- mobj = re.match(self._VALID_URL, url)
- if mobj is None:
- raise ExtractorError(u'Invalid URL: %s' % url)
- showName = mobj.group('showname')
- videoId = mobj.group('episode')
-
- self.report_extraction(videoId)
- webpage = self._download_webpage(url, videoId)
-
- videoDesc = self._html_search_regex('<meta name="description" content="([^"]*)"',
- webpage, u'description', fatal=False)
-
- playerUrl = self._og_search_video_url(webpage, name='player url')
-
- title = self._html_search_regex('<meta name="title" content="([^"]*)"',
- webpage, u'player url').split(' : ')[-1]
-
- configUrl = self._search_regex('config=(.*)$', playerUrl, u'config url')
- configUrl = compat_urllib_parse.unquote(configUrl)
-
- configJSON = self._download_webpage(configUrl, videoId,
- u'Downloading configuration',
- u'unable to download configuration')
-
- # Technically, it's JavaScript, not JSON
- configJSON = configJSON.replace("'", '"')
-
+ video_id = self._match_id(url)
+ webpage = self._download_webpage(url, video_id)
+
+ uploader_id = self._html_search_regex(
+ r"<h1 class='headline'><a href='/videos/view/(.*?)'",
+ webpage, 'uploader ID', fatal=False)
+ uploader = self._html_search_regex(
+ r"<h1 class='headline'>(.*?)</a>",
+ webpage, 'uploader', fatal=False)
+ description = self._html_search_meta('description', webpage)
+
+ raw_title = self._html_search_meta('title', webpage, fatal=True)
+ title = raw_title.partition(' : ')[2]
+
+ player_url = self._og_search_video_url(webpage, name='player URL')
+ config_url = compat_urllib_parse.unquote(self._search_regex(
+ r'config=(.*)$', player_url, 'config URL'))
+
+ formats = []
+
+ def _add_format(name, cfgurl, quality):
+ config = self._download_json(
+ cfgurl, video_id,
+ 'Downloading ' + name + ' configuration',
+ 'Unable to download ' + name + ' configuration',
+ transform_source=js_to_json)
+
+ playlist = config['playlist']
+ video_url = next(
+ p['url'] for p in playlist
+ if p.get('eventCategory') == 'Video')
+ formats.append({
+ 'url': video_url,
+ 'format_id': name,
+ 'quality': quality,
+ })
+
+ _add_format('normal', config_url, quality=0)
+ hq_url = (config_url +
+ ('&hq=1' if '?' in config_url else config_url + '?hq=1'))