Document Object Model

If you need simple interface to DOM check out phpQuery - jQuery port to PHP:http://code.google.com/p/phpquery/It uses CSS selectors to fetch nodes.Here's example how it works:<?php// just one file to includerequire('phpQuery/phpQuery.php');$html = '<div>    mydiv    <ul>        <li>1</li>        <li>2</li>        <li>3</li>    </ul></div>';// intialize new DOM from markupphpQuery::newDocument($markup)    ->find('ul > li')        ->addClass('my-new-class')        ->filter(':last')            ->addClass('last-li');// query all unordered lists in last used DOMpq('ul')->insertAfter('div');// iterate all LIs from last used DOMforeach(pq('li') as $li) {    // iteration returns plain DOM nodes, not phpQuery objects    pq($li)->addClass('my-second-new-class');}// same as pq('anything')->htmlOuter()// but on document root (returns doctype etc)print phpQuery::getDocument();?>It uses DOM extension and XPath so it works only in PHP5.

If  you want to print the DOM XML file content, you can use the next code:$doc = new DOMDocument();$doc->load($xmlFileName);echo "<br>" . $doc->documentURI;$x = $doc->documentElement;getNodeContent($x->childNodes, 0);function getNodeContent($nodes, $level){    foreach ($nodes AS $item) {    //  print "<br><br>TIPO: " . $item->nodeType ;      printValues($item, $level);      if ($item->nodeType == 1) { //DOMElement          foreach ($item->attributes AS $itemAtt) {              printValues($itemAtt, $level+3);          }          if($item->childNodes || $item->childNodes->lenth > 0) {              getNodeContent($item->childNodes, $level+5);          }      }    }}function printValues($item, $level){    if ($item->nodeType == 1) { //DOMElement        printLevel($level);        print $item->nodeName . " = " . $item->nodeValue;    }    if ($item->nodeType == 2) { //DOMAttr        printLevel($level);        print $item->name . " = " . $item->value ;    }    if ($item->nodeType == 3) { //DOMText      if ($item->isWhitespaceInElementContent() == false){        printLevel($level);        print $item->wholeText ;      }    }}function printLevel($level){    print "<br>";    if ($level == 0) {        print "<br>";    }    for($i=0; $i < $level; $i++) {        print "-";    }}

When I tried to parse my XHTML Strict files with DOM extension, it couldn't understand xhtml entities (like &copy;). I found post about it here (14-Jul-2005 09:05) which adviced to add resolveExternals = true, but it was very slow. There was some small note about xml catalogs but without any glue. Here it is:XML catalogs is something like cache. Download all needed dtd's to /etc/xml, edit file /etc/xml/catalog and add this line: <public publicId="-//W3C//DTD XHTML 1.0 Strict//EN" uri="file:///etc/xml/xhtml1-strict.dtd" />Thats all. Thanks to http://www.whump.com/moreLikeThis/link/03815

The project I'm currently working on uses XPaths to dynamically navigate through chunks of an XML file.  I couldn't find any PHP code on the net that would build the XPath to a node for me, so I wrote my own function.  Turns out it wasn't as hard as I thought it might be (yay recursion), though it does entail using some PHP shenanigans...  Hopefully it'll save someone else the trouble of reinventing this wheel. <?php    function getNodeXPath( $node ) {        // REMEMBER THAT XPATHS USE BASE-1 INSTEAD OF BASE-0!!!                // Get the index for the current node by looping through the siblings.        $parentNode = $node->parentNode;        if( $parentNode != null ) {            $nodeIndex = 0;            do {                $testNode = $parentNode->childNodes->item( $nodeIndex );                $nodeName = $testNode->nodeName;                $nodeIndex++;                                // PHP trickery!  Here we create a counter based on the node                 //  name of the test node to use in the XPath.                if( !isset( $$nodeName ) ) $$nodeName = 1;                else $$nodeName++;                                // Failsafe return value.                if( $nodeIndex > $parentNode->childNodes->length ) return( "/" );            } while( !$node->isSameNode( $testNode ) );                        // Recursively get the XPath for the parent.            return( getNodeXPath( $parentNode ) . "/{$node->nodeName}[{$$nodeName}]" );        } else {            // Hit the root node!  Note that the slash is added when             //  building the XPath, so we return just an empty string.            return( "" );        }    }?>

This module is not included by default either in the CentOS 4 "centosplus" repository. For those using PHP5 on CentOS 4, a simple "yum --enablerepo=centosplus install php-xml" will do the trick (this will install both the XML and DOM modules).

