Changeset 6314

Timestamp:

11/05/2007 05:13:43 PM (17 years ago)

Author:

ryan

Message:

Improve performance of seems_utf8() and uf8_uri_encode(). fixes #5320

File:

• trunk/wp-includes/formatting.php (modified) (4 diffs)

trunk/wp-includes/formatting.php

@@ -95 +95 @@
 
 function seems_utf8($Str) { # by bmorel at ssi dot fr
-    for ($i=0; $i<strlen($Str); $i++) {
+    $length = strlen($Str);
+    for ($i=0; $i < $length; $i++) {
         if (ord($Str[$i]) < 0x80) continue; # 0bbbbbbb
         elseif ((ord($Str[$i]) & 0xE0) == 0xC0) $n=1; # 110bbbbb
@@ -104 +105 @@
         else return false; # Does not match any model
         for ($j=0; $j<$n; $j++) { # n bytes matching 10bbbbbb follow ?
-            if ((++$i == strlen($Str)) || ((ord($Str[$i]) & 0xC0) != 0x80))
+            if ((++$i == $length) || ((ord($Str[$i]) & 0xC0) != 0x80))
             return false;
         }
@@ -133 +134 @@
     $values = array();
     $num_octets = 1;
-
-    for ($i = 0; $i < strlen( $utf8_string ); $i++ ) {
+    $unicode_length = 0;
+
+    $string_length = strlen( $utf8_string );
+    for ($i = 0; $i < $string_length; $i++ ) {
 
         $value = ord( $utf8_string[ $i ] );
 
         if ( $value < 128 ) {
-            if ( $length && ( strlen($unicode) + 1 > $length ) )
+            if ( $length && ( $unicode_length >= $length ) )
                 break;
             $unicode .= chr($value);
+            $unicode_length++;
         } else {
             if ( count( $values ) == 0 ) $num_octets = ( $value < 224 ) ? 2 : 3;
@@ -147 +151 @@
             $values[] = $value;
 
-            if ( $length && ( (strlen($unicode) + ($num_octets * 3)) > $length ) )
+            if ( $length && ( $unicode_length + ($num_octets * 3) ) > $length )
                 break;
             if ( count( $values ) == $num_octets ) {
                 if ($num_octets == 3) {
                     $unicode .= '%' . dechex($values[0]) . '%' . dechex($values[1]) . '%' . dechex($values[2]);
+                    $unicode_length += 9;
                 } else {
                     $unicode .= '%' . dechex($values[0]) . '%' . dechex($values[1]);
+                    $unicode_length += 6;
                 }