Merge pull request #311 from discomrade/arabic-fix

Fix combining character set, allow limit for combining characters

inc/config.php

@@ -484,8 +484,11 @@
 
     // Strip superfluous new lines at the end of a post.
     $config['strip_superfluous_returns'] = true;
-    // Strip combining characters from Unicode strings (eg. "Zalgo").
+    // Strip combining characters from Unicode strings (eg. "Zalgo").  This will impact some non-English languages.
     $config['strip_combining_chars'] = true;
+    // Maximum number of combining characters in a row allowed in Unicode strings so that they can still be used in moderation.
+    // Requires $config['strip_combining_chars'] = true;
+    $config['max_combining_chars'] = 0;
 
     // Maximum post body length.
     $config['max_body'] = 1800;

inc/functions.php

@@ -2257,19 +2257,11 @@ function ordutf8($string, &$offset) {
 	return $code;
 }
 
+// Limit Non_Spacing_Mark and Enclosing_Mark characters
 function strip_combining_chars($str) {
-	$chars = preg_split('//u', $str, -1, PREG_SPLIT_NO_EMPTY);
-	$str = '';
-	foreach ($chars as $char) {
-		$o = 0;
-		$ord = ordutf8($char, $o);
-
-		if ( ($ord >= 768 && $ord <= 879) || ($ord >= 1536 && $ord <= 1791) || ($ord >= 3655 && $ord <= 3659) || ($ord >= 7616 && $ord <= 7679) || ($ord >= 8400 && $ord <= 8447) || ($ord >= 65056 && $ord <= 65071))
-			continue;
-
-		$str .= $char;
-	}
-	return $str;
+	global $config;
+	$limit = strval($config['max_combining_chars']+1);
+	return preg_replace('/(\p{Me}|\p{Mn}){'.$limit.',}/u','', $str);
 }
 
 function buildThread($id, $return = false, $mod = false) {

inc/instance-config.php

@@ -106,6 +106,10 @@ $config['threads_preview'] = 5;
 $config['root'] = '/';
 $config['secure_trip_salt'] = 'ODQ2NDM0ODlmMmRhNzk2M2EyNjJlOW';
 
+$config['strip_combining_chars'] = true;
+// Maximum number of combining characters in a row allowed so that they can still be used in moderation.
+$config['max_combining_chars'] = 3;
+
 //Banners
 $config['url_banner'] = '/banners.php';