youtube_dl/extractor/criterion.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5
   6
   7 class CriterionIE(InfoExtractor):
   8     _VALID_URL = r'https?://(?:www\.)?criterion\.com/films/(?P<id>[0-9]+)-.+'
   9     _TEST = {
  10         'url': 'http://www.criterion.com/films/184-le-samourai',
  11         'md5': 'bc51beba55685509883a9a7830919ec3',
  12         'info_dict': {
  13             'id': '184',
  14             'ext': 'mp4',
  15             'title': 'Le Samouraï',
  16             'description': 'md5:a2b4b116326558149bef81f76dcbb93f',
  17             'thumbnail': r're:^https?://.*\.jpg$',
  18         }
  19     }
  20
  21     def _real_extract(self, url):
  22         video_id = self._match_id(url)
  23         webpage = self._download_webpage(url, video_id)
  24
  25         final_url = self._search_regex(
  26             r'so\.addVariable\("videoURL", "(.+?)"\)\;', webpage, 'video url')
  27         title = self._og_search_title(webpage)
  28         description = self._html_search_meta('description', webpage)
  29         thumbnail = self._search_regex(
  30             r'so\.addVariable\("thumbnailURL", "(.+?)"\)\;',
  31             webpage, 'thumbnail url')
  32
  33         return {
  34             'id': video_id,
  35             'url': final_url,
  36             'title': title,
  37             'description': description,
  38             'thumbnail': thumbnail,
  39         }