From 4c3250e0214b90f8352b2250c674a0f143c30c20 Mon Sep 17 00:00:00 2001
From: Daniel Kraus <bovender@bovender.de>
Date: Wed, 30 Aug 2017 22:23:03 +0200
Subject: [PATCH] Add tests for headings.

---
 includes/Splitter.php          | 17 ++++++++++++-----
 tests/phpunit/SplitterTest.php | 20 ++++++++++++++++++++
 2 files changed, 32 insertions(+), 5 deletions(-)

diff --git a/includes/Splitter.php b/includes/Splitter.php
index 8b8d2c8..c13952c 100644
--- a/includes/Splitter.php
+++ b/includes/Splitter.php
@@ -114,14 +114,21 @@ class Splitter {
 		// wiki headings ("= ... =", "== ... ==" etc.),
 		// urls ("http://example.com", "[http://example.com]", "[http://example.com Description]",
 		// and email addresses ("mail@example.com").
-		// Since there is a user option to skip headings, we make this part of the expression
-		// optional. Note that in order to use preg_split(), it is important to have only one
-		// capturing subpattern (which precludes the use of conditional subpatterns).
-		( $this->config->parseHeadings ) ? $delimiter = '' : $delimiter = '^=+.+?=+|';
+
+		// Match WikiText headings.
+		// Since there is a user option to skip headings, we make this part of the
+		// expression optional. Note that in order to use preg_split(), it is
+		// important to have only one capturing subpattern (which precludes the use
+		// of conditional subpatterns).
+		// Caveat: This regex pattern should be improved to deal with balanced '='s
+		// only. However, this would require grouping in the pattern which does not
+		// agree with preg_split.
+		$headingsDelimiter = $this->config->parseHeadings ? '' : '^=+[^=]+=+$|';
+
 		$urlPattern = '[a-z]+?\:\/\/(?:\S+\.)+\S+(?:\/.*)?';
 		$this->splitter = '/(' .                     // exclude from linking:
 			'\[\[.*?\]\]|' .                            // links
-			$delimiter .                                // titles (if requested)
+			$headingsDelimiter .                        // headings (if requested)
 			$templatesDelimiter .                       // templates (if requested)
 			'^ .+?\n|\n .+?\n|\n .+?$|^ .+?$|' .        // preformatted text
 			'<nowiki>.*?<.nowiki>|<code>.*?<\/code>|' . // nowiki/code
diff --git a/tests/phpunit/SplitterTest.php b/tests/phpunit/SplitterTest.php
index 2ac2450..c503b73 100644
--- a/tests/phpunit/SplitterTest.php
+++ b/tests/phpunit/SplitterTest.php
@@ -73,6 +73,26 @@ class SplitterTest extends MediaWikiTestCase {
 				'With skipTemplates = true, this may be linked {{mytemplate|param={{transcluded}}}}',
 				[ 'With skipTemplates = true, this may be linked ', '{{mytemplate|param={{transcluded}}}}', '' ]
 			],
+			[
+				true, // skipTemplates
+				true, // parseHeadings
+				"With parseHeadings = true,\n==a heading may be linked==\n",
+				[ "With parseHeadings = true,\n==a heading may be linked==\n" ]
+			],
+			[
+				true, // skipTemplates
+				false, // parseHeadings
+				// no trailing newline in the following string because it would be swallowed
+				"With parseHeadings = false,\n==a heading may not be linked==",
+				[ "With parseHeadings = false,\n", "==a heading may not be linked==", '' ]
+			],
+			// Improperly formatted headings cannot be dealt with appropriately for now
+			// [
+			// 	true, // skipTemplates
+			// 	false, // parseHeadings
+			// 	"With parseHeadings = false,\n==an improperly formatted heading may be linked=\n",
+			// 	[ "With parseHeadings = false,\n==an improperly formatted heading may be linked=\n" ]
+			// ],
 		];
 	}
 }