From 5690c9e0a2994a8064ba6fa264319799a15f4d55 Mon Sep 17 00:00:00 2001
From: "Edward Z. Yang" <edwardzyang@thewritingpot.com>
Date: Tue, 15 Aug 2006 21:19:45 +0000
Subject: [PATCH] Further optimization: 20% - 12%. Also fixed broken
 benchmarks.

git-svn-id: http://htmlpurifier.org/svnroot/htmlpurifier/trunk@266 48356398-32a2-884e-a903-53898d9a118a
---
 benchmarks/Lexer.php                  |  2 ++
 benchmarks/ProfileDirectLex.php       |  2 ++
 library/HTMLPurifier/Lexer/DOMLex.php | 27 +++++++++++++++------------
 3 files changed, 19 insertions(+), 12 deletions(-)

diff --git a/benchmarks/Lexer.php b/benchmarks/Lexer.php
index 5fe3a55b..da112fe4 100644
--- a/benchmarks/Lexer.php
+++ b/benchmarks/Lexer.php
@@ -3,6 +3,8 @@
 // emulates inserting a dir called HTMLPurifier into your class dir
 set_include_path(get_include_path() . PATH_SEPARATOR . '../library/');
 
+require_once 'HTMLPurifier/ConfigDef.php';
+require_once 'HTMLPurifier/Config.php';
 require_once 'HTMLPurifier/Lexer/DirectLex.php';
 require_once 'HTMLPurifier/Lexer/PEARSax3.php';
 
diff --git a/benchmarks/ProfileDirectLex.php b/benchmarks/ProfileDirectLex.php
index 11781dcb..175d2894 100644
--- a/benchmarks/ProfileDirectLex.php
+++ b/benchmarks/ProfileDirectLex.php
@@ -2,6 +2,8 @@
 
 set_include_path(get_include_path() . PATH_SEPARATOR . '../library/');
 
+require_once 'HTMLPurifier/ConfigDef.php';
+require_once 'HTMLPurifier/Config.php';
 require_once 'HTMLPurifier/Lexer/DirectLex.php';
 
 $input = file_get_contents('samples/Lexer/4.html');
diff --git a/library/HTMLPurifier/Lexer/DOMLex.php b/library/HTMLPurifier/Lexer/DOMLex.php
index 07b12d22..adf10636 100644
--- a/library/HTMLPurifier/Lexer/DOMLex.php
+++ b/library/HTMLPurifier/Lexer/DOMLex.php
@@ -81,19 +81,19 @@ class HTMLPurifier_Lexer_DOMLex extends HTMLPurifier_Lexer
         
         // intercept non element nodes
         
-        if ( !($node instanceof DOMElement) ) {
-            if ($node instanceof DOMComment) {
-                $tokens[] = $this->factory->createComment($node->data);
-            } elseif ($node instanceof DOMText ||
-                      $node instanceof DOMCharacterData) {
+        if ( isset($node->data) ) {
+            if ($node->nodeType === XML_TEXT_NODE ||
+                      $node->nodeType === XML_CDATA_SECTION_NODE) {
                 $tokens[] = $this->factory->createText($node->data);
+            } elseif ($node->nodeType === XML_COMMENT_NODE) {
+                $tokens[] = $this->factory->createComment($node->data);
             }
             // quite possibly, the object wasn't handled, that's fine
             return;
         }
         
         // We still have to make sure that the element actually IS empty
-        if (!$node->hasChildNodes()) {
+        if (!$node->childNodes->length) {
             if ($collect) {
                 $tokens[] = $this->factory->createEmpty(
                     $node->tagName,
@@ -125,13 +125,16 @@ class HTMLPurifier_Lexer_DOMLex extends HTMLPurifier_Lexer
      * @param $attribute_list DOMNamedNodeMap of DOMAttr objects.
      * @returns Associative array of attributes.
      */
-    protected function transformAttrToAssoc($attribute_list) {
-        $attribute_array = array();
-        // undocumented behavior
-        foreach ($attribute_list as $key => $attr) {
-            $attribute_array[$key] = $attr->value;
+    protected function transformAttrToAssoc($node_map) {
+        // NamedNodeMap is documented very well, so we're using undocumented
+        // features, namely, the fact that it implements Iterator and
+        // has a ->length attribute
+        if ($node_map->length === 0) return array();
+        $array = array();
+        foreach ($node_map as $attr) {
+            $array[$attr->name] = $attr->value;
         }
-        return $attribute_array;
+        return $array;
     }
     
 }