_ Git - youtube-dl/blob - 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         'file': '48863.flv',
  17         'md5': '29aca1e47ae68fc28804aca89f29507e',
  18         'info_dict': {
  19             'title': 'Ready To Go',
  20         },
  21     }
  22
  23     def _real_extract(self, url):
  24         mobj = re.match(self._VALID_URL, url)
  25         video_id = mobj.group('id')
  26         webpage = self._download_webpage(url, video_id)
  27
  28         title = self._search_regex(r'(?:G_TITLE=|G_MOVIE_TITLE = )[\'"](.+?)[\'"]', webpage, 'video title')
  29         surls = re.search(r'surls:\[\'.+?\'\]|lurl:\'.+?\.flv\'', webpage).group(0)
  30         gcids = re.findall(r"http://.+?/.+?/(.+?)/", surls)
  31         gcid = gcids[-1]
  32
  33         info_url = 'http://p2s.cl.kankan.com/getCdnresource_flv?gcid=%s' % gcid
  34         video_info_page = self._download_webpage(
  35             info_url, video_id, 'Downloading video url info')
  36         ip = self._search_regex(r'ip:"(.+?)"', video_info_page, 'video url ip')
  37         path = self._search_regex(r'path:"(.+?)"', video_info_page, 'video url path')
  38         param1 = self._search_regex(r'param1:(\d+)', video_info_page, 'param1')
  39         param2 = self._search_regex(r'param2:(\d+)', video_info_page, 'param2')
  40         key = _md5('xl_mp43651' + param1 + param2)
  41         video_url = 'http://%s%s?key=%s&key1=%s' % (ip, path, key, param2)
  42
  43         return {
  44             'id': video_id,
  45             'title': title,
  46             'url': video_url,
  47         }