From ea3b95e20d7f5cf6f585b622b8bae0bf42f11f1c Mon Sep 17 00:00:00 2001
From: Subramanya Sastry <ssastry@wikimedia.org>
Date: Wed, 6 Sep 2017 17:47:54 -0500
Subject: [PATCH] Fix bug in dl-dt list output generation

* An open <dt> (;) should be closed when we encounter a new <dd> (:)
  char even if it is on a new line that has other nested lists inside.

* Tidy was hiding this PHP parser bug by closing a <dt> and opening
  a <dd> when given this HTML: "<dl><dt>a<ul><li>b</li></ul></dt></dl>"
  It generates "<dl><dt>a</dt><dd><ul><li>b</li></ul></dd></dl>"

  However, a HTML5 parser like RemexHTML, domino (used by Parsoid),
  or browsers don't do this fixup.

* So, what I thought was a bug in RemexHTML turned out to be a bug
  in the PHP parser that was being hidden by the use of Tidy.

* Added a regression test.

Bug: T175099
Change-Id: I6d5b225b82cecf9a43f23837ed8ec359b31aadad
---
 includes/parser/BlockLevelPass.php |  7 ++++++-
 tests/parser/parserTests.txt       | 17 ++++++++++++++++-
 2 files changed, 22 insertions(+), 2 deletions(-)

diff --git a/includes/parser/BlockLevelPass.php b/includes/parser/BlockLevelPass.php
index 599fbf61de..fab9ab7fb1 100644
--- a/includes/parser/BlockLevelPass.php
+++ b/includes/parser/BlockLevelPass.php
@@ -257,12 +257,17 @@ class BlockLevelPass {
 					$output .= $this->nextItem( $prefix[$commonPrefixLength - 1] );
 				}
 
+				# Close an open <dt> if we have a <dd> (":") starting on this line
+				if ( $this->DTopen && $commonPrefixLength > 0 && $prefix[$commonPrefixLength - 1] === ':' ) {
+					$output .= $this->nextItem( ':' );
+				}
+
 				# Open prefixes where appropriate.
 				if ( $lastPrefix && $prefixLength > $commonPrefixLength ) {
 					$output .= "\n";
 				}
 				while ( $prefixLength > $commonPrefixLength ) {
-					$char = substr( $prefix, $commonPrefixLength, 1 );
+					$char = $prefix[$commonPrefixLength];
 					$output .= $this->openList( $char );
 
 					if ( ';' === $char ) {
diff --git a/tests/parser/parserTests.txt b/tests/parser/parserTests.txt
index 3f93793263..380fe6f9b6 100644
--- a/tests/parser/parserTests.txt
+++ b/tests/parser/parserTests.txt
@@ -4332,6 +4332,21 @@ Definition Lists: Mixed Lists: Test 10
 
 !! end
 
+# This is a regression test for T175099
+# html/php+tidy is insufficient since Tidy covers up the bug.
+# But once Tidy is replaced with RemexHTML, html/php+tidy is good enough
+!! test
+Definition Lists: Mixed Lists: Test 11
+!! wikitext
+;a
+:*b
+!! html
+<dl><dt>a</dt>
+<dd>
+<ul><li>b</li></ul></dd></dl>
+
+!! end
+
 # The Parsoid team disagrees with the PHP parser's seemingly-random
 # rules regarding dd/dt on the next two tests.  Parsoid is more
 # consistent, and recognizes the shared nesting and keeps the
@@ -4339,7 +4354,7 @@ Definition Lists: Mixed Lists: Test 10
 # (And tidy again converts <dt> to <dd> before 'bar'.)
 
 !! test
-Definition Lists: Mixed Lists: Test 11
+Definition Lists: Mixed Lists: Test 12
 !! wikitext
 *#*#;*;;foo :bar
 *#*#;boo :baz
-- 
2.20.1