Browse Source

[extractor/common] Allow HTML5 unquoted attribute values

Fixes #7108

HTML5 allows unquoted attribute values. See the "Unquoted attribute value
syntax" section [1] for more information

[1] http://www.w3.org/TR/html5/syntax.html
Yen Chi Hsuan 10 years ago
parent
commit
57935b2564
1 changed files with 1 additions and 1 deletions
  1. 1 1
      youtube_dl/extractor/common.py

+ 1 - 1
youtube_dl/extractor/common.py

@@ -646,7 +646,7 @@ class InfoExtractor(object):
     @staticmethod
     @staticmethod
     def _og_regexes(prop):
     def _og_regexes(prop):
         content_re = r'content=(?:"([^>]+?)"|\'([^>]+?)\')'
         content_re = r'content=(?:"([^>]+?)"|\'([^>]+?)\')'
-        property_re = r'(?:name|property)=[\'"]og:%s[\'"]' % re.escape(prop)
+        property_re = r'(?:name|property)=[\'"]?og:%s[\'"]?' % re.escape(prop)
         template = r'<meta[^>]+?%s[^>]+?%s'
         template = r'<meta[^>]+?%s[^>]+?%s'
         return [
         return [
             template % (property_re, content_re),
             template % (property_re, content_re),