youtube_dl/extractor/kankan.py

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