youtube_dl/extractor/vk.py

   1 # encoding: utf-8
   2 import re
   3 import json
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     compat_str,
   8     unescapeHTML,
   9 )
  10
  11
  12 class VKIE(InfoExtractor):
  13     IE_NAME = u'vk.com'
  14     _VALID_URL = r'https?://vk\.com/(?:videos.*?\?.*?z=)?video(?P<id>.*?)(?:\?|%2F|$)'
  15
  16     _TEST = {
  17         u'url': u'http://vk.com/videos-77521?z=video-77521_162222515%2Fclub77521',
  18         u'md5': u'0deae91935c54e00003c2a00646315f0',
  19         u'info_dict': {
  20             u'id': u'162222515',
  21             u'ext': u'flv',
  22             u'title': u'ProtivoGunz - Хуёвая песня',
  23             u'uploader': u'Noize MC',
  24         },
  25     }
  26
  27     def _real_extract(self, url):
  28         mobj = re.match(self._VALID_URL, url)
  29         video_id = mobj.group('id')
  30         info_url = 'http://vk.com/al_video.php?act=show&al=1&video=%s' % video_id
  31         info_page = self._download_webpage(info_url, video_id)
  32         m_yt = re.search(r'src="(http://www.youtube.com/.*?)"', info_page)
  33         if m_yt is not None:
  34             self.to_screen(u'Youtube video detected')
  35             return self.url_result(m_yt.group(1), 'Youtube')
  36         vars_json = self._search_regex(r'var vars = ({.*?});', info_page, u'vars')
  37         vars = json.loads(vars_json)
  38
  39         return {
  40             'id': compat_str(vars['vid']),
  41             'url': vars['url240'],
  42             'title': unescapeHTML(vars['md_title']),
  43             'thumbnail': vars['jpg'],
  44             'uploader': vars['md_author'],
  45         }