youtube_dl/extractor/breakcom.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4 import json
   5
   6 from .common import InfoExtractor
   7
   8
   9 class BreakIE(InfoExtractor):
  10     _VALID_URL = r'http://(?:www\.)?break\.com/video/([^/]+)'
  11     _TEST = {
  12         'url': 'http://www.break.com/video/when-girls-act-like-guys-2468056',
  13         'md5': 'a3513fb1547fba4fb6cfac1bffc6c46b',
  14         'info_dict': {
  15             'id': '2468056',
  16             'ext': 'mp4',
  17             'title': 'When Girls Act Like D-Bags',
  18         }
  19     }
  20
  21     def _real_extract(self, url):
  22         mobj = re.match(self._VALID_URL, url)
  23         video_id = mobj.group(1).split("-")[-1]
  24         embed_url = 'http://www.break.com/embed/%s' % video_id
  25         webpage = self._download_webpage(embed_url, video_id)
  26         info_json = self._search_regex(r'var embedVars = ({.*})\s*?</script>',
  27             webpage, 'info json', flags=re.DOTALL)
  28         info = json.loads(info_json)
  29         video_url = info['videoUri']
  30         youtube_id = info.get('youtubeId')
  31         if youtube_id:
  32             return self.url_result(youtube_id, 'Youtube')
  33
  34         final_url = video_url + '?' + info['AuthToken']
  35         return {
  36             'id': video_id,
  37             'url': final_url,
  38             'title': info['contentName'],
  39             'thumbnail': info['thumbUri'],
  40         }