2 from __future__ import unicode_literals
6 from .common import InfoExtractor
13 class SinaIE(InfoExtractor):
14 _VALID_URL = r'''https?://(.*?\.)?video\.sina\.com\.cn/
16 (.+?/(((?P<pseudo_id>\d+).html)|(.*?(\#|(vid=))(?P<id>\d+?)($|&))))
18 # This is used by external sites like Weibo
19 (api/sinawebApi/outplay.php/(?P<token>.+?)\.swf)
24 'url': 'http://video.sina.com.cn/news/vlist/zt/chczlj2013/?opsubject_id=top12#110028898',
25 'file': '110028898.flv',
26 'md5': 'd65dd22ddcf44e38ce2bf58a10c3e71f',
28 'title': '《中国新闻》 朝鲜要求巴拿马立即释放被扣船员',
33 def suitable(cls, url):
34 return re.match(cls._VALID_URL, url, flags=re.VERBOSE) is not None
36 def _extract_video(self, video_id):
37 data = compat_urllib_parse.urlencode({'vid': video_id})
38 url_doc = self._download_xml('http://v.iask.com/v_play.php?%s' % data,
39 video_id, 'Downloading video url')
40 image_page = self._download_webpage(
41 'http://interface.video.sina.com.cn/interface/common/getVideoImage.php?%s' % data,
42 video_id, 'Downloading thumbnail info')
44 return {'id': video_id,
45 'url': url_doc.find('./durl/url').text,
47 'title': url_doc.find('./vname').text,
48 'thumbnail': image_page.split('=')[1],
51 def _real_extract(self, url):
52 mobj = re.match(self._VALID_URL, url, flags=re.VERBOSE)
53 video_id = mobj.group('id')
54 if mobj.group('token') is not None:
55 # The video id is in the redirected url
56 self.to_screen('Getting video id')
57 request = compat_urllib_request.Request(url)
58 request.get_method = lambda: 'HEAD'
59 (_, urlh) = self._download_webpage_handle(request, 'NA', False)
60 return self._real_extract(urlh.geturl())
61 elif video_id is None:
62 pseudo_id = mobj.group('pseudo_id')
63 webpage = self._download_webpage(url, pseudo_id)
64 video_id = self._search_regex(r'vid:\'(\d+?)\'', webpage, 'video id')
66 return self._extract_video(video_id)