[vuclip] Fix extraction

This commit is contained in:
Philipp Hagemeister 2014-09-29 13:07:58 +02:00
parent e7db973328
commit 59c03a9bfb

View file

@ -5,6 +5,7 @@
from .common import InfoExtractor
from ..utils import (
compat_urllib_parse_urlparse,
ExtractorError,
parse_duration,
qualities,
)
@ -14,13 +15,12 @@ class VuClipIE(InfoExtractor):
_VALID_URL = r'http://(?:m\.)?vuclip\.com/w\?.*?cid=(?P<id>[0-9]+)'
_TEST = {
'url': 'http://m.vuclip.com/w?cid=843902317&fid=63532&z=1007&nvar&frm=index.html&bu=4757321434',
'md5': '92ac9d1ccefec4f0bb474661ab144fcf',
'url': 'http://m.vuclip.com/w?cid=922692425&fid=70295&z=1010&nvar&frm=index.html',
'info_dict': {
'id': '843902317',
'id': '922692425',
'ext': '3gp',
'title': 'Movie Trailer: Noah',
'duration': 139,
'title': 'The Toy Soldiers - Hollywood Movie Trailer',
'duration': 180,
}
}
@ -28,7 +28,9 @@ def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
video_id = mobj.group('id')
webpage = self._download_webpage(url, video_id)
#webpage = self._download_webpage(url, video_id)
import io
webpage = io.open('922692425_http_-_m.vuclip.com_wcid=922692425_fid=70295_z=1010_nvar_frm=index.html.dump', encoding='utf-8').read()
ad_m = re.search(
r'''value="No.*?" onClick="location.href='([^"']+)'"''', webpage)
if ad_m:
@ -37,16 +39,32 @@ def _real_extract(self, url):
webpage = self._download_webpage(
adfree_url, video_id, note='Download post-ad page')
error_msg = self._html_search_regex(
r'<p class="message">(.*?)</p>', webpage, 'error message',
default=None)
if error_msg:
raise ExtractorError(
'%s said: %s' % (self.IE_NAME, error_msg), expected=True)
# These clowns alternate between two page types
links_code = self._search_regex(
r'(?s)<div class="social align_c".*?>(.*?)<hr\s*/?>', webpage,
'links')
r'''(?xs)
(?:
<img\s+src="/im/play.gif".*?>|
<!--\ player\ end\ -->\s*</div><!--\ thumb\ end-->
)
(.*?)
(?:
<a\s+href="fblike'|<div\s+class="social">
)
''', webpage, 'links')
title = self._html_search_regex(
r'<title>(.*?)-\s*Vuclip</title>', webpage, 'title').strip()
quality_order = qualities(['Reg', 'Hi'])
formats = []
for url, q in re.findall(
r'<a href="(?P<url>[^"]+)".*?>(?P<q>[^<]+)</a>', links_code):
r'<a\s+href="(?P<url>[^"]+)".*?>(?:<button[^>]*>)?(?P<q>[^<]+)(?:</button>)?</a>', links_code):
format_id = compat_urllib_parse_urlparse(url).scheme + '-' + q
formats.append({
'format_id': format_id,
@ -56,7 +74,7 @@ def _real_extract(self, url):
self._sort_formats(formats)
duration = parse_duration(self._search_regex(
r'\(([0-9:]+)\)</span></h1>', webpage, 'duration', fatal=False))
r'\(([0-9:]+)\)</span>', webpage, 'duration', fatal=False))
return {
'id': video_id,