youtube_dl/extractor/c56.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5 import json
   6
   7 from .common import InfoExtractor
   8
   9
  10 class C56IE(InfoExtractor):
  11     _VALID_URL = r'https?://((www|player)\.)?56\.com/(.+?/)?(v_|(play_album.+-))(?P<textid>.+?)\.(html|swf)'
  12     IE_NAME = '56.com'
  13     _TEST = {
  14         'url': 'http://www.56.com/u39/v_OTM0NDA3MTY.html',
  15         'file': '93440716.flv',
  16         'md5': 'e59995ac63d0457783ea05f93f12a866',
  17         'info_dict': {
  18             'title': '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, 'Downloading video info')
  27         info = json.loads(info_page)['info']
  28         formats = [{
  29             'format_id': f['type'],
  30             'filesize': int(f['filesize']),
  31             'url': f['url']
  32         } for f in info['rfiles']]
  33         self._sort_formats(formats)
  34
  35         return {
  36             'id': info['vid'],
  37             'title': info['Subject'],
  38             'formats': formats,
  39             'thumbnail': info.get('bimg') or info.get('img'),
  40         }