[goshgay] Fix extraction

This commit is contained in:
Philipp Hagemeister 2014-12-12 12:55:13 +01:00
parent 73aeb2dc56
commit 82b34105d3

View file

@ -2,22 +2,25 @@
from __future__ import unicode_literals
from .common import InfoExtractor
from ..compat import compat_urlparse
from ..compat import (
compat_parse_qs,
)
from ..utils import (
ExtractorError,
parse_duration,
)
class GoshgayIE(InfoExtractor):
_VALID_URL = r'https?://www\.goshgay\.com/video(?P<id>\d+?)($|/)'
_TEST = {
'url': 'http://www.goshgay.com/video4116282',
'md5': '268b9f3c3229105c57859e166dd72b03',
'url': 'http://www.goshgay.com/video299069/diesel_sfw_xxx_video',
'md5': '027fcc54459dff0feb0bc06a7aeda680',
'info_dict': {
'id': '4116282',
'id': '299069',
'ext': 'flv',
'title': 'md5:089833a4790b5e103285a07337f245bf',
'thumbnail': 're:http://.*\.jpg',
'title': 'DIESEL SFW XXX Video',
'thumbnail': 're:^http://.*\.jpg$',
'duration': 79,
'age_limit': 18,
}
}
@ -26,33 +29,25 @@ def _real_extract(self, url):
video_id = self._match_id(url)
webpage = self._download_webpage(url, video_id)
title = self._og_search_title(webpage)
thumbnail = self._og_search_thumbnail(webpage)
title = self._html_search_regex(
r'<h2>(.*?)<', webpage, 'title')
duration = parse_duration(self._html_search_regex(
r'<span class="duration">\s*-?\s*(.*?)</span>',
webpage, 'duration', fatal=False))
family_friendly = self._html_search_meta(
'isFamilyFriendly', webpage, default='false')
config_url = self._search_regex(
r"'config'\s*:\s*'([^']+)'", webpage, 'config URL')
config = self._download_xml(
config_url, video_id, 'Downloading player config XML')
if config is None:
raise ExtractorError('Missing config XML')
if config.tag != 'config':
raise ExtractorError('Missing config attribute')
fns = config.findall('file')
if len(fns) < 1:
raise ExtractorError('Missing media URI')
video_url = fns[0].text
url_comp = compat_urlparse.urlparse(url)
ref = "%s://%s%s" % (url_comp[0], url_comp[1], url_comp[2])
flashvars = compat_parse_qs(self._html_search_regex(
r'<embed.+?id="flash-player-embed".+?flashvars="([^"]+)"',
webpage, 'flashvars'))
thumbnail = flashvars.get('url_bigthumb', [None])[0]
video_url = flashvars['flv_url'][0]
return {
'id': video_id,
'url': video_url,
'title': title,
'thumbnail': thumbnail,
'http_referer': ref,
'duration': duration,
'age_limit': 0 if family_friendly == 'true' else 18,
}