Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/viki.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from ..compat import compat_urlparse
   6 from ..utils import (
   7     ExtractorError,
   8     unescapeHTML,
   9     unified_strdate,
  10     US_RATINGS,
  11 )
  12 from .common import InfoExtractor
  13
  14
  15 class VikiIE(InfoExtractor):
  16     IE_NAME = 'viki'
  17
  18     _VALID_URL = r'^https?://(?:www\.)?viki\.com/videos/(?P<id>[0-9]+v)'
  19     _TEST = {
  20         'url': 'http://www.viki.com/videos/1023585v-heirs-episode-14',
  21         'info_dict': {
  22             'id': '1023585v',
  23             'ext': 'mp4',
  24             'title': 'Heirs Episode 14',
  25             'uploader': 'SBS',
  26             'description': 'md5:c4b17b9626dd4b143dcc4d855ba3474e',
  27             'upload_date': '20131121',
  28             'age_limit': 13,
  29         },
  30         'skip': 'Blocked in the US',
  31     }
  32
  33     def _real_extract(self, url):
  34         video_id = self._match_id(url)
  35
  36         webpage = self._download_webpage(url, video_id)
  37         title = self._og_search_title(webpage)
  38         description = self._og_search_description(webpage)
  39         thumbnail = self._og_search_thumbnail(webpage)
  40
  41         uploader_m = re.search(
  42             r'<strong>Broadcast Network: </strong>\s*([^<]*)<', webpage)
  43         if uploader_m is None:
  44             uploader = None
  45         else:
  46             uploader = uploader_m.group(1).strip()
  47
  48         rating_str = self._html_search_regex(
  49             r'<strong>Rating: </strong>\s*([^<]*)<', webpage,
  50             'rating information', default='').strip()
  51         age_limit = US_RATINGS.get(rating_str)
  52
  53         info_url = 'http://www.viki.com/player5_fragment/%s?action=show&controller=videos' % video_id
  54         info_webpage = self._download_webpage(
  55             info_url, video_id, note='Downloading info page')
  56         if re.match(r'\s*<div\s+class="video-error', info_webpage):
  57             raise ExtractorError(
  58                 'Video %s is blocked from your location.' % video_id,
  59                 expected=True)
  60         video_url = self._html_search_regex(
  61             r'<source[^>]+src="([^"]+)"', info_webpage, 'video URL')
  62
  63         upload_date_str = self._html_search_regex(
  64             r'"created_at":"([^"]+)"', info_webpage, 'upload date')
  65         upload_date = (
  66             unified_strdate(upload_date_str)
  67             if upload_date_str is not None
  68             else None
  69         )
  70
  71         # subtitles
  72         video_subtitles = self.extract_subtitles(video_id, info_webpage)
  73
  74         return {
  75             'id': video_id,
  76             'title': title,
  77             'url': video_url,
  78             'description': description,
  79             'thumbnail': thumbnail,
  80             'age_limit': age_limit,
  81             'uploader': uploader,
  82             'subtitles': video_subtitles,
  83             'upload_date': upload_date,
  84         }
  85
  86     def _get_subtitles(self, video_id, info_webpage):
  87         res = {}
  88         for sturl_html in re.findall(r'<track src="([^"]+)"', info_webpage):
  89             sturl = unescapeHTML(sturl_html)
  90             m = re.search(r'/(?P<lang>[a-z]+)\.vtt', sturl)
  91             if not m:
  92                 continue
  93             res[m.group('lang')] = [{
  94                 'url': compat_urlparse.urljoin('http://www.viki.com', sturl),
  95                 'ext': 'vtt',
  96             }]
  97         return res