From 87c7ccd9bcada35f5645e727f1524f69d0345c23 Mon Sep 17 00:00:00 2001
From: Subramanya Sastry <ssastry@wikimedia.org>
Date: Tue, 20 Mar 2018 20:01:55 -0500
Subject: [PATCH] Fix whitespace trimming in headings

* b3dd3881 was trimming whitespace in wikitext as well as HTML headings
  whereas the whitespace-trimming proposal was going to leave HTML tags
  untouched.

* 30495ea1 missed this because coincidentally, the test I added there
  for HTML headings had a typo and used <h2>...<h2> instead of
  <h2>...</h2> which caused the test to magically pass.

* This patch trims whitespace in
  doHeadings (which deals with wikitext headings) instead of
  formatHeadings (which deals with all headings).

* Updated parser tests to account for this.

Change-Id: I854f20b4c39a0a8e03d70155b269de77acf02cae
---
 includes/parser/Parser.php   |  9 ++++++---
 tests/parser/parserTests.txt | 14 ++++++++------
 2 files changed, 14 insertions(+), 9 deletions(-)

diff --git a/includes/parser/Parser.php b/includes/parser/Parser.php
index e54588750b..d34257fa5b 100644
--- a/includes/parser/Parser.php
+++ b/includes/parser/Parser.php
@@ -1623,7 +1623,9 @@ class Parser {
 	public function doHeadings( $text ) {
 		for ( $i = 6; $i >= 1; --$i ) {
 			$h = str_repeat( '=', $i );
-			$text = preg_replace( "/^$h(.+)$h\\s*$/m", "<h$i>\\1</h$i>", $text );
+			// Trim non-newline whitespace from headings
+			// Using \s* will break for: "==\n===\n" and parse as <h2>=</h2>
+			$text = preg_replace( "/^(?:$h)[ \\t]*(.+?)[ \\t]*(?:$h)\\s*$/m", "<h$i>\\1</h$i>", $text );
 		}
 		return $text;
 	}
@@ -4056,10 +4058,11 @@ class Parser {
 
 		# Get all headlines for numbering them and adding funky stuff like [edit]
 		# links - this is for later, but we need the number of headlines right now
-		# This regexp also trims whitespace in the heading's content
+		# NOTE: white space in headings have been trimmed in doHeadings. They shouldn't
+		# be trimmed here since whitespace in HTML headings is significant.
 		$matches = [];
 		$numMatches = preg_match_all(
-			'/<H(?P<level>[1-6])(?P<attrib>.*?>)\s*(?P<header>[\s\S]*?)\s*<\/H[1-6] *>/i',
+			'/<H(?P<level>[1-6])(?P<attrib>.*?>)(?P<header>[\s\S]*?)<\/H[1-6] *>/i',
 			$text,
 			$matches
 		);
diff --git a/tests/parser/parserTests.txt b/tests/parser/parserTests.txt
index 3f4ecadc9e..9a4c2735f1 100644
--- a/tests/parser/parserTests.txt
+++ b/tests/parser/parserTests.txt
@@ -16941,7 +16941,8 @@ __NOEDITSECTION__
 <h1><span class="mw-headline" id="Header_1">Header 1</span></h1>
 <h2><span class="mw-headline" id="Header_1.1">Header 1.1</span></h2>
 <h2><span class="mw-headline" id="Header_1.2">Header 1.2</span></h2>
-<h1><span class="mw-headline" id="Header_2">Header 2</span></h1>
+<h1><span class="mw-headline" id="Header_2">Header 2
+</span></h1>
 <h2><span class="mw-headline" id="Header_2.1">Header 2.1</span></h2>
 <h2><span class="mw-headline" id="Header_2.2">Header 2.2</span></h2>
 
@@ -30485,7 +30486,8 @@ headings, and cells. HTML versions of the same should preserve whitespace.
 Trim whitespace in wikitext headings, list items, table captions, headings, and cells
 !! wikitext
 __NOTOC__
-==    <!--c1-->   <!--c2--> Heading    <!--c3--> <!--c4-->  ==
+==    <!--c1-->  <!--c2--> Spaces   <!--c3--> <!--c4-->  ==
+==		<!--c2-->	<!--c2--> Tabs		<!--c3--><!--c4-->	==
 *     <!--c1-->   <!--c2-->  List item <!--c3--> <!--c4-->
 ; <!--term to define--> term : <!--term's definition--> definition
 {|
@@ -30503,7 +30505,8 @@ __NOTOC__
   |  <!--c1--> <!--c2--> Table Cell 1 <!--c3--> <!--c4--> ||   Table Cell 2 <!--c5-->
   |} foo   <!--c1-->
 !! html/php+tidy
-<h2><span class="mw-headline" id="Heading">Heading</span><span class="mw-editsection"><span class="mw-editsection-bracket">[</span><a href="/index.php?title=Parser_test&amp;action=edit&amp;section=1" title="Edit section: Heading">edit</a><span class="mw-editsection-bracket">]</span></span></h2>
+<h2><span class="mw-headline" id="Spaces">Spaces</span><span class="mw-editsection"><span class="mw-editsection-bracket">[</span><a href="/index.php?title=Parser_test&amp;action=edit&amp;section=1" title="Edit section: Spaces">edit</a><span class="mw-editsection-bracket">]</span></span></h2>
+<h2><span class="mw-headline" id="Tabs">Tabs</span><span class="mw-editsection"><span class="mw-editsection-bracket">[</span><a href="/index.php?title=Parser_test&amp;action=edit&amp;section=2" title="Edit section: Tabs">edit</a><span class="mw-editsection-bracket">]</span></span></h2>
 <ul><li>List item</li></ul>
 <dl><dt>term&#160;</dt>
 <dd>definition</dd></dl>
@@ -30537,20 +30540,19 @@ __NOTOC__
 Do not trim whitespace in HTML headings, list items, table captions, headings, and cells
 !! wikitext
 __NOTOC__
-<h2>    <!--c1-->   <!--c2--> Heading    <!--c3--> <!--c4-->  <h2>
+<h2>    <!--c1-->   <!--c2--> Heading    <!--c3--> <!--c4-->  </h2>
 <ul><li>     <!--c1-->   <!--c2-->  List item <!--c3--> <!--c4-->  </li></ul>
 <table>
 <tr><th> <!--c1--> <!--c2--> Table Heading <!--c3--> <!--c4--> <th></tr>
 <tr><td> <!--c1--> <!--c2--> Table Cell <!--c3--> <!--c4--> <th></tr>
 </table>
 !! html/php+tidy
-<h2>        Heading       </h2><h2>
+<h2><span class="mw-headline" id="Heading">        Heading       </span></h2>
 <ul><li>          List item    </li></ul>
 <table>
 <tbody><tr><th>   Table Heading   </th><th></th></tr>
 <tr><td>   Table Cell   </td><th></th></tr>
 </tbody></table>
-</h2>
 !! end
 
 !! test
-- 
2.20.1