[extractor/common] Require closing quote in _og_regexes (Closes #7174)

E.g. do not match `property='og:video:type'` when `og:video` is requested.
This commit is contained in:
Sergey M․ 2015-10-14 20:49:39 +06:00
parent 4aa353673b
commit 7a6d76a64d

View file

@ -646,7 +646,8 @@ def _get_tfa_info(self, note='two-factor verification code'):
@staticmethod
def _og_regexes(prop):
content_re = r'content=(?:"([^>]+?)"|\'([^>]+?)\'|\s*([^\s"\'=<>`]+?))'
property_re = r'(?:name|property)=[\'"]?og:%s[\'"]?' % re.escape(prop)
property_re = (r'(?:name|property)=(?:\'og:%(prop)s\'|"og:%(prop)s"|\s*og:%(prop)s\b)'
% {'prop': re.escape(prop)})
template = r'<meta[^>]+?%s[^>]+?%s'
return [
template % (property_re, content_re),