From 4bdc8a3fdcc3111c1e2f7de983faaaf5bb6fffb1 Mon Sep 17 00:00:00 2001
From: Yorick Peterse <yorickpeterse@gmail.com>
Date: Wed, 8 Apr 2015 14:30:02 +0200
Subject: [PATCH] Don't convert entities in script/style elements

In HTML the text of a script/style tag should be left untouched, no
entities must be converted. Doing so would break Javascript such as the
following:

    foo&&bar;

Such code is often the result of minifiers doing their dirty business.
---
 lib/oga/xml/text.rb       | 28 +++++++++++++++++++++-------
 spec/oga/xml/text_spec.rb | 32 ++++++++++++++++++++++++++++++++
 2 files changed, 53 insertions(+), 7 deletions(-)

diff --git a/lib/oga/xml/text.rb b/lib/oga/xml/text.rb
index 6d8115c..a9eb796 100644
--- a/lib/oga/xml/text.rb
+++ b/lib/oga/xml/text.rb
@@ -26,7 +26,7 @@ module Oga
       # @return [String]
       #
       def text
-        unless @decoded
+        if decode_entities?
           @text    = EntityDecoder.try_decode(@text, html?)
           @decoded = true
         end
@@ -38,15 +38,29 @@ module Oga
       # @see [Oga::XML::CharacterNode#to_xml]
       #
       def to_xml
-        node = parent
-
-        if node.is_a?(Element) and html? \
-        and Lexer::LITERAL_HTML_ELEMENTS.include?(node.name)
-          return super
-        end
+        return super if inside_literal_html?
 
         return Entities.encode(super)
       end
+
+      private
+
+      ##
+      # @return [TrueClass|FalseClass]
+      #
+      def decode_entities?
+        return !@decoded && !inside_literal_html?
+      end
+
+      ##
+      # @return [TrueClass|FalseClass]
+      #
+      def inside_literal_html?
+        node = parent
+
+        return node.is_a?(Element) && html? &&
+          Lexer::LITERAL_HTML_ELEMENTS.include?(node.name)
+      end
     end # Text
   end # XML
 end # Oga
diff --git a/spec/oga/xml/text_spec.rb b/spec/oga/xml/text_spec.rb
index b29ef2a..95998af 100644
--- a/spec/oga/xml/text_spec.rb
+++ b/spec/oga/xml/text_spec.rb
@@ -85,6 +85,38 @@ describe Oga::XML::Text do
         node.text.should == [160].pack('U')
       end
     end
+
+    describe 'inside an HTML script element' do
+      before do
+        @element  = Oga::XML::Element.new(:name => 'script')
+        @document = Oga::XML::Document.new(
+          :type     => :html,
+          :children => [@element]
+        )
+      end
+
+      it 'does not decode any entities' do
+        @element.inner_text = '&foo;'
+
+        @element.inner_text.should == '&foo;'
+      end
+    end
+
+    describe 'inside an HTML style element' do
+      before do
+        @element  = Oga::XML::Element.new(:name => 'style')
+        @document = Oga::XML::Document.new(
+          :type     => :html,
+          :children => [@element]
+        )
+      end
+
+      it 'does not decode any entities' do
+        @element.inner_text = '&foo;'
+
+        @element.inner_text.should == '&foo;'
+      end
+    end
   end
 
   describe '#to_xml' do