[wshh] Extract html5 entries and delegate to generic extractor (closes #12676)

This commit is contained in:
Sergey M․ 2017-04-08 16:01:56 +07:00
parent 4c03973296
commit f8f2da25ab
No known key found for this signature in database
GPG Key ID: 2C393E0F18A9236D

View File

@ -1,12 +1,10 @@
from __future__ import unicode_literals
import re
from .common import InfoExtractor
class WorldStarHipHopIE(InfoExtractor):
_VALID_URL = r'https?://(?:www|m)\.worldstar(?:candy|hiphop)\.com/(?:videos|android)/video\.php\?v=(?P<id>.*)'
_VALID_URL = r'https?://(?:www|m)\.worldstar(?:candy|hiphop)\.com/(?:videos|android)/video\.php\?.*?\bv=(?P<id>[^&]+)'
_TESTS = [{
'url': 'http://www.worldstarhiphop.com/videos/video.php?v=wshh6a7q1ny0G34ZwuIO',
'md5': '9d04de741161603bf7071bbf4e883186',
@ -17,48 +15,26 @@ class WorldStarHipHopIE(InfoExtractor):
}
}, {
'url': 'http://m.worldstarhiphop.com/android/video.php?v=wshh6a7q1ny0G34ZwuIO',
'md5': 'dc1c76c83ecc4190bb1eb143899b87d3',
'info_dict': {
'id': 'wshh6a7q1ny0G34ZwuIO',
'ext': 'mp4',
'title': 'KO Of The Week: MMA Fighter Gets Knocked Out By Swift Head Kick!'
}
'only_matching': True,
}]
def _real_extract(self, url):
video_id = self._match_id(url)
webpage = self._download_webpage(url, video_id)
m_vevo_id = re.search(r'videoId=(.*?)&amp?', webpage)
if m_vevo_id is not None:
return self.url_result('vevo:%s' % m_vevo_id.group(1), ie='Vevo')
entries = self._parse_html5_media_entries(url, webpage, video_id)
video_url = self._search_regex(
[r'so\.addVariable\("file","(.*?)"\)',
r'<div class="artlist">\s*<a[^>]+href="([^"]+)">'],
webpage, 'video URL')
if not entries:
return self.url_result(url, 'Generic')
if 'youtube' in video_url:
return self.url_result(video_url, ie='Youtube')
video_title = self._html_search_regex(
title = self._html_search_regex(
[r'(?s)<div class="content-heading">\s*<h1>(.*?)</h1>',
r'<span[^>]+class="tc-sp-pinned-title">(.*)</span>'],
webpage, 'title')
# Getting thumbnail and if not thumbnail sets correct title for WSHH candy video.
thumbnail = self._html_search_regex(
r'rel="image_src" href="(.*)" />', webpage, 'thumbnail',
default=None)
if not thumbnail:
_title = r'candytitles.*>(.*)</span>'
mobj = re.search(_title, webpage)
if mobj is not None:
video_title = mobj.group(1)
return {
info = entries[0]
info.update({
'id': video_id,
'url': video_url,
'title': video_title,
'thumbnail': thumbnail,
}
'title': title,
})
return info