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 ..utils import (
   8     compat_urllib_request,
   9     compat_urllib_parse,
  10 )
  11
  12
  13 class SinaIE(InfoExtractor):
  14     _VALID_URL = r'''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             'file': '110028898.flv',
  27             'md5': 'd65dd22ddcf44e38ce2bf58a10c3e71f',
  28             'info_dict': {
  29                 'title': '《中国新闻》 朝鲜要求巴拿马立即释放被扣船员',
  30             }
  31         },
  32         {
  33             'url': 'http://video.sina.com.cn/v/b/101314253-1290078633.html',
  34             'info_dict': {
  35                 'id': '101314253',
  36                 'ext': 'flv',
  37                 'title': '军方提高对朝情报监视级别',
  38             },
  39         },
  40     ]
  41
  42     @classmethod
  43     def suitable(cls, url):
  44         return re.match(cls._VALID_URL, url, flags=re.VERBOSE) is not None
  45
  46     def _extract_video(self, video_id):
  47         data = compat_urllib_parse.urlencode({'vid': video_id})
  48         url_doc = self._download_xml('http://v.iask.com/v_play.php?%s' % data,
  49                                      video_id, 'Downloading video url')
  50         image_page = self._download_webpage(
  51             'http://interface.video.sina.com.cn/interface/common/getVideoImage.php?%s' % data,
  52             video_id, 'Downloading thumbnail info')
  53
  54         return {'id': video_id,
  55                 'url': url_doc.find('./durl/url').text,
  56                 'ext': 'flv',
  57                 'title': url_doc.find('./vname').text,
  58                 'thumbnail': image_page.split('=')[1],
  59                 }
  60
  61     def _real_extract(self, url):
  62         mobj = re.match(self._VALID_URL, url, flags=re.VERBOSE)
  63         video_id = mobj.group('id')
  64         if mobj.group('token') is not None:
  65             # The video id is in the redirected url
  66             self.to_screen('Getting video id')
  67             request = compat_urllib_request.Request(url)
  68             request.get_method = lambda: 'HEAD'
  69             (_, urlh) = self._download_webpage_handle(request, 'NA', False)
  70             return self._real_extract(urlh.geturl())
  71         elif video_id is None:
  72             pseudo_id = mobj.group('pseudo_id')
  73             webpage = self._download_webpage(url, pseudo_id)
  74             video_id = self._search_regex(r'vid:\'(\d+?)\'', webpage, 'video id')
  75
  76         return self._extract_video(video_id)