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

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from ..utils import (
   6     ExtractorError,
   7     unescapeHTML,
   8     unified_strdate,
   9 )
  10 from .subtitles import SubtitlesInfoExtractor
  11
  12
  13 class VikiIE(SubtitlesInfoExtractor):
  14     IE_NAME = 'viki'
  15
  16     _VALID_URL = r'^https?://(?:www\.)?viki\.com/videos/(?P<id>[0-9]+v)'
  17     _TEST = {
  18         'url': 'http://www.viki.com/videos/1023585v-heirs-episode-14',
  19         'md5': 'a21454021c2646f5433514177e2caa5f',
  20         'info_dict': {
  21             'id': '1023585v',
  22             'ext': 'mp4',
  23             'title': 'Heirs Episode 14',
  24             'uploader': 'SBS',
  25             'description': 'md5:c4b17b9626dd4b143dcc4d855ba3474e',
  26             'upload_date': '20131121',
  27             'age_limit': 13,
  28         },
  29         'skip': 'Blocked in the US',
  30     }
  31
  32     def _real_extract(self, url):
  33         mobj = re.match(self._VALID_URL, url)
  34         video_id = mobj.group(1)
  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         RATINGS = {
  52             'G': 0,
  53             'PG': 10,
  54             'PG-13': 13,
  55             'R': 16,
  56             'NC': 18,
  57         }
  58         age_limit = RATINGS.get(rating_str)
  59
  60         info_url = 'http://www.viki.com/player5_fragment/%s?action=show&controller=videos' % video_id
  61         info_webpage = self._download_webpage(
  62             info_url, video_id, note='Downloading info page')
  63         if re.match(r'\s*<div\s+class="video-error', info_webpage):
  64             raise ExtractorError(
  65                 'Video %s is blocked from your location.' % video_id,
  66                 expected=True)
  67         video_url = self._html_search_regex(
  68             r'<source[^>]+src="([^"]+)"', info_webpage, 'video URL')
  69
  70         upload_date_str = self._html_search_regex(
  71             r'"created_at":"([^"]+)"', info_webpage, 'upload date')
  72         upload_date = (
  73             unified_strdate(upload_date_str)
  74             if upload_date_str is not None
  75             else None
  76         )
  77
  78         # subtitles
  79         video_subtitles = self.extract_subtitles(video_id, info_webpage)
  80         if self._downloader.params.get('listsubtitles', False):
  81             self._list_available_subtitles(video_id, info_webpage)
  82             return
  83
  84         return {
  85             'id': video_id,
  86             'title': title,
  87             'url': video_url,
  88             'description': description,
  89             'thumbnail': thumbnail,
  90             'age_limit': age_limit,
  91             'uploader': uploader,
  92             'subtitles': video_subtitles,
  93             'upload_date': upload_date,
  94         }
  95
  96     def _get_available_subtitles(self, video_id, info_webpage):
  97         res = {}
  98         for sturl_html in re.findall(r'<track src="([^"]+)"/>', info_webpage):
  99             sturl = unescapeHTML(sturl_html)
 100             m = re.search(r'/(?P<lang>[a-z]+)\.vtt', sturl)
 101             if not m:
 102                 continue
 103             res[m.group('lang')] = sturl
 104         return res