yt_dlp/extractor/c56.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4
   5 from .common import InfoExtractor
   6 from ..utils import js_to_json
   7
   8
   9 class C56IE(InfoExtractor):
  10     _VALID_URL = r'https?://(?:(?:www|player)\.)?56\.com/(?:.+?/)?(?:v_|(?:play_album.+-))(?P<textid>.+?)\.(?:html|swf)'
  11     IE_NAME = '56.com'
  12     _TESTS = [{
  13         'url': 'http://www.56.com/u39/v_OTM0NDA3MTY.html',
  14         'md5': 'e59995ac63d0457783ea05f93f12a866',
  15         'info_dict': {
  16             'id': '93440716',
  17             'ext': 'flv',
  18             'title': '网事知多少 第32期：车怒',
  19             'duration': 283.813,
  20         },
  21     }, {
  22         'url': 'http://www.56.com/u47/v_MTM5NjQ5ODc2.html',
  23         'md5': '',
  24         'info_dict': {
  25             'id': '82247482',
  26             'title': '爱的诅咒之杜鹃花开',
  27         },
  28         'playlist_count': 7,
  29         'add_ie': ['Sohu'],
  30     }]
  31
  32     def _real_extract(self, url):
  33         mobj = self._match_valid_url(url)
  34         text_id = mobj.group('textid')
  35
  36         webpage = self._download_webpage(url, text_id)
  37         sohu_video_info_str = self._search_regex(
  38             r'var\s+sohuVideoInfo\s*=\s*({[^}]+});', webpage, 'Sohu video info', default=None)
  39         if sohu_video_info_str:
  40             sohu_video_info = self._parse_json(
  41                 sohu_video_info_str, text_id, transform_source=js_to_json)
  42             return self.url_result(sohu_video_info['url'], 'Sohu')
  43
  44         page = self._download_json(
  45             'http://vxml.56.com/json/%s/' % text_id, text_id, 'Downloading video info')
  46
  47         info = page['info']
  48
  49         formats = [
  50             {
  51                 'format_id': f['type'],
  52                 'filesize': int(f['filesize']),
  53                 'url': f['url']
  54             } for f in info['rfiles']
  55         ]
  56         self._sort_formats(formats)
  57
  58         return {
  59             'id': info['vid'],
  60             'title': info['Subject'],
  61             'duration': int(info['duration']) / 1000.0,
  62             'formats': formats,
  63             'thumbnail': info.get('bimg') or info.get('img'),
  64         }