From 31625b54a03781f7504d67c3dba414e6ed285707 Mon Sep 17 00:00:00 2001
From: Gabriel Wicke <gwicke@wikimedia.org>
Date: Tue, 17 Apr 2012 18:58:52 +0200
Subject: [PATCH] Bug 11748: Handle optionally-closed HTML tags

* Fixes bug 11748 (Parser issue for HTML definition list) and similar
  issues for nested unordered / ordered lists

* Stops wrapping HTML-syntax definition lists into paragraphs
  for consistency with their wikitext variants

* Enables one previously disabled test and adds another for nested
  definition lists with HTML syntax

Change-Id: If75ed54e11452dbcf5e6213cc20923064f811715
---
 RELEASE-NOTES-1.21           |  3 ++-
 includes/Sanitizer.php       |  8 ++++++--
 includes/parser/Parser.php   |  4 ++--
 tests/parser/parserTests.txt | 22 +++++++++++++++++++---
 4 files changed, 29 insertions(+), 8 deletions(-)

diff --git a/RELEASE-NOTES-1.21 b/RELEASE-NOTES-1.21
index 29bef9f087..07cba04d40 100644
--- a/RELEASE-NOTES-1.21
+++ b/RELEASE-NOTES-1.21
@@ -49,7 +49,8 @@ production.
   to support Squid configured in forward-proxy mode, set 
   $wgSquidPurgeUseHostHeader to false.
 * (bug 37020) sql.php with readline eats semicolon
-
+* (bug 11748) Properly handle optionally-closed HTML tags when Tidy is
+  disabled, and don't wrap HTML-syntax definition lists in paragraphs.
 
 === API changes in 1.21 ===
 * prop=revisions can now report the contentmodel and contentformat, see docs/contenthandler.txt
diff --git a/includes/Sanitizer.php b/includes/Sanitizer.php
index 8919f10ad6..2c4ea47372 100644
--- a/includes/Sanitizer.php
+++ b/includes/Sanitizer.php
@@ -387,9 +387,9 @@ class Sanitizer {
 			$htmlsingleonly = array( # Elements that cannot have close tags
 				'br', 'hr'
 			);
-			$htmlnest = array( # Tags that can be nested--??
+			$htmlnest = array( # Tags that can be nested directly or indirectly
 				'table', 'tr', 'td', 'th', 'div', 'blockquote', 'ol', 'ul',
-				'dl', 'font', 'big', 'small', 'sub', 'sup', 'span'
+				'li', 'dl', 'dt', 'dd', 'font', 'big', 'small', 'sub', 'sup', 'span'
 			);
 			$tabletags = array( # Can only appear inside table, we will close them
 				'td', 'th', 'tr',
@@ -510,6 +510,10 @@ class Sanitizer {
 						} elseif ( isset( $htmlsingle[$t] ) ) {
 							# Hack to not close $htmlsingle tags
 							$brace = null;
+							# Still need to push this optionally-closed tag to
+							# the tag stack so that we can match end tags
+							# instead of marking them as bad.
+							array_push( $tagstack, $t );
 						} elseif ( isset( $tabletags[$t] )
 						&& in_array( $t, $tagstack ) ) {
 							// New table tag but forgot to close the previous one
diff --git a/includes/parser/Parser.php b/includes/parser/Parser.php
index 8671665ffc..b31288f58d 100644
--- a/includes/parser/Parser.php
+++ b/includes/parser/Parser.php
@@ -2377,10 +2377,10 @@ class Parser {
 				wfProfileIn( __METHOD__."-paragraph" );
 				# No prefix (not in list)--go to paragraph mode
 				# XXX: use a stack for nestable elements like span, table and div
-				$openmatch = preg_match('/(?:<table|<blockquote|<h1|<h2|<h3|<h4|<h5|<h6|<pre|<tr|<p|<ul|<ol|<li|<\\/tr|<\\/td|<\\/th)/iS', $t );
+				$openmatch = preg_match('/(?:<table|<blockquote|<h1|<h2|<h3|<h4|<h5|<h6|<pre|<tr|<p|<ul|<ol|<dl|<li|<\\/tr|<\\/td|<\\/th)/iS', $t );
 				$closematch = preg_match(
 					'/(?:<\\/table|<\\/blockquote|<\\/h1|<\\/h2|<\\/h3|<\\/h4|<\\/h5|<\\/h6|'.
-					'<td|<th|<\\/?div|<hr|<\\/pre|<\\/p|'.$this->mUniqPrefix.'-pre|<\\/li|<\\/ul|<\\/ol|<\\/?center)/iS', $t );
+					'<td|<th|<\\/?div|<hr|<\\/pre|<\\/p|'.$this->mUniqPrefix.'-pre|<\\/li|<\\/ul|<\\/ol|<\\/dl|<\\/?center)/iS', $t );
 				if ( $openmatch or $closematch ) {
 					$paragraphStack = false;
 					# TODO bug 5718: paragraph closed
diff --git a/tests/parser/parserTests.txt b/tests/parser/parserTests.txt
index dbbcd478cf..27843a31de 100644
--- a/tests/parser/parserTests.txt
+++ b/tests/parser/parserTests.txt
@@ -1202,7 +1202,6 @@ Definition lists: colon in HTML attribute
 
 !! end
 
-
 !! test
 Definition lists: self-closed tag
 !! input
@@ -1215,8 +1214,6 @@ Definition lists: self-closed tag
 
 !! test
 Bug 11748: Literal closing tags
-!! options
-disabled
 !! input
 <dl>
 <dt>test 1</dt>
@@ -1231,6 +1228,7 @@ disabled
 <dt>test 2</dt>
 <dd>test test test test test</dd>
 </dl>
+
 !! end
 
 !! test
@@ -1253,6 +1251,7 @@ Definition and unordered list using wiki syntax nested in unordered list using h
 !! end
 
 !! test
+
 Definition list with empty definition and following paragraph
 !! input
 ; term:
@@ -1264,6 +1263,23 @@ Paragraph text
 </p>
 !! end
 
+!! test
+Nested definition lists using html syntax
+!! input
+<dl><dd>
+<dl>
+<dd>Foo</dd>
+</dl>
+</dd></dl>
+!! result
+<dl><dd>
+<dl>
+<dd>Foo</dd>
+</dl>
+</dd></dl>
+
+!! end
+
 !! test
 Definition Lists: No nesting: Multiple dd's
 !! input
-- 
2.20.1