From e42ff9ee7908a968d4a53c1587938a2de2e8e544 Mon Sep 17 00:00:00 2001
From: Tim Starling <tstarling@users.mediawiki.org>
Date: Sun, 29 Apr 2007 08:28:34 +0000
Subject: [PATCH] =?utf8?q?Added=20character=20entity=20aliases=20&=D7=A8?=
 =?utf8?q?=D7=9C=D7=9E;=20and=20&=D8=B1=D9=84=D9=85;=20as=20per=20http://l?=
 =?utf8?q?ists.wikimedia.org/pipermail/wikitech-l/2007-April/031064.html?=
MIME-Version: 1.0
Content-Type: text/plain; charset=utf8
Content-Transfer-Encoding: 8bit

---
 includes/Sanitizer.php | 28 ++++++++++++++++++++++------
 1 file changed, 22 insertions(+), 6 deletions(-)

diff --git a/includes/Sanitizer.php b/includes/Sanitizer.php
index 572dafddec..fa5416dc26 100644
--- a/includes/Sanitizer.php
+++ b/includes/Sanitizer.php
@@ -28,7 +28,7 @@
  * Sanitizer::normalizeCharReferences and Sanitizer::decodeCharReferences
  */
 define( 'MW_CHAR_REFS_REGEX',
-	'/&([A-Za-z0-9]+);
+	'/&([A-Za-z0-9\x80-\xff]+);
 	 |&\#([0-9]+);
 	 |&\#x([0-9A-Za-z]+);
 	 |&\#X([0-9A-Za-z]+);
@@ -315,6 +315,16 @@ $wgHtmlEntities = array(
 	'zwj'      => 8205,
 	'zwnj'     => 8204 );
 
+/**
+ * Character entity aliases accepted by MediaWiki
+ */
+global $wgHtmlEntityAliases;
+$wgHtmlEntityAliases = array(
+	'רלמ' => 'rlm',
+	'رلم' => 'rlm',
+);
+
+
 /**
  * XHTML sanitizer for MediaWiki
  * @addtogroup Parser
@@ -902,16 +912,19 @@ class Sanitizer {
 
 	/**
 	 * If the named entity is defined in the HTML 4.0/XHTML 1.0 DTD,
-	 * return the named entity reference as is. Otherwise, returns
-	 * HTML-escaped text of pseudo-entity source (eg &amp;foo;)
+	 * return the named entity reference as is. If the entity is a 
+	 * MediaWiki-specific alias, returns the HTML equivalent. Otherwise, 
+	 * returns HTML-escaped text of pseudo-entity source (eg &amp;foo;)
 	 *
 	 * @param string $name
 	 * @return string
 	 * @static
 	 */
 	static function normalizeEntity( $name ) {
-		global $wgHtmlEntities;
-		if( isset( $wgHtmlEntities[$name] ) ) {
+		global $wgHtmlEntities, $wgHtmlEntityAliases;
+		if ( isset( $wgHtmlEntityAliases[$name] ) ) {
+			return "&{$wgHtmlEntityAliases[$name]};";
+		} elseif( isset( $wgHtmlEntities[$name] ) ) {
 			return "&$name;";
 		} else {
 			return "&amp;$name;";
@@ -1008,7 +1021,10 @@ class Sanitizer {
 	 * @return string
 	 */
 	static function decodeEntity( $name ) {
-		global $wgHtmlEntities;
+		global $wgHtmlEntities, $wgHtmlEntityAliases;
+		if ( isset( $wgHtmlEntityAliases[$name] ) ) {
+			$name = $wgHtmlEntityAliases[$name];
+		}
 		if( isset( $wgHtmlEntities[$name] ) ) {
 			return codepointToUtf8( $wgHtmlEntities[$name] );
 		} else {
-- 
2.20.1