_ Git - youtube-dl/blob - youtube_dl/extractor/viki.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from ..compat import (
   6     compat_urlparse,
   7     compat_urllib_request,
   8 )
   9 from ..utils import (
  10     ExtractorError,
  11     unescapeHTML,
  12     unified_strdate,
  13     US_RATINGS,
  14 )
  15 from .common import InfoExtractor
  16
  17
  18 class VikiIE(InfoExtractor):
  19     IE_NAME = 'viki'
  20
  21     # iPad2
  22     _USER_AGENT = 'Mozilla/5.0(iPad; U; CPU OS 4_3 like Mac OS X; en-us) AppleWebKit/533.17.9 (KHTML, like Gecko) Version/5.0.2 Mobile/8F191 Safari/6533.18.5'
  23
  24     _VALID_URL = r'^https?://(?:www\.)?viki\.com/videos/(?P<id>[0-9]+v)'
  25     _TESTS = [{
  26         'url': 'http://www.viki.com/videos/1023585v-heirs-episode-14',
  27         'info_dict': {
  28             'id': '1023585v',
  29             'ext': 'mp4',
  30             'title': 'Heirs Episode 14',
  31             'uploader': 'SBS',
  32             'description': 'md5:c4b17b9626dd4b143dcc4d855ba3474e',
  33             'upload_date': '20131121',
  34             'age_limit': 13,
  35         },
  36         'skip': 'Blocked in the US',
  37     }, {
  38         'url': 'http://www.viki.com/videos/1067139v-the-avengers-age-of-ultron-press-conference',
  39         'md5': 'ca6493e6f0a6ec07da9aa8d6304b4b2c',
  40         'info_dict': {
  41             'id': '1067139v',
  42             'ext': 'mp4',
  43             'description': 'md5:d70b2f9428f5488321bfe1db10d612ea',
  44             'upload_date': '20150430',
  45             'title': '\'The Avengers: Age of Ultron\' Press Conference',
  46         }
  47     }]
  48
  49     def _real_extract(self, url):
  50         video_id = self._match_id(url)
  51
  52         webpage = self._download_webpage(url, video_id)
  53         title = self._og_search_title(webpage)
  54         description = self._og_search_description(webpage)
  55         thumbnail = self._og_search_thumbnail(webpage)
  56
  57         uploader_m = re.search(
  58             r'<strong>Broadcast Network: </strong>\s*([^<]*)<', webpage)
  59         if uploader_m is None:
  60             uploader = None
  61         else:
  62             uploader = uploader_m.group(1).strip()
  63
  64         rating_str = self._html_search_regex(
  65             r'<strong>Rating: </strong>\s*([^<]*)<', webpage,
  66             'rating information', default='').strip()
  67         age_limit = US_RATINGS.get(rating_str)
  68
  69         req = compat_urllib_request.Request(
  70             'http://www.viki.com/player5_fragment/%s?action=show&controller=videos' % video_id)
  71         req.add_header('User-Agent', self._USER_AGENT)
  72         info_webpage = self._download_webpage(
  73             req, video_id, note='Downloading info page')
  74         if re.match(r'\s*<div\s+class="video-error', info_webpage):
  75             raise ExtractorError(
  76                 'Video %s is blocked from your location.' % video_id,
  77                 expected=True)
  78         video_url = self._html_search_regex(
  79             r'<source[^>]+src="([^"]+)"', info_webpage, 'video URL')
  80
  81         upload_date_str = self._html_search_regex(
  82             r'"created_at":"([^"]+)"', info_webpage, 'upload date')
  83         upload_date = (
  84             unified_strdate(upload_date_str)
  85             if upload_date_str is not None
  86             else None
  87         )
  88
  89         # subtitles
  90         video_subtitles = self.extract_subtitles(video_id, info_webpage)
  91
  92         return {
  93             'id': video_id,
  94             'title': title,
  95             'url': video_url,
  96             'description': description,
  97             'thumbnail': thumbnail,
  98             'age_limit': age_limit,
  99             'uploader': uploader,
 100             'subtitles': video_subtitles,
 101             'upload_date': upload_date,
 102         }
 103
 104     def _get_subtitles(self, video_id, info_webpage):
 105         res = {}
 106         for sturl_html in re.findall(r'<track src="([^"]+)"', info_webpage):
 107             sturl = unescapeHTML(sturl_html)
 108             m = re.search(r'/(?P<lang>[a-z]+)\.vtt', sturl)
 109             if not m:
 110                 continue
 111             res[m.group('lang')] = [{
 112                 'url': compat_urlparse.urljoin('http://www.viki.com', sturl),
 113                 'ext': 'vtt',
 114             }]
 115         return res