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 (
   8     compat_urllib_request,
   9     compat_urllib_parse,
  10 )
  11
  12
  13 class SinaIE(InfoExtractor):
  14     _VALID_URL = r'''(?x)https?://(.*?\.)?video\.sina\.com\.cn/
  15                         (
  16                             (.+?/(((?P<pseudo_id>\d+).html)|(.*?(\#|(vid=)|b/)(?P<id>\d+?)($|&|\-))))
  17                             |
  18                             # This is used by external sites like Weibo
  19                             (api/sinawebApi/outplay.php/(?P<token>.+?)\.swf)
  20                         )
  21                   '''
  22
  23     _TESTS = [
  24         {
  25             'url': 'http://video.sina.com.cn/news/vlist/zt/chczlj2013/?opsubject_id=top12#110028898',
  26             'md5': 'd65dd22ddcf44e38ce2bf58a10c3e71f',
  27             'info_dict': {
  28                 'id': '110028898',
  29                 'ext': 'flv',
  30                 'title': '《中国新闻》 朝鲜要求巴拿马立即释放被扣船员',
  31             }
  32         },
  33         {
  34             'url': 'http://video.sina.com.cn/v/b/101314253-1290078633.html',
  35             'info_dict': {
  36                 'id': '101314253',
  37                 'ext': 'flv',
  38                 'title': '军方提高对朝情报监视级别',
  39             },
  40         },
  41     ]
  42
  43     def _extract_video(self, video_id):
  44         data = compat_urllib_parse.urlencode({'vid': video_id})
  45         url_doc = self._download_xml('http://v.iask.com/v_play.php?%s' % data,
  46                                      video_id, 'Downloading video url')
  47         image_page = self._download_webpage(
  48             'http://interface.video.sina.com.cn/interface/common/getVideoImage.php?%s' % data,
  49             video_id, 'Downloading thumbnail info')
  50
  51         return {'id': video_id,
  52                 'url': url_doc.find('./durl/url').text,
  53                 'ext': 'flv',
  54                 'title': url_doc.find('./vname').text,
  55                 'thumbnail': image_page.split('=')[1],
  56                 }
  57
  58     def _real_extract(self, url):
  59         mobj = re.match(self._VALID_URL, url)
  60         video_id = mobj.group('id')
  61         if mobj.group('token') is not None:
  62             # The video id is in the redirected url
  63             self.to_screen('Getting video id')
  64             request = compat_urllib_request.Request(url)
  65             request.get_method = lambda: 'HEAD'
  66             (_, urlh) = self._download_webpage_handle(request, 'NA', False)
  67             return self._real_extract(urlh.geturl())
  68         elif video_id is None:
  69             pseudo_id = mobj.group('pseudo_id')
  70             webpage = self._download_webpage(url, pseudo_id)
  71             video_id = self._search_regex(r'vid:\'(\d+?)\'', webpage, 'video id')
  72
  73         return self._extract_video(video_id)