From caa558a25886638427d3005030260af8518612ea Mon Sep 17 00:00:00 2001
From: Jan Klimo <jan.klimo@gmail.com>
Date: Tue, 20 Mar 2018 01:18:17 +0700
Subject: [PATCH] Optimize memory usage

---
 lib/sanitize.rb                            |  6 ++++--
 lib/sanitize/transformers/clean_element.rb |  2 +-
 test/test_clean_element.rb                 | 19 ++++++++++++++++++-
 3 files changed, 23 insertions(+), 4 deletions(-)

diff --git a/lib/sanitize.rb b/lib/sanitize.rb
index 4dc1174..935fcaa 100644
--- a/lib/sanitize.rb
+++ b/lib/sanitize.rb
@@ -198,7 +198,7 @@ def to_html(node)
       # the original document didn't actually include a content-type meta tag.
       replace_meta = !@config[:elements].include?('meta') ||
         node.xpath('/html/head/meta[@http-equiv]').none? do |meta|
-          meta['http-equiv'].downcase == 'content-type'
+          meta['http-equiv'].casecmp('content-type').zero?
         end
     end
 
@@ -217,12 +217,14 @@ def to_html(node)
   end
 
   def transform_node!(node, node_whitelist)
+    node_name = node.name.downcase
+
     @transformers.each do |transformer|
       result = transformer.call(
         :config         => @config,
         :is_whitelisted => node_whitelist.include?(node),
         :node           => node,
-        :node_name      => node.name.downcase,
+        :node_name      => node_name,
         :node_whitelist => node_whitelist
       )
 
diff --git a/lib/sanitize/transformers/clean_element.rb b/lib/sanitize/transformers/clean_element.rb
index f883d52..3431be4 100644
--- a/lib/sanitize/transformers/clean_element.rb
+++ b/lib/sanitize/transformers/clean_element.rb
@@ -99,7 +99,7 @@ def call(env)
           if @protocols.include?(name) && @protocols[name].include?(attr_name)
             attr_protocols = @protocols[name][attr_name]
 
-            if attr.value.to_s.downcase =~ REGEX_PROTOCOL
+            if attr.value =~ REGEX_PROTOCOL
               attr.unlink unless attr_protocols.include?($1.downcase)
             else
               attr.unlink unless attr_protocols.include?(:relative)
diff --git a/test/test_clean_element.rb b/test/test_clean_element.rb
index 6355d78..562464c 100644
--- a/test/test_clean_element.rb
+++ b/test/test_clean_element.rb
@@ -402,6 +402,23 @@
       s.fragment('foo<div>bar</div>baz').must_equal "foo\nbar\nbaz"
       s.fragment('foo<br>bar<br>baz').must_equal "foo\nbar\nbaz"
     end
-  end
 
+    it 'handles protocols correctly regardless of case' do
+      input = '<a href="hTTpS://foo.com/">Text</a>'
+
+      Sanitize.fragment(input, {
+        :elements   => ['a'],
+        :attributes => {'a' => ['href']},
+        :protocols  => {'a' => {'href' => ['https']}}
+      }).must_equal input
+
+      input = '<a href="mailto:someone@example.com?Subject=Hello">Text</a>'
+
+      Sanitize.fragment(input, {
+        :elements   => ['a'],
+        :attributes => {'a' => ['href']},
+        :protocols  => {'a' => {'href' => ['https']}}
+      }).must_equal "<a>Text</a>"
+    end
+  end
 end