[worldstarhiphop] Correct title extraction

This commit is contained in:
Philipp Hagemeister 2014-09-29 05:02:58 +02:00
parent 2593039522
commit 12548cd933
1 changed files with 10 additions and 12 deletions

View File

@ -13,37 +13,35 @@ class WorldStarHipHopIE(InfoExtractor):
"info_dict": { "info_dict": {
"id": "wshh6a7q1ny0G34ZwuIO", "id": "wshh6a7q1ny0G34ZwuIO",
"ext": "mp4", "ext": "mp4",
"title": "Video: KO Of The Week: MMA Fighter Gets Knocked Out By Swift Head Kick!" "title": "KO Of The Week: MMA Fighter Gets Knocked Out By Swift Head Kick!"
} }
} }
def _real_extract(self, url): def _real_extract(self, url):
m = re.match(self._VALID_URL, url) video_id = self._match_id(url)
video_id = m.group('id') webpage = self._download_webpage(url, video_id)
webpage_src = self._download_webpage(url, video_id) m_vevo_id = re.search(r'videoId=(.*?)&amp?', webpage)
m_vevo_id = re.search(r'videoId=(.*?)&amp?',
webpage_src)
if m_vevo_id is not None: if m_vevo_id is not None:
return self.url_result('vevo:%s' % m_vevo_id.group(1), ie='Vevo') return self.url_result('vevo:%s' % m_vevo_id.group(1), ie='Vevo')
video_url = self._search_regex( video_url = self._search_regex(
r'so\.addVariable\("file","(.*?)"\)', webpage_src, 'video URL') r'so\.addVariable\("file","(.*?)"\)', webpage, 'video URL')
if 'youtube' in video_url: if 'youtube' in video_url:
return self.url_result(video_url, ie='Youtube') return self.url_result(video_url, ie='Youtube')
video_title = self._html_search_regex( video_title = self._html_search_regex(
r"<title>(.*)</title>", webpage_src, 'title') r'(?s)<div class="content-heading">\s*<h1>(.*?)</h1>',
webpage, 'title')
# Getting thumbnail and if not thumbnail sets correct title for WSHH candy video. # Getting thumbnail and if not thumbnail sets correct title for WSHH candy video.
thumbnail = self._html_search_regex( thumbnail = self._html_search_regex(
r'rel="image_src" href="(.*)" />', webpage_src, 'thumbnail', r'rel="image_src" href="(.*)" />', webpage, 'thumbnail',
fatal=False) fatal=False)
if not thumbnail: if not thumbnail:
_title = r"""candytitles.*>(.*)</span>""" _title = r'candytitles.*>(.*)</span>'
mobj = re.search(_title, webpage_src) mobj = re.search(_title, webpage)
if mobj is not None: if mobj is not None:
video_title = mobj.group(1) video_title = mobj.group(1)