improved html_escape to favor lower case entities

mojolicious · Sep 27, 2012 · 19b74a6 · 19b74a6
1 parent 5c8169c
commit 19b74a6
Show file tree

Hide file tree

Showing 5 changed files with 9 additions and 6 deletions.
diff --git a/Changes b/Changes
@@ -1,5 +1,6 @@
 
 3.44  2012-09-27
+  - Improved html_escape to favor lower case entities. (judofyr)
   - Improved javascript and stylesheet helpers to not generate type
     attributes.
   - Improved documentation.

diff --git a/lib/Mojo/DOM.pm b/lib/Mojo/DOM.pm
@@ -433,7 +433,7 @@ use it for validation.
 =head1 CASE SENSITIVITY
 
 L<Mojo::DOM> defaults to HTML semantics, that means all tags and attributes
-are lowercased and selectors need to be lowercase as well.
+are lowercased and selectors need to be lower case as well.
 
   my $dom = Mojo::DOM->new('<P ID="greeting">Hi!</P>');
   say $dom->at('p')->text;

diff --git a/lib/Mojo/Util.pm b/lib/Mojo/Util.pm
@@ -33,7 +33,9 @@ my %ENTITIES;
 
 # Reverse entities for html_escape (without "apos")
 my %REVERSE = ("\x{0027}" => '#39;');
-$REVERSE{$ENTITIES{$_}} //= $_ for sort grep {/;/} keys %ENTITIES;
+$REVERSE{$ENTITIES{$_}} //= $_
+  for sort { split(/[A-Z]/, $a) <=> split(/[A-Z]/, $b) }
+  sort grep {/;/} keys %ENTITIES;
 
 our @EXPORT_OK = (
   qw(b64_decode b64_encode camelize class_to_file class_to_path decamelize),

diff --git a/t/mojo/bytestream.t b/t/mojo/bytestream.t
@@ -33,7 +33,7 @@ is b('foo%C3%9F%C4%80bar%E2%98%BA')->url_unescape->decode('UTF-8'),
   "foo\x{df}\x{0100}bar\x{263a}", 'right url unescaped result';
 
 # html_escape
-is b("foo bar'<baz>")->html_escape, 'foo bar&#39;&LT;baz&GT;',
+is b("foo bar'<baz>")->html_escape, 'foo bar&#39;&lt;baz&gt;',
   'right html escaped result';
 
 # html_unescape

diff --git a/t/mojo/util.t b/t/mojo/util.t
@@ -103,7 +103,7 @@ is decode('UTF-8', url_unescape 'foo%C3%9F%C4%80bar%E2%98%BA'),
   "foo\x{df}\x{0100}bar\x{263a}", 'right url unescaped result';
 
 # html_escape
-is html_escape("foo bar'<baz>"), 'foo bar&#39;&LT;baz&GT;',
+is html_escape("foo bar'<baz>"), 'foo bar&#39;&lt;baz&gt;',
   'right html escaped result';
 
 # html_escape (nothing to escape)
@@ -131,7 +131,7 @@ is html_unescape('&Ltf&amp&0oo&nbspba;&ltr'), "&Ltf&&0oo\x{00a0}ba;<r",
 
 # UTF-8 html_escape
 is html_escape("fo\nobar<baz>&\"\x{152}\x{02ae4}"),
-  "fo\nobar&LT;baz&GT;&AMP;&QUOT;&OElig;&Dashv;", 'right html escaped result';
+  "fo\nobar&lt;baz&gt;&amp;&quot;&OElig;&Dashv;", 'right html escaped result';
 
 # UTF-8 html_unescape
 is html_unescape(decode 'UTF-8', 'foo&lt;baz&gt;&#x26;&#34;&OElig;&Foo;'),
@@ -142,7 +142,7 @@ is html_escape('/home/sri/perl/site_perl/5.10.0/Mojo.pm'),
   '/home/sri/perl/site_perl/5.10.0/Mojo.pm', 'right html escaped result';
 
 # html_escape (custom pattern)
-is html_escape("fo\no b<a>r", 'o<'), "f&#111;\n&#111; b&LT;a>r",
+is html_escape("fo\no b<a>r", 'o<'), "f&#111;\n&#111; b&lt;a>r",
   'right html escaped result';
 
 # xml_escape