youtube_dl/extractor/weibo.py

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