + clip_path,
 355                         'ProSiebenSat1')
 356                     for clip_path in playlist_clips]
 357                 return self.playlist_result(entries, playlist_id, title, description)
 358
 359     def _real_extract(self, url):
 360         video_id = self._match_id(url)
 361         webpage = self._download_webpage(url, video_id)
 362         page_type = self._search_regex(
 363             self._PAGE_TYPE_REGEXES, webpage,
 364             'page type', default='clip').lower()
 365         if page_type == 'clip':
 366             return self._extract_clip(url, webpage)
 367         elif page_type == 'playlist':
 368             return self._extract_playlist(url, webpage)