Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/sina.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..compat import compat_urllib_parse
   8 from ..utils import sanitized_Request
   9
  10
  11 class SinaIE(InfoExtractor):
  12     _VALID_URL = r'''(?x)https?://(.*?\.)?video\.sina\.com\.cn/
  13                         (
  14                             (.+?/(((?P<pseudo_id>\d+).html)|(.*?(\#|(vid=)|b/)(?P<id>\d+?)($|&|\-))))
  15                             |
  16                             # This is used by external sites like Weibo
  17                             (api/sinawebApi/outplay.php/(?P<token>.+?)\.swf)
  18                         )
  19                   '''
  20
  21     _TESTS = [
  22         {
  23             'url': 'http://video.sina.com.cn/news/vlist/zt/chczlj2013/?opsubject_id=top12#110028898',
  24             'md5': 'd65dd22ddcf44e38ce2bf58a10c3e71f',
  25             'info_dict': {
  26                 'id': '110028898',
  27                 'ext': 'flv',
  28                 'title': '《中国新闻》 朝鲜要求巴拿马立即释放被扣船员',
  29             }
  30         },
  31         {
  32             'url': 'http://video.sina.com.cn/v/b/101314253-1290078633.html',
  33             'info_dict': {
  34                 'id': '101314253',
  35                 'ext': 'flv',
  36                 'title': '军方提高对朝情报监视级别',
  37             },
  38         },
  39     ]
  40
  41     def _extract_video(self, video_id):
  42         data = compat_urllib_parse.urlencode({'vid': video_id})
  43         url_doc = self._download_xml('http://v.iask.com/v_play.php?%s' % data,
  44                                      video_id, 'Downloading video url')
  45         image_page = self._download_webpage(
  46             'http://interface.video.sina.com.cn/interface/common/getVideoImage.php?%s' % data,
  47             video_id, 'Downloading thumbnail info')
  48
  49         return {'id': video_id,
  50                 'url': url_doc.find('./durl/url').text,
  51                 'ext': 'flv',
  52                 'title': url_doc.find('./vname').text,
  53                 'thumbnail': image_page.split('=')[1],
  54                 }
  55
  56     def _real_extract(self, url):
  57         mobj = re.match(self._VALID_URL, url)
  58         video_id = mobj.group('id')
  59         if mobj.group('token') is not None:
  60             # The video id is in the redirected url
  61             self.to_screen('Getting video id')
  62             request = sanitized_Request(url)
  63             request.get_method = lambda: 'HEAD'
  64             (_, urlh) = self._download_webpage_handle(request, 'NA', False)
  65             return self._real_extract(urlh.geturl())
  66         elif video_id is None:
  67             pseudo_id = mobj.group('pseudo_id')
  68             webpage = self._download_webpage(url, pseudo_id)
  69             video_id = self._search_regex(r'vid:\'(\d+?)\'', webpage, 'video id')
  70
  71         return self._extract_video(video_id)