Update html-scanner to handle CDATA sections better. Closes #2970. [Jamis Buck]

git-svn-id: http://svn-commit.rubyonrails.org/rails/trunk@3153 5ecf4fe2-1ee6-0310-87b1-e25e094e27de

2 files changed, 16 insertions, 0 deletions

diff --git a/actionpack/lib/action_controller/vendor/html-scanner/html/node.rb b/actionpack/lib/action_controller/vendor/html-scanner/html/node.rb
index 4f02e2d421..fb961570c4 100644
--- a/actionpack/lib/action_controller/vendor/html-scanner/html/node.rb
+++ b/actionpack/lib/action_controller/vendor/html-scanner/html/node.rb
@@ -150,6 +150,11 @@ module HTML #:nodoc:
             end
           end
 
+          if scanner.skip(/!\[CDATA\[/)
+            scanner.scan_until(/\]\]>/)
+            return CDATA.new(parent, line, pos, scanner.pre_match)
+          end
+          
           closing = ( scanner.scan(/\//) ? :close : nil )
           return Text.new(parent, line, pos, content) unless name = scanner.scan(/[\w:]+/)
           name.downcase!
@@ -256,6 +261,14 @@ module HTML #:nodoc:
       content == node.content
     end
   end
+  
+  # A CDATA node is simply a text node with a specialized way of displaying
+  # itself.
+  class CDATA < Text
+    def to_s
+      "<![CDATA[#{super}]>"
+    end
+  end
 
   # A Tag is any node that represents markup. It may be an opening tag, a
   # closing tag, or a self-closing tag. It has a name, and may have a hash of
diff --git a/actionpack/lib/action_controller/vendor/html-scanner/html/tokenizer.rb b/actionpack/lib/action_controller/vendor/html-scanner/html/tokenizer.rb
index ce9d3b2800..b950e84628 100644
--- a/actionpack/lib/action_controller/vendor/html-scanner/html/tokenizer.rb
+++ b/actionpack/lib/action_controller/vendor/html-scanner/html/tokenizer.rb
@@ -52,6 +52,9 @@ module HTML #:nodoc:
         if @scanner.scan(/!--/) # comment
           tag << @scanner.matched
           tag << (@scanner.scan_until(/--\s*>/) || @scanner.scan_until(/\Z/))
+        elsif @scanner.scan(/!\[CDATA\[/)
+          tag << @scanner.matched
+          tag << @scanner.scan_until(/\]\]>/)
         elsif @scanner.scan(/!/) # doctype
           tag << @scanner.matched
           tag << consume_quoted_regions