@@ -53,7 +53,7 @@ class Sanitize
# or more characters followed by a colon is considered a match, even if the
# colon is encoded as an entity and even if it's an incomplete entity (which
# IE6 and Opera will still parse).
- REGEX_PROTOCOL = /^([^:]*)(?:\:|�*58|�*3a)/i
+ REGEX_PROTOCOL = /^([A-Za-z0-9\+\-\.\&\;\#\s]*?)(?:\:|�*58|�*3a)/i
#--
# Instance Methods
lib/sanitize.rb
@@ -248,6 +248,11 @@ describe 'Custom configs' do
Sanitize.clean(input, {:elements => ['p'], :attributes => {'div' => ['class']}}).should.equal('<p>bar</p>')
Sanitize.clean(input, {:elements => ['p'], :attributes => {'p' => ['title'], :all => ['class']}}).should.equal(input)
end
+
+ should 'allow relative URLs containing colons where the colon is not in the first path segment' do
+ input = '<a href="/wiki/Special:Random">Random Page</a>'
+ Sanitize.clean(input, { :elements => ['a'], :attributes => {'a' => ['href']}, :protocols => { 'a' => { 'href' => [:relative] }} }).should.equal(input)
+ end
end
describe 'Sanitize.clean' do
test/spec_sanitize.rb
5d01e561fa1c554ea46b621d77f2e278c7771050
peter
peter@biggie.local
http://github.com/rgrove/sanitize/commit/8550b8afae09ad5a22f70c3d2e4838cbfeebdcfc
8550b8afae09ad5a22f70c3d2e4838cbfeebdcfc
2009-05-19T18:29:16-07:00
2009-05-19T18:29:16-07:00
Changed protocol regex to ensure Sanitize doesn't kill URLs with colons in path segments
e31088cadc587d34c7e8c01ec3ba14110c1ffc1a
peter
peter@biggie.local