DOMNode::cloneNode

frame at dynamiccreated dot de (2012-12-02 02:19:48)

Remeber that DOMNode always needs a reference to a parent node or DOMDocument.
For example, if you try to clone a node - copy all children - and overwrite or delete the variable which holds the cloned node - all children will loose any reference and getting invalid.
This will cause a nice message like "Couldn't... node no longer exists" if you have luck. In most cases PHP only gives you the poor information "Couldn't fetch DOM[...]" which makes hard to find out whats going on, depending on the current operation.

cemkalyoncu at gmail dot com (2009-04-28 03:04:14)

If you need some function to clone a node without touching namespaces you can use the following.

<?php
private function cloneNode($node,$doc){
    $nd=$doc->createElement($node->nodeName);
            
    foreach($node->attributes as $value)
        $nd->setAttribute($value->nodeName,$value->value);
            
    if(!$node->childNodes) 
        return $nd;
                
    foreach($node->childNodes as $child) {
        if($child->nodeName=="#text")
            $nd->appendChild($doc->createTextNode($child->nodeValue));
        else
            $nd->appendChild(cloneNode($child,$doc));
    }
            
    return $nd;
}
?>

[montana] at [percepticon] dot [com] (2009-02-16 02:23:59)

<?php

//@oliver thanks for example source...

/*
 cloneNode(false) does not omit 
 Attributes of cloned node, 
 to achieve this an iteration is required. 
 this is probably less efficient 
 than merely creating a new 
 node from the desired nodeName 
 but in some cases could be useful.

use case: 

omit subnodes and attributes of 
secured portions of an xml document 
without altering expected general structure;
*/
//xml to use

$file="<?xml version='1.0'?>
<book type='paperback'>
    <title name='MAP'>Red Nails</title>
    <price>$12.99</price>
    <author>
        <name first='Robert' middle='E' last='Howard'/>
        <birthdate disco='false' nirvana='definitely'>
            9/21/1977 
            <month title='september' />
        </birthdate>
    </author>
    <author>
        <name first='Arthur' middle='Mc' last='Kayn'/>
    </author>
</book>";

$doc = new domDocument;

$doc->loadXML($file);

$xpath = new domXPath($doc);

$query = "//author/birthdate";
$xpathQuery = $xpath->query($query);

//would be a loop in production code...
$child = $xpathQuery->item(0); 

$parent = $child->parentNode;

$doppel = $child->cloneNode(false);

$limit = $doppel->attributes->length;

for ($a=0;$a<$limit;$a++) {
    $doppel->removeAttributeNode($doppel->attributes->item(0));
}
//swap for now empty node
$parent->replaceChild( $doppel, $child); 

print $doc->saveXML();

?>

oliver dot christen at camptocamp dot com (2004-11-19 08:48:29)

simple exemple of node cloning

<?xml version="1.0"?>

<book type="paperback">
    <title name='MAP'>Red Nails</title>
    <price>$12.99</price>
    <author>
        <name first="Robert" middle="E" last="Howard"/>
        <birthdate>9/21/1977</birthdate>
    </author>
    <author>
        <name first="Arthur" middle="Mc" last="Kayn"/>
    </author>
</book>

<?php

//filename xml file to use
$file = 'book.xml';

$doc = new domDocument;

if (file_exists($file)) {
    $doc->load($file);
} else {
    exit('Erreur !.');
}

$xpath = new domXPath($doc);

$query = "//author/*";
$xpathQuery = $xpath->query($query);

$size = $xpathQuery->length;
for ($i=0; $i<$size; $i++){
    $node = $xpathQuery->item($i);
    if ($node->nodeName == 'birthdate' && $node->hasChildNodes() && $node->firstChild->textContent != ''){
        $clonenode = $node->cloneNode(true);
        $refnode = $node;
    }
}
for ($i=0; $i<$size; $i++){
    $node = $xpathQuery->item($i);
    if (!$node->isSameNode($refnode)){
        $newnode = $node->appendChild($clonenode);
    }
}

print $doc->saveXML();

?>