From e47082d6c2279fb6151f057d459f1e0bad33fd91 Mon Sep 17 00:00:00 2001
From: Daniel Friesen <pub-github@nadir-seen-fire.com>
Date: Mon, 6 Aug 2012 03:02:49 -0700
Subject: [PATCH] (bug 39067) Add support for HTML5 <mark> element.

* whitelist <mark> in tidy and sanitizer
* provides a default styling for mark elements

Change-Id: I23fc2fc558ff0590be04771ef1e75fcfdf240aac
---
 RELEASE-NOTES-1.21                       |  1 +
 includes/Sanitizer.php                   |  3 +-
 includes/tidy.conf                       |  2 +-
 skins/common/shared.css                  |  6 ++++
 tests/phpunit/includes/SanitizerTest.php | 40 ++++++++++++++++++++++++
 5 files changed, 50 insertions(+), 2 deletions(-)

diff --git a/RELEASE-NOTES-1.21 b/RELEASE-NOTES-1.21
index 66b3927c3a..70c6bf86b9 100644
--- a/RELEASE-NOTES-1.21
+++ b/RELEASE-NOTES-1.21
@@ -39,6 +39,7 @@ production.
 * (bug 7851) Implemented one-click AJAX patrolling.
 * The <data>, <time>, <meta>, and <link> elements are allowed within WikiText
   for use with Microdata.
+* The HTML5 <mark> tag has been whitelisted.
 
 === Bug fixes in 1.21 ===
 * (bug 40353) SpecialDoubleRedirect should support interwiki redirects.
diff --git a/includes/Sanitizer.php b/includes/Sanitizer.php
index c529417d29..7bb3d939e4 100644
--- a/includes/Sanitizer.php
+++ b/includes/Sanitizer.php
@@ -385,7 +385,7 @@ class Sanitizer {
 				'kbd', 'samp'
 			);
 			if ( $wgHtml5 ) {
-				$htmlpairsStatic = array_merge( $htmlpairsStatic, array( 'data', 'time' ) );
+				$htmlpairsStatic = array_merge( $htmlpairsStatic, array( 'data', 'time', 'mark' ) );
 			}
 			$htmlsingle = array(
 				'br', 'hr', 'li', 'dt', 'dd'
@@ -1696,6 +1696,7 @@ class Sanitizer {
 			$whitelist += array(
 				'data' => array_merge( $common, array( 'value' ) ),
 				'time' => array_merge( $common, array( 'datetime' ) ),
+				'mark' => $common,
 
 				// meta and link are only permitted by removeHTMLtags when Microdata
 				// is enabled so we don't bother adding a conditional to hide these
diff --git a/includes/tidy.conf b/includes/tidy.conf
index bced6fd6c6..6c947295a6 100644
--- a/includes/tidy.conf
+++ b/includes/tidy.conf
@@ -19,4 +19,4 @@ fix-uri: no
 # Don't strip html5 elements we support
 # html-{meta,link} is a hack we use to prevent Tidy from stripping <meta> and <link> used in the body for Microdata
 new-empty-tags: html-meta, html-link
-new-inline-tags: video, audio, source, track, bdi, data, time
+new-inline-tags: video, audio, source, track, bdi, data, time, mark
diff --git a/skins/common/shared.css b/skins/common/shared.css
index ed4e8c55a1..34d3357c43 100644
--- a/skins/common/shared.css
+++ b/skins/common/shared.css
@@ -31,6 +31,12 @@
 	direction: rtl;
 }
 
+/* User-Agent styles for new HTML5 elements */
+mark {
+	background-color: yellow;
+	color: black;
+}
+
 /* Input types that should follow user direction, like buttons */
 /* TODO: What about buttons in wikipage content ? */
 input[type="submit"],
diff --git a/tests/phpunit/includes/SanitizerTest.php b/tests/phpunit/includes/SanitizerTest.php
index dc672ba534..6e618e392d 100644
--- a/tests/phpunit/includes/SanitizerTest.php
+++ b/tests/phpunit/includes/SanitizerTest.php
@@ -64,6 +64,46 @@ class SanitizerTest extends MediaWikiTestCase {
 		$this->assertEquals( UTF8_REPLACEMENT, Sanitizer::decodeCharReferences( "&#88888888888888;" ), 'Invalid numbered entity' );
 	}
 
+	/**
+	 * @cover Sanitizer::removeHTMLtags
+	 * @dataProvider provideHtml5Tags
+	 *
+	 * @param String $tag Name of an HTML5 element (ie: 'video')
+	 * @param Boolean $escaped Wheter sanitizer let the tag in or escape it (ie: '&lt;video&gt;')
+	 */
+	function testRemovehtmltagsOnHtml5Tags( $tag, $escaped ) {
+
+		# Enable HTML5 mode
+		global $wgHTML5;
+		$save = $wgHTML5;
+		$wgHTML5 = true;
+
+		if( $escaped ) {
+			$this->assertEquals( "&lt;$tag&gt;",
+				Sanitizer::removeHTMLtags( "<$tag>" )
+			);
+		} else {
+			$this->assertEquals( "<$tag></$tag>\n",
+				Sanitizer::removeHTMLtags( "<$tag>" )
+			);
+		}
+		$wgHTML5 = $save;
+	}
+
+	/**
+	 * Provide HTML5 tags
+	 */
+	function provideHtml5Tags() {
+		$ESCAPED  = true; # We want tag to be escaped
+		$VERBATIM = false;  # We want to keep the tag
+		return array(
+			array( 'data', $VERBATIM ),
+			array( 'mark', $VERBATIM ),
+			array( 'time', $VERBATIM ),
+			array( 'video', $ESCAPED ),
+		);
+	}
+
 	function testSelfClosingTag() {
 		$GLOBALS['wgUseTidy'] = false;
 		$this->assertEquals(
-- 
2.20.1