Performance

My REXML::Element.to_ncr (and REXML::Element.to_utf8) is horribly slow. For long documents, it proves more efficient to serialize to a string, apply String.to_ncr (or String.to_utf8) and then Sanitize the string.

test/unit/sanitize_test.rb

@@ -0,0 +1,32 @@
+#!/usr/bin/env ruby
+
+require File.expand_path(File.dirname(__FILE__) + '/../test_helper')
+require 'sanitize'
+
+class SanitizeTest < Test::Unit::TestCase
+
+  def setup
+
+  end
+
+  def rexml_doc(string)
+    REXML::Document.new(
+      "<div xmlns='http://www.w3.org/1999/xhtml'>#{string}</div>")
+  end
+  
+  def my_rex(string)
+    sanitize_rexml(rexml_doc(string)).gsub(/\A<div xmlns="http:\/\/www.w3.org\/1999\/xhtml">(.*)<\/div>\Z/m, '\1')
+  end
+
+  def test_sanitize_named_entities
+    input = '<p>Greek &phi;, double-struck &Aopf;, numeric &#x1D538; &#8279;</p>'
+    output = "<p>Greek \317\225, double-struck \360\235\224\270, numeric \360\235\224\270 \342\201\227</p>"
+    output2 = "<p>Greek \317\225, double-struck \360\235\224\270, numeric &#x1D538; &#8279;</p>"
+    assert_equal(output, sanitize_xhtml(input))
+    assert_equal(output, sanitize_html(input))
+    assert_equal(output, my_rex(input))
+    assert_equal(output2, input.to_utf8)
+  end
+
+
+end