From 0880f444b1f8476dd2ab85720e5cfe4648f2f6fc Mon Sep 17 00:00:00 2001
From: Andrew Garrett <werdna@users.mediawiki.org>
Date: Wed, 11 Feb 2009 18:23:21 +0000
Subject: [PATCH] Abuse Filter Parser updates: * Use strcspn to scan ahead for
 long regions of uninteresting text in string handling (performance). * Remove
 cruft specific to my system in phpTest.php. * Remove a test that was in
 incorrect syntax, and useless without adding variable support.

---
 AbuseFilter.parser.php | 12 +++++++-----
 phpTest.php            | 12 ++++++++----
 tests/whitespace.t     |  2 --
 3 files changed, 15 insertions(+), 11 deletions(-)
 delete mode 100644 tests/whitespace.t

diff --git a/AbuseFilter.parser.php b/AbuseFilter.parser.php
index dde505a5a..3663af261 100644
--- a/AbuseFilter.parser.php
+++ b/AbuseFilter.parser.php
@@ -743,7 +743,13 @@ class AbuseFilterParser {
 					return array( $tok, AFPToken::TString, $code, $offset );
 				}
 				
-				if( $code[$offset] == '\\' ) {
+				// Performance: Use a PHP function (implemented in C)
+				//  to scan ahead.
+				$addLength = strcspn( $code, $type."\\", $offset );
+				if ($addLength) {
+					$tok .= substr( $code, $offset, $addLength );
+					$offset += $addLength;
+				} elseif( $code[$offset] == '\\' ) {
 					if( $code[$offset + 1] == '\\' )
 						$tok .= '\\';
 					elseif( $code[$offset + 1] == $type )
@@ -893,10 +899,6 @@ class AbuseFilterParser {
 			'unrecognisedtoken', $offset, array( substr( $code, $offset ) ) );
 	}
 	
-	protected static function isDigitOrDot( $chr ) {
-		return ctype_alnum( $chr ) || $chr == '.';
-	}
-	
 	protected static function isValidIdSymbol( $chr ) {
 		return ctype_alnum( $chr ) || $chr == '_';
 	}
diff --git a/phpTest.php b/phpTest.php
index b109aeda7..56a553bfb 100644
--- a/phpTest.php
+++ b/phpTest.php
@@ -4,7 +4,10 @@
  * Runs tests against the PHP parser.
  */
 
-require( '/home/andrew/mediawiki/maintenance/commandLine.inc' );	
+require_once ( getenv('MW_INSTALL_PATH') !== false
+	? getenv('MW_INSTALL_PATH')."/maintenance/commandLine.inc"
+	: dirname( __FILE__ ) . '/../../maintenance/commandLine.inc' );
+	
 $tester = new AbuseFilterParser;
 
 wfLoadExtensionMessages( 'AbuseFilter' );
@@ -16,9 +19,6 @@ $check = 0;
 $pass = 0;
 
 foreach( $tests as $test ) {
-	if( in_string( 'whitespace.t', $test ) )
-		continue;	// Skip it. Or add preset variables support to the parser
-
 	$result = substr($test,0,-2).".r";
 
 	$rule = trim(file_get_contents( $test ));
@@ -45,6 +45,10 @@ foreach( $tests as $test ) {
 		}
 	} catch (AFPException $excep) {
 		print "-FAILED - exception ".$excep->getMessage()." with input $rule\n";
+		
+		// export
+		$vars = var_export( $tester->mTokens, true );
+		file_put_contents( $test.'.parsed', $vars );
 	}
 }
 
diff --git a/tests/whitespace.t b/tests/whitespace.t
deleted file mode 100644
index cf595a46a..000000000
--- a/tests/whitespace.t
+++ /dev/null
@@ -1,2 +0,0 @@
-1 == FOO
-FOO=1