youtube_dl/extractor/kankan.py

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..utils import determine_ext
   5
   6
   7 class KankanIE(InfoExtractor):
   8     _VALID_URL = r'https?://(?:.*?\.)?kankan\.com/.+?/(?P<id>\d+)\.shtml'
   9
  10     _TEST = {
  11         u'url': u'http://yinyue.kankan.com/vod/48/48863.shtml',
  12         u'file': u'48863.flv',
  13         u'md5': u'29aca1e47ae68fc28804aca89f29507e',
  14         u'info_dict': {
  15             u'title': u'Ready To Go',
  16         },
  17     }
  18
  19     def _real_extract(self, url):
  20         mobj = re.match(self._VALID_URL, url)
  21         video_id = mobj.group('id')
  22         webpage = self._download_webpage(url, video_id)
  23
  24         title = self._search_regex(r'(?:G_TITLE=|G_MOVIE_TITLE = )[\'"](.+?)[\'"]', webpage, u'video title')
  25         surls = re.search(r'surls:\[\'.+?\'\]|lurl:\'.+?\.flv\'', webpage).group(0)
  26         gcids = re.findall(r"http://.+?/.+?/(.+?)/", surls)
  27         gcid = gcids[-1]
  28
  29         video_info_page = self._download_webpage('http://p2s.cl.kankan.com/getCdnresource_flv?gcid=%s' % gcid,
  30                                                  video_id, u'Downloading video url info')
  31         ip = self._search_regex(r'ip:"(.+?)"', video_info_page, u'video url ip')
  32         path = self._search_regex(r'path:"(.+?)"', video_info_page, u'video url path')
  33         video_url = 'http://%s%s' % (ip, path)
  34
  35         return {'id': video_id,
  36                 'title': title,
  37                 'url': video_url,
  38                 'ext': determine_ext(video_url),
  39                 }