From 758325923f24dae5b0f03cc419b95d46a53fee28 Mon Sep 17 00:00:00 2001
From: Jacques Distler <distler@golem.ph.utexas.edu>
Date: Sun, 30 Nov 2008 21:44:52 -0600
Subject: [PATCH] Fix another ill-Formedness hole

The html5lib sanitizer does not necessarily produce well-formed output.
Take some "bad" input, wrap it in a <nowiki> tag and -- bingo! -- you get
ill-formed output.

Fixed. (Though, probably, one should fix the html5lib sanitizer, instead.)
---
 lib/chunks/nowiki.rb            | 11 ++++++++++-
 test/unit/chunks/nowiki_test.rb |  6 ++++++
 2 files changed, 16 insertions(+), 1 deletion(-)

diff --git a/lib/chunks/nowiki.rb b/lib/chunks/nowiki.rb
index 6d9641ef..f3cee86f 100644
--- a/lib/chunks/nowiki.rb
+++ b/lib/chunks/nowiki.rb
@@ -1,5 +1,6 @@
 require 'chunks/chunk'
 require 'sanitize'
+require 'rexml/document'
 
 # This chunks allows certain parts of a wiki page to be hidden from the
 # rest of the rendering pipeline. It should be run at the beginning
@@ -26,7 +27,15 @@ class NoWiki < Chunk::Abstract
 
   def initialize(match_data, content)
     super
-    @plain_text = @unmask_text = sanitize_xhtml(match_data[1])
+    begin
+      sanitized = sanitize_xhtml(match_data[1])
+      doc = REXML::Document.new("<div xmlns='http://www.w3.org/1999/xhtml'>#{sanitized}</div>")
+      sanitized = doc.to_s.gsub(/\A<div xmlns='http:\/\/www.w3.org\/1999\/xhtml'>(.*)<\/div>\Z/m, '\1')
+    rescue REXML::ParseException
+      sanitized = %{<pre class='markdown-html-error' style='border: solid 3px red; background-color: pink;'>HTML parse error:
+#{sanitized.escapeHTML}</pre>}
+    end
+    @plain_text = @unmask_text = sanitized
   end
 
 end
diff --git a/test/unit/chunks/nowiki_test.rb b/test/unit/chunks/nowiki_test.rb
index a8915ef3..8c068b63 100755
--- a/test/unit/chunks/nowiki_test.rb
+++ b/test/unit/chunks/nowiki_test.rb
@@ -24,4 +24,10 @@ class NoWikiTest < Test::Unit::TestCase
 	)
   end
 
+  def test_sanitize_nowiki_ill_formed
+    match(NoWiki, "<nowiki><animateColor xlink:href='#foo'/></nowiki>",
+                :plain_text => "<pre class='markdown-html-error' style='border: solid 3px red; background-color: pink;'>HTML parse error:\n&lt;animateColor xlink:href=&#39;#foo&#39;&gt;&lt;/animateColor&gt;</pre>"
+    )
+  end
+
 end