class MetacafeIE(InfoExtractor):
- _VALID_URL = r'(?:http://)?(?:www\.)?metacafe\.com/watch/([^/]+)/([^/]+)/.*'
+ _VALID_URL = r'http://(?:www\.)?metacafe\.com/watch/([^/]+)/([^/]+)/.*'
_DISCLAIMER = 'http://www.metacafe.com/family_filter/'
_FILTER_POST = 'http://www.metacafe.com/f/index.php?inputType=filter&controllerGroup=user'
IE_NAME = 'metacafe'
'id': '_aUehQsCQtM',
'ext': 'mp4',
'upload_date': '20090102',
- 'title': 'The Electric Company | \"Short I\" | PBS KIDS GO!',
+ 'title': 'The Electric Company | "Short I" | PBS KIDS GO!',
'description': 'md5:2439a8ef6d5a70e380c22f5ad323e5a8',
'uploader': 'PBS',
'uploader_id': 'PBS'
},
# cbs video
{
- 'url': 'http://www.metacafe.com/watch/cb-0rOxMBabDXN6/samsung_galaxy_note_2_samsungs_next_generation_phablet/',
+ 'url': 'http://www.metacafe.com/watch/cb-8VD4r_Zws8VP/open_this_is_face_the_nation_february_9/',
'info_dict': {
- 'id': '0rOxMBabDXN6',
+ 'id': '8VD4r_Zws8VP',
'ext': 'flv',
- 'title': 'Samsung Galaxy Note 2: Samsung\'s next-generation phablet',
- 'description': 'md5:54d49fac53d26d5a0aaeccd061ada09d',
- 'duration': 129,
+ 'title': 'Open: This is Face the Nation, February 9',
+ 'description': 'md5:8a9ceec26d1f7ed6eab610834cc1a476',
+ 'duration': 96,
},
'params': {
# rtmp download
]
def report_disclaimer(self):
- """Report disclaimer retrieval."""
self.to_screen('Retrieving disclaimer')
def _real_initialize(self):
video_url = mobj.group(1)
video_ext = 'mp4'
else:
- mobj = re.search(r' name="flashvars" value="(.*?)"', webpage)
- if mobj is None:
- raise ExtractorError('Unable to extract media URL')
- vardict = compat_parse_qs(mobj.group(1))
+ flashvars = self._search_regex(
+ r' name="flashvars" value="(.*?)"', webpage, 'flashvars')
+ vardict = compat_parse_qs(flashvars)
if 'mediaData' not in vardict:
raise ExtractorError('Unable to extract media URL')
mobj = re.search(
video_url = '%s?__gda__=%s' % (mediaURL, mobj.group('key'))
video_ext = determine_ext(video_url)
- video_title = self._html_search_regex(r'(?im)<title>(.*) - Video</title>', webpage, 'title')
+ video_title = self._html_search_regex(
+ r'(?im)<title>(.*) - Video</title>', webpage, 'title')
description = self._og_search_description(webpage)
thumbnail = self._og_search_thumbnail(webpage)
video_uploader = self._html_search_regex(
'description': description,
'uploader': video_uploader,
'title': video_title,
- 'thumbnail':thumbnail,
+ 'thumbnail': thumbnail,
'ext': video_ext,
'age_limit': age_limit,
}