_ Git - youtube-dl/blob - youtube_dl/extractor/vidzi.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .jwplatform import JWPlatformBaseIE
   7 from ..utils import (
   8     base36,
   9     js_to_json,
  10 )
  11
  12
  13 class VidziIE(JWPlatformBaseIE):
  14     _VALID_URL = r'https?://(?:www\.)?vidzi\.tv/(?P<id>\w+)'
  15     _TEST = {
  16         'url': 'http://vidzi.tv/cghql9yq6emu.html',
  17         'md5': '4f16c71ca0c8c8635ab6932b5f3f1660',
  18         'info_dict': {
  19             'id': 'cghql9yq6emu',
  20             'ext': 'mp4',
  21             'title': 'youtube-dl test video  1\\\\2\'3/4<5\\\\6ä7↭',
  22         },
  23         'params': {
  24             # m3u8 download
  25             'skip_download': True,
  26         },
  27     }
  28
  29     def _real_extract(self, url):
  30         video_id = self._match_id(url)
  31
  32         webpage = self._download_webpage(url, video_id)
  33         title = self._html_search_regex(
  34             r'(?s)<h2 class="video-title">(.*?)</h2>', webpage, 'title')
  35
  36         mobj = re.search(r"}\('(.+)',36,(\d+),'([^']+)'\.split\('\|'\)", webpage)
  37         code, count, symbols = mobj.groups()
  38
  39         count = int(count)
  40         symbols = symbols.split('|')
  41
  42         while count:
  43             count -= 1
  44             if symbols[count]:
  45                 code = re.sub(r'\b%s\b' % base36(count), symbols[count], code)
  46
  47         code = code.replace('\\\'', '\'')
  48         jwplayer_data = self._parse_json(
  49             self._search_regex(r'setup\(([^)]+)\)', code, 'jwplayer data'),
  50             video_id, transform_source=js_to_json)
  51
  52         info_dict = self._parse_jwplayer_data(jwplayer_data, video_id, require_title=False)
  53         info_dict['title'] = title
  54
  55         return info_dict