Changeset 34747 for trunk/src/wp-includes/shortcodes.php

Timestamp:

10/01/2015 06:04:13 PM (10 years ago)

Author:

wonderboymusic

Message:

Shortcodes: Fix PCRE performance bugs in get_shortcode_regexp() and related to wptexturize(), do_shortcode(), and strip_shortcodes()

Alters unit tests.

Props miqrogroove.
Fixes #33517.

File:

• trunk/src/wp-includes/shortcodes.php (modified) (5 diffs)

trunk/src/wp-includes/shortcodes.php

@@ -209 +209 @@
         return $content;
 
-    $tagnames = array_keys($shortcode_tags);
-    $tagregexp = join( '|', array_map('preg_quote', $tagnames) );
-    $pattern = "/\\[($tagregexp)/s";
-
-    if ( 1 !== preg_match( $pattern, $content ) ) {
-        // Avoids parsing HTML when there are no shortcodes or embeds anyway.
+    // Find all registered tag names in $content.
+    preg_match_all( '@\[([^<>&/\[\]\x00-\x20]++)@', $content, $matches );
+    $tagnames = array_intersect( array_keys( $shortcode_tags ), $matches[1] );
+
+    if ( empty( $tagnames ) ) {
         return $content;
     }
 
-    $content = do_shortcodes_in_html_tags( $content, $ignore_html );
-
-    $pattern = get_shortcode_regex();
+    $content = do_shortcodes_in_html_tags( $content, $ignore_html, $tagnames );
+
+    $pattern = get_shortcode_regex( $tagnames );
     $content = preg_replace_callback( "/$pattern/s", 'do_shortcode_tag', $content );
 
@@ -248 +247 @@
  * @global array $shortcode_tags
  *
+ * @param array $tagnames List of shortcodes to find. Optional. Defaults to all registered shortcodes.
  * @return string The shortcode search regular expression
  */
-function get_shortcode_regex() {
-    global $shortcode_tags;
-    $tagnames = array_keys($shortcode_tags);
+function get_shortcode_regex( $tagnames = null ) {
+    global $shortcode_tags;
+
+    if ( empty( $tagnames ) ) {
+        $tagnames = array_keys( $shortcode_tags );
+    }
     $tagregexp = join( '|', array_map('preg_quote', $tagnames) );
 
@@ -338 +341 @@
  * @param string $content Content to search for shortcodes
  * @param bool $ignore_html When true, all square braces inside elements will be encoded.
+ * @param array $tagnames List of shortcodes to find.
  * @return string Content with shortcodes filtered out.
  */
-function do_shortcodes_in_html_tags( $content, $ignore_html ) {
+function do_shortcodes_in_html_tags( $content, $ignore_html, $tagnames ) {
     // Normalize entities in unfiltered HTML before adding placeholders.
     $trans = array( '&#91;' => '&#091;', '&#93;' => '&#093;' );
@@ -346 +350 @@
     $trans = array( '[' => '&#91;', ']' => '&#93;' );
 
-    $pattern = get_shortcode_regex();
+    $pattern = get_shortcode_regex( $tagnames );
     $textarr = wp_html_split( $content );
 
@@ -558 +562 @@
         return $content;
 
-    $content = do_shortcodes_in_html_tags( $content, true );
-
-    $pattern = get_shortcode_regex();
+    // Find all registered tag names in $content.
+    preg_match_all( '@\[([^<>&/\[\]\x00-\x20]++)@', $content, $matches );
+    $tagnames = array_intersect( array_keys( $shortcode_tags ), $matches[1] );
+
+    if ( empty( $tagnames ) ) {
+        return $content;
+    }
+
+    $content = do_shortcodes_in_html_tags( $content, true, $tagnames );
+
+    $pattern = get_shortcode_regex( $tagnames );
     $content = preg_replace_callback( "/$pattern/s", 'strip_shortcode_tag', $content );