[pornhd] Fix extraction (Closes #3739)

This commit is contained in:
Sergey M․ 2014-09-13 21:45:53 +07:00
parent 7e6a715380
commit ceff3fd8ef
1 changed files with 22 additions and 29 deletions

View File

@ -27,47 +27,40 @@ class PornHdIE(InfoExtractor):
webpage = self._download_webpage(url, video_id) webpage = self._download_webpage(url, video_id)
title = self._og_search_title(webpage) title = self._html_search_regex(
TITLE_SUFFIX = ' porn HD Video | PornHD.com ' r'<title>(.+) porn HD.+?</title>', webpage, 'title')
if title.endswith(TITLE_SUFFIX):
title = title[:-len(TITLE_SUFFIX)]
description = self._html_search_regex( description = self._html_search_regex(
r'<div class="description">([^<]+)</div>', webpage, 'description', fatal=False) r'<div class="description">([^<]+)</div>', webpage, 'description', fatal=False)
view_count = int_or_none(self._html_search_regex( view_count = int_or_none(self._html_search_regex(
r'(\d+) views </span>', webpage, 'view count', fatal=False)) r'(\d+) views\s*</span>', webpage, 'view count', fatal=False))
formats = [ videos = re.findall(
{
'url': format_url,
'ext': format.lower(),
'format_id': '%s-%s' % (format.lower(), quality.lower()),
'quality': 1 if quality.lower() == 'high' else 0,
} for format, quality, format_url in re.findall(
r'var __video([\da-zA-Z]+?)(Low|High)StreamUrl = \'(http://.+?)\?noProxy=1\'', webpage) r'var __video([\da-zA-Z]+?)(Low|High)StreamUrl = \'(http://.+?)\?noProxy=1\'', webpage)
]
mobj = re.search(r'flashVars = (?P<flashvars>{.+?});', webpage) mobj = re.search(r'flashVars = (?P<flashvars>{.+?});', webpage)
if mobj: if mobj:
flashvars = json.loads(mobj.group('flashvars')) flashvars = json.loads(mobj.group('flashvars'))
formats.extend([ for key, quality in [('hashlink', 'low'), ('hd', 'high')]:
{ redirect_url = flashvars.get(key)
'url': flashvars['hashlink'].replace('?noProxy=1', ''), if redirect_url:
'ext': 'flv', videos.append(('flv', quality, redirect_url))
'format_id': 'flv-low',
'quality': 0,
},
{
'url': flashvars['hd'].replace('?noProxy=1', ''),
'ext': 'flv',
'format_id': 'flv-high',
'quality': 1,
}
])
thumbnail = flashvars['urlWallpaper'] thumbnail = flashvars['urlWallpaper']
else: else:
thumbnail = self._og_search_thumbnail(webpage) thumbnail = self._og_search_thumbnail(webpage)
formats = []
for format_, quality, redirect_url in videos:
format_id = '%s-%s' % (format_.lower(), quality.lower())
video_url = self._download_webpage(
redirect_url, video_id, 'Downloading %s video link' % format_id, fatal=False)
if not video_url:
continue
formats.append({
'url': video_url,
'ext': format_.lower(),
'format_id': format_id,
'quality': 1 if quality.lower() == 'high' else 0,
})
self._sort_formats(formats) self._sort_formats(formats)
return { return {