Changeset 58613 for trunk/src/wp-includes/html-api/class-wp-html-decoder.php

Timestamp:

07/01/2024 11:34:19 PM (5 days ago)

Author:

dmsnell

Message:

HTML API: Optimize low-level parsing details in Tag Processor.

Introduces a number of micro-level optimizations in the Tag Processor to
improve token-scanning performance. Should contain no functional changes.

Based on benchmarking against a list of the 100 most-visited websites,
these changes result in an average improvement in performance of the Tag
Processor for scanning tags from between 3.5% and 7.5%.

Developed in ​https://github.com/WordPress/wordpress-develop/pull/6890
Discussed in https://core.trac.wordpress.org/ticket/61545

Follow-up to [55203].

See #61545.

File:

• trunk/src/wp-includes/html-api/class-wp-html-decoder.php (modified) (2 diffs)

trunk/src/wp-includes/html-api/class-wp-html-decoder.php

@@ -142 +142 @@
         while ( $at < $end ) {
             $next_character_reference_at = strpos( $text, '&', $at );
-            if ( false === $next_character_reference_at || $next_character_reference_at >= $end ) {
+            if ( false === $next_character_reference_at ) {
                 break;
             }
@@ -437 +437 @@
 
         if ( $code_point <= 0x7FF ) {
-            $byte1 = ( $code_point >> 6 ) | 0xC0;
-            $byte2 = $code_point & 0x3F | 0x80;
-
-            return pack( 'CC', $byte1, $byte2 );
+            $byte1 = chr( ( $code_point >> 6 ) | 0xC0 );
+            $byte2 = chr( $code_point & 0x3F | 0x80 );
+
+            return "{$byte1}{$byte2}";
         }
 
         if ( $code_point <= 0xFFFF ) {
-            $byte1 = ( $code_point >> 12 ) | 0xE0;
-            $byte2 = ( $code_point >> 6 ) & 0x3F | 0x80;
-            $byte3 = $code_point & 0x3F | 0x80;
-
-            return pack( 'CCC', $byte1, $byte2, $byte3 );
+            $byte1 = chr( ( $code_point >> 12 ) | 0xE0 );
+            $byte2 = chr( ( $code_point >> 6 ) & 0x3F | 0x80 );
+            $byte3 = chr( $code_point & 0x3F | 0x80 );
+
+            return "{$byte1}{$byte2}{$byte3}";
         }
 
         // Any values above U+10FFFF are eliminated above in the pre-check.
-        $byte1 = ( $code_point >> 18 ) | 0xF0;
-        $byte2 = ( $code_point >> 12 ) & 0x3F | 0x80;
-        $byte3 = ( $code_point >> 6 ) & 0x3F | 0x80;
-        $byte4 = $code_point & 0x3F | 0x80;
-
-        return pack( 'CCCC', $byte1, $byte2, $byte3, $byte4 );
+        $byte1 = chr( ( $code_point >> 18 ) | 0xF0 );
+        $byte2 = chr( ( $code_point >> 12 ) & 0x3F | 0x80 );
+        $byte3 = chr( ( $code_point >> 6 ) & 0x3F | 0x80 );
+        $byte4 = chr( $code_point & 0x3F | 0x80 );
+
+        return "{$byte1}{$byte2}{$byte3}{$byte4}";
     }
 }