I hate DOM model !so I wrote dom2array simple function (simple for use):function dom2array($node) {  $res = array();  print $node->nodeType.'<br/>';  if($node->nodeType == XML_TEXT_NODE){      $res = $node->nodeValue;  }  else{      if($node->hasAttributes()){          $attributes = $node->attributes;          if(!is_null($attributes)){              $res['@attributes'] = array();              foreach ($attributes as $index=>$attr) {                  $res['@attributes'][$attr->name] = $attr->value;              }          }      }      if($node->hasChildNodes()){          $children = $node->childNodes;          for($i=0;$i<$children->length;$i++){              $child = $children->item($i);              $res[$child->nodeName] = dom2array($child);          }      }  }  return $res;}

When dealing with validation or loading, the output errors can be quite annoying.PHP 5.1 introduces libxml_get_errors().http://php.net/libxml_get_errors

As of PHP 5.1, libxml options may be set using constants rather than the use of proprietary DomDocument properties.DomDocument->resolveExternals is equivilant to settingLIBXML_DTDLOADLIBXML_DTDATTRDomDocument->validateOnParse is equivilant to settingLIBXML_DTDLOADLIBXML_DTDVALIDPHP 5.1 users are encouraged to use the new constants. Example:DomDocument->load($file, LIBXML_DTDLOAD|LIBXML_DTDATTR);DomDocument->load($file, LIBXML_DTDLOAD|LIBXML_DTDVALID);

I had problems with the dom2array_full function by "nospam at ya dot ru". Here's my function, which works correctly for my project, and might work for yours:

<?php
function dom_to_array($root)
{
    $result = array();

    if ($root->hasAttributes())
    {
        $attrs = $root->attributes;

        foreach ($attrs as $i => $attr)
            $result[$attr->name] = $attr->value;
    }

    $children = $root->childNodes;

    if ($children->length == 1)
    {
        $child = $children->item(0);

        if ($child->nodeType == XML_TEXT_NODE)
        {
            $result['_value'] = $child->nodeValue;

            if (count($result) == 1)
                return $result['_value'];
            else
                return $result;
        }
    }

    $group = array();

    for($i = 0; $i < $children->length; $i++)
    {
        $child = $children->item($i);

        if (!isset($result[$child->nodeName]))
            $result[$child->nodeName] = dom_to_array($child);
        else
        {
            if (!isset($group[$child->nodeName]))
            {
                $tmp = $result[$child->nodeName];
                $result[$child->nodeName] = array($tmp);
                $group[$child->nodeName] = 1;
            }

            $result[$child->nodeName][] = dom_to_array($child);
        }
    }

    return $result;
}
?>

I had the hardest time updating a complex XML document. Here's a quick example on how to do it.<?php // Load the XML from a file.$xml = "a2062.xml"; // This is an XFDL form previously unencoded and ungzipped.$dom = new DOMDocument();$dom->preserveWhiteSpace = false;$dom->Load($xml);// Create an XPath query.// Note: you must define the namespace if the XML document has defined namespaces.$xpath = new DOMXPath($dom);$xpath->registerNamespace('xfdl', "http://www.PureEdge.com/XFDL/6.5");// Locate the value for the first Item Description field.$query = "//xfdl:page/xfdl:field[@sid='ITEMDESA']/xfdl:value";$nodeList = $xpath->query($query);$nodeList->item(0)->nodeValue = "This is the text in the value node of the first Item Description field inside the DA 2062 PureEdge form.";$dom->save($xml);?>I hope this helps someone.

In response to "simlee at indiana dot edu",- First of all thanks for sharing your funciton.- It didn't work for me so i rewrite it from scratch using different method.Here is the new version, hope it helps someone : <?php/*** result sample : /html[1]/body[1]/span[1]/fieldset[1]/div[1]* @return string*/function getNodeXPath( $node ) {        $result='';    while ($parentNode = $node->parentNode) {        $nodeIndex=-1;        $nodeTagIndex=0;        do {            $nodeIndex++;            $testNode = $parentNode->childNodes->item( $nodeIndex );                            if ($testNode->nodeName==$node->nodeName and $testNode->parentNode->isSameNode($node->parentNode) and $testNode->childNodes->length>0) {                //echo "{$testNode->parentNode->nodeName}-{$testNode->nodeName}-{}<br/>";                $nodeTagIndex++;            }                        } while (!$node->isSameNode($testNode));        $result="/{$node->nodeName}[{$nodeTagIndex}]".$result;        $node=$parentNode;    };    return $result;}?>By Sina.Salek.ws

