_ 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     clean_html,
  15 )
  16 from .common import InfoExtractor
  17
  18
  19 class VikiIE(InfoExtractor):
  20     IE_NAME = 'viki'
  21
  22     # iPad2
  23     _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'
  24
  25     _VALID_URL = r'^https?://(?:www\.)?viki\.com/videos/(?P<id>[0-9]+v)'
  26     _TESTS = [{
  27         'url': 'http://www.viki.com/videos/1023585v-heirs-episode-14',
  28         'info_dict': {
  29             'id': '1023585v',
  30             'ext': 'mp4',
  31             'title': 'Heirs Episode 14',
  32             'uploader': 'SBS',
  33             'description': 'md5:c4b17b9626dd4b143dcc4d855ba3474e',
  34             'upload_date': '20131121',
  35             'age_limit': 13,
  36         },
  37         'skip': 'Blocked in the US',
  38     }, {
  39         'url': 'http://www.viki.com/videos/1067139v-the-avengers-age-of-ultron-press-conference',
  40         'md5': 'ca6493e6f0a6ec07da9aa8d6304b4b2c',
  41         'info_dict': {
  42             'id': '1067139v',
  43             'ext': 'mp4',
  44             'description': 'md5:d70b2f9428f5488321bfe1db10d612ea',
  45             'upload_date': '20150430',
  46             'title': '\'The Avengers: Age of Ultron\' Press Conference',
  47         }
  48     }]
  49
  50     def _real_extract(self, url):
  51         video_id = self._match_id(url)
  52
  53         webpage = self._download_webpage(url, video_id)
  54         title = self._og_search_title(webpage)
  55         description = self._og_search_description(webpage)
  56         thumbnail = self._og_search_thumbnail(webpage)
  57
  58         uploader_m = re.search(
  59             r'<strong>Broadcast Network: </strong>\s*([^<]*)<', webpage)
  60         if uploader_m is None:
  61             uploader = None
  62         else:
  63             uploader = uploader_m.group(1).strip()
  64
  65         rating_str = self._html_search_regex(
  66             r'<strong>Rating: </strong>\s*([^<]*)<', webpage,
  67             'rating information', default='').strip()
  68         age_limit = US_RATINGS.get(rating_str)
  69
  70         req = compat_urllib_request.Request(
  71             'http://www.viki.com/player5_fragment/%s?action=show&controller=videos' % video_id)
  72         req.add_header('User-Agent', self._USER_AGENT)
  73         info_webpage = self._download_webpage(
  74             req, video_id, note='Downloading info page')
  75         err_msg = self._html_search_regex(r'<div[^>]+class="video-error[^>]+>(.+)</div>', info_webpage, 'error message', default=None)
  76         if err_msg:
  77             err_msg = clean_html(err_msg)
  78             if 'not available in your region' in err_msg:
  79                 raise ExtractorError(
  80                     'Video %s is blocked from your location.' % video_id,
  81                     expected=True)
  82             else:
  83                 raise ExtractorError('Viki said: ' + err_msg)
  84         video_url = self._html_search_regex(
  85             r'<source[^>]+src="([^"]+)"', info_webpage, 'video URL')
  86
  87         upload_date_str = self._html_search_regex(
  88             r'"created_at":"([^"]+)"', info_webpage, 'upload date')
  89         upload_date = (
  90             unified_strdate(upload_date_str)
  91             if upload_date_str is not None
  92             else None
  93         )
  94
  95         # subtitles
  96         video_subtitles = self.extract_subtitles(video_id, info_webpage)
  97
  98         return {
  99             'id': video_id,
 100             'title': title,
 101             'url': video_url,
 102             'description': description,
 103             'thumbnail': thumbnail,
 104             'age_limit': age_limit,
 105             'uploader': uploader,
 106             'subtitles': video_subtitles,
 107             'upload_date': upload_date,
 108         }
 109
 110     def _get_subtitles(self, video_id, info_webpage):
 111         res = {}
 112         for sturl_html in re.findall(r'<track src="([^"]+)"', info_webpage):
 113             sturl = unescapeHTML(sturl_html)
 114             m = re.search(r'/(?P<lang>[a-z]+)\.vtt', sturl)
 115             if not m:
 116                 continue
 117             res[m.group('lang')] = [{
 118                 'url': compat_urlparse.urljoin('http://www.viki.com', sturl),
 119                 'ext': 'vtt',
 120             }]
 121         return res