youtube_dl/extractor/c56.py

   1 # coding: utf-8
   2
   3 import re
   4 import json
   5
   6 from .common import InfoExtractor
   7 from ..utils import determine_ext
   8
   9 class C56IE(InfoExtractor):
  10     _VALID_URL = r'https?://((www|player)\.)?56\.com/(.+?/)?(v_|(play_album.+-))(?P<textid>.+?)\.(html|swf)'
  11     IE_NAME = u'56.com'
  12
  13     _TEST ={
  14         u'url': u'http://www.56.com/u39/v_OTM0NDA3MTY.html',
  15         u'file': u'93440716.flv',
  16         u'md5': u'e59995ac63d0457783ea05f93f12a866',
  17         u'info_dict': {
  18             u'title': u'网事知多少 第32期：车怒',
  19         },
  20     }
  21
  22     def _real_extract(self, url):
  23         mobj = re.match(self._VALID_URL, url, flags=re.VERBOSE)
  24         text_id = mobj.group('textid')
  25         info_page = self._download_webpage('http://vxml.56.com/json/%s/' % text_id,
  26                                            text_id, u'Downloading video info')
  27         info = json.loads(info_page)['info']
  28         best_format = sorted(info['rfiles'], key=lambda f: int(f['filesize']))[-1]
  29         video_url = best_format['url']
  30
  31         return {'id': info['vid'],
  32                 'title': info['Subject'],
  33                 'url': video_url,
  34                 'ext': determine_ext(video_url),
  35                 'thumbnail': info.get('bimg') or info.get('img'),
  36                 }