youtube_dl/extractor/kankan.py

   1 import re
   2 import hashlib
   3
   4 from .common import InfoExtractor
   5 from ..utils import determine_ext
   6
   7 _md5 = lambda s: hashlib.md5(s.encode('utf-8')).hexdigest()
   8
   9 class KankanIE(InfoExtractor):
  10     _VALID_URL = r'https?://(?:.*?\.)?kankan\.com/.+?/(?P<id>\d+)\.shtml'
  11
  12     _TEST = {
  13         u'url': u'http://yinyue.kankan.com/vod/48/48863.shtml',
  14         u'file': u'48863.flv',
  15         u'md5': u'29aca1e47ae68fc28804aca89f29507e',
  16         u'info_dict': {
  17             u'title': u'Ready To Go',
  18         },
  19     }
  20
  21     def _real_extract(self, url):
  22         mobj = re.match(self._VALID_URL, url)
  23         video_id = mobj.group('id')
  24         webpage = self._download_webpage(url, video_id)
  25
  26         title = self._search_regex(r'(?:G_TITLE=|G_MOVIE_TITLE = )[\'"](.+?)[\'"]', webpage, u'video title')
  27         surls = re.search(r'surls:\[\'.+?\'\]|lurl:\'.+?\.flv\'', webpage).group(0)
  28         gcids = re.findall(r"http://.+?/.+?/(.+?)/", surls)
  29         gcid = gcids[-1]
  30
  31         video_info_page = self._download_webpage('http://p2s.cl.kankan.com/getCdnresource_flv?gcid=%s' % gcid,
  32                                                  video_id, u'Downloading video url info')
  33         ip = self._search_regex(r'ip:"(.+?)"', video_info_page, u'video url ip')
  34         path = self._search_regex(r'path:"(.+?)"', video_info_page, u'video url path')
  35         param1 = self._search_regex(r'param1:(\d+)', video_info_page, u'param1')
  36         param2 = self._search_regex(r'param2:(\d+)', video_info_page, u'param2')
  37         key = _md5('xl_mp43651' + param1 + param2)
  38         video_url = 'http://%s%s?key=%s&key1=%s' % (ip, path, key, param2)
  39
  40         return {'id': video_id,
  41                 'title': title,
  42                 'url': video_url,
  43                 'ext': determine_ext(video_url),
  44                 }