youtube_dl/extractor/weibo.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7
   8
   9 class WeiboIE(InfoExtractor):
  10     """
  11     The videos in Weibo come from different sites, this IE just finds the link
  12     to the external video and returns it.
  13     """
  14     _VALID_URL = r'https?://video\.weibo\.com/v/weishipin/t_(?P<id>.+?)\.htm'
  15
  16     _TEST = {
  17         'url': 'http://video.weibo.com/v/weishipin/t_zjUw2kZ.htm',
  18         'info_dict': {
  19             'id': '98322879',
  20             'ext': 'flv',
  21             'title': '魔声耳机最新广告“All Eyes On Us”',
  22         },
  23         'params': {
  24             'skip_download': True,
  25         },
  26         'add_ie': ['Sina'],
  27     }
  28
  29     # Additional example videos from different sites
  30     # Youku: http://video.weibo.com/v/weishipin/t_zQGDWQ8.htm
  31     # 56.com: http://video.weibo.com/v/weishipin/t_zQ44HxN.htm
  32
  33     def _real_extract(self, url):
  34         mobj = re.match(self._VALID_URL, url, flags=re.VERBOSE)
  35         video_id = mobj.group('id')
  36         info_url = 'http://video.weibo.com/?s=v&a=play_list&format=json&mix_video_id=t_%s' % video_id
  37         info = self._download_json(info_url, video_id)
  38
  39         videos_urls = map(lambda v: v['play_page_url'], info['result']['data'])
  40         # Prefer sina video since they have thumbnails
  41         videos_urls = sorted(videos_urls, key=lambda u: 'video.sina.com' in u)
  42         player_url = videos_urls[-1]
  43         m_sina = re.match(r'https?://video\.sina\.com\.cn/v/b/(\d+)-\d+\.html',
  44             player_url)
  45         if m_sina is not None:
  46             self.to_screen('Sina video detected')
  47             sina_id = m_sina.group(1)
  48             player_url = 'http://you.video.sina.com.cn/swf/quotePlayer.swf?vid=%s' % sina_id
  49         return self.url_result(player_url)