Being an experienced ASP developer I was wondering how to replace textual content of a node (with msxml this is simply acheived by setting the 'text' property of a node). Out of frustration I started to play around with SimpleXml but I could not get it to work in combination with xPath.I took me a lot of time to find out so I hope this helps others:function replaceNodeText($objXml, $objNode, $strNewContent){    /*    This function replaces a node's string content with strNewContent    */    $objNodeListNested = &$objNode->childNodes;    foreach ( $objNodeListNested as $objNodeNested ){         if ($objNodeNested->nodeType == XML_TEXT_NODE)$objNode->removeChild ($objNodeNested);    }        $objNode->appendChild($objXml->createTextNode($strNewContent));}$objXml= new DOMDocument();$objXml->loadXML('<root><node id="1">bla</note></root>');$objXpath = new domxpath($objXml);$strXpath="/root/node[@id='1']";$objNodeList = $objXpath ->query($strXpath);foreach ($objNodeList as $objNode){//pass the node by referencereplaceNodeText($objXml, &$objNode, $strImportedValue);}

<?php    // this note is about how to get a DOMNode's outerHTML and innerHTML.    $dom = new DOMDocument('1.0','UTF-8');    $dom->loadHTML('<html><body><div><p>p1</p><p>p2</p></div></body></html>');        $node = $dom->getElementsByTagName('div')->item(0);        $outerHTML = $node->ownerDocument->saveHTML($node);        $innerHTML = '';    foreach ($node->childNodes as $childNode){        $innerHTML .= $childNode->ownerDocument->saveHTML($childNode);    }    echo '<h2>outerHTML: </h2>';    echo htmlspecialchars($outerHTML);    echo '<h2>innerHTML: </h2>';        echo htmlspecialchars($innerHTML);    ?>

This particular W3C page provides invaluable documentation for the DOM classes implemented in php5 (via libxml2). It fills in plenty of php.net's gaps:http://www.w3.org/TR/DOM-Level-2-Core/core.htmlSome key examples:* concise summary of the class heirachy (1.1.1)* clarification that DOM level 2 doesn't allow for population of internal DTDs* explanation of DOMNode->normalize()* explanation of the DOMImplementation classThe interfaces are described in OMG's Interface Definition Language

If you are using not object-oriented functions and it takes too much time to change them all (or you'll be replacing them later) then as a temporary decision can be used this modules:For DOM XML:http://alexandre.alapetite.net/doc-alex/domxml-php4-php5/For XSLT:http://alexandre.alapetite.net/doc-alex/xslt-php4-php5/

After searching how to extend the DOMDocument and DOMElement I found a way in the bug: http://bugs.php.net/bug.php?id=35104. The following code shows how:<?phpclass extDOMDocument extends DOMDocument { public function createElement($name, $value=null) {  $orphan = new extDOMElement($name, $value); // new  sub-class object  $docFragment = $this->createDocumentFragment(); // lightweight container maintains "ownerDocument"  $docFragment->appendChild($orphan); // attach  $ret = $docFragment->removeChild($orphan); // remove  return $ret; // ownerDocument set; won't be destroyed on  method exit } // .. more class definition}class extDOMElement extends DOMElement { function __construct($name, $value='', $namespaceURI=null) {  parent::__construct($name, $value, $namespaceURI); }  //  ... more class definition here}$doc = new extDOMDocument('test');$el = $doc->createElement('tagname');$el->setAttribute("attr", "val");$doc->appendChild($el); // append discards the DOMDocumentFragment and just adds its child nodes, but ownerDocument is maintained.echo get_class($el)."<br/>";echo get_class($doc->documentElement)."<br/>";echo "<xmp>".$doc->saveXML()."</xmp>";?>

innerHTML in PHP DOM 

<?php
function DOMinnerHTML($element)
{
    $innerHTML = "";
    $children = $element->childNodes;
    foreach ($children as $child) 
    {
        $tmp_dom = new DOMDocument();
        $tmp_dom->appendChild($tmp_dom->importNode($child, true));
        $innerHTML.=trim($tmp_dom->saveHTML());
    }
    return $innerHTML;
}
?>

Example:

<?php
$dom= new DOMDocument(); 
$dom->load($html_string);
$dom->preserveWhiteSpace = false;

$domTable = $dom->getElementsByTagName("table");

foreach ($domTable as $tables)
{
    echo DOMinnerHTML($tables);
}
?>

This extension allows editing an HTML snippet, for example removing an element:‌‌$dom = new DOMDocument();‌‌$dom->loadHTML('keep<b id="simple">REMOVE</b>');$‌‌remove = $dom->getElementById('simple');‌‌$dom->removeChild($‌‌remove);echo ‌‌$dom->saveHTML();...would output "keep".But it does not preserve formatting. If your code defines attributes using double quotes, saveHTML() may for instance return single quotes instead.

If you want to  use DOMDocument in your PHPUnit Tests drive on Symfony Controller (testing form)! Use like :namespace Tests\YourBundle\Controller;use Symfony\Bundle\FrameworkBundle\Test\WebTestCase;use YourBundle\Controller\TextController;class DefaultControllerTest extends WebTestCase{    public function testIndex()    {        $client = static::createClient(array(), array());        $crawler = $client->request('GET', '/text/add');        $this->assertTrue($crawler->filter("form")->count() > 0, "Text form exist !");        $form = $crawler->filter("form")->form();        $domDocument = new \DOMDocument;        $domInput = $domDocument->createElement('input');        $dom = $domDocument->appendChild($domInput);        $dom->setAttribute('slug', 'bloc');               $formInput = new \Symfony\Component\DomCrawler\Field\InputFormField($domInput);        $form->set($formInput);         $crawler = $client->submit($form);                    if ($client->getResponse()->isRedirect())        {            $crawler = $client->followRedirect(false);        }       // $this->assertTrue($client->getResponse()->isSuccessful());        //$this->assertEquals(200, $client->getResponse()->getStatusCode(),        //             "Unexpected HTTP status code for GET /backoffice/login");            }}

Here is a fast innerHTML function that returns the result without iterating over child nodes.<?phpfunction innerHTML($el) {  $doc = new DOMDocument();  $doc->appendChild($doc->importNode($el, TRUE));  $html = trim($doc->saveHTML());  $tag = $el->nodeName;  return preg_replace('@^<' . $tag . '[^>]*>|</' . $tag . '>$@', '', $html);}?>Example<?php$doc = new DOMDocument();// A corrupt HTML string$doc->loadHTML('<HTML><A HREF="ss">asd</A>');$body = $doc->getElementsByTagName('body')->item(0);print htmlspecialchars(innerHTML($body));// Prints <a href="ss">asd</a>?>

A function among several others to parse a google results page, I wrote this some time ago - google has probably changed their site since then, but I thought this might be helpful to someone.I'm moving servers, but I will probably throw this up on my blog when I get it back up.<?phpfunction googleResult($listItem) {    // given a LIST ITEM element, this will validate, and return an array for that LI entry as an inline result from google.    /*     * <li class='g w0'>     *   <h3 class='r'>     *     <a href='the URL' class='l'>     *       Description <em>description</em>     *     </a>     *   </h3>     * </li>     *       UPDATE:       This function will now look for any subcontainer that has an href, it doesn't have to be an H3       this will make it work with a few more formatted search results.     */    $listItem = $listItem->childNodes;        // Yes I don't use instanceof - I guess you'll have to deal.    foreach($listItem as $element) {        if(is_object($element) && get_class($element) == 'DOMElement' && $element->hasChildNodes()) {            $hrefContainer = $element->childNodes;            foreach($hrefContainer as $element2) {                if(is_object($element2) && get_class($element2) == 'DOMElement' && $element2->nodeName == 'a' && $element2->hasAttribute('href')) {                    $anchor = $element2;                    unset($h3);                    unset($element2);                    break;                } else {                  //print __LINE__ ." :: Breaking out of loop (normal result) element is not an annchor Element='".$element2->nodeName."'\n";                }            }            unset($element);            unset($listItem);            break;        }    }    if(empty($anchor) || !is_object($anchor) || get_class($anchor) != 'DOMElement') {        //print __LINE__ ." :: Returning false, did not locate anchor through iteration...";        return false;    }    $href = $anchor->getAttribute('href');    if(empty($href)) {        //print __LINE__ ." :: Found anchor object, could not read href attribute / href is empty? href='$href'\n";        return false;    }    $description = $anchor->childNodes;    $urlDescription = '';    foreach($description as $words) {        $name =  trim($words->nodeName);        if($name == 'em' || $name == '#text' || $name == 'b') {            if(!empty($words->nodeValue)) {                $text = trim($words->nodeValue);                $urlDescription = $urlDescription . $text . ' ';            }        }    }    $urlDescription = htmlspecialchars_decode($urlDescription, ENT_QUOTES);    $urlDescription = trim($urlDescription);    return array('description' => $urlDescription, 'href' => $href);}

You can get the "innerHTML" by nodeValue so<?php    $doc = new DOMDocument( );    $ele = $doc->createElement( 'p', 'Sensei Ninja' );    print $ele->nodeValue;?>You can even set it if you want<?php    $doc = new DOMDocument( );    $ele = $doc->createElement( 'p' );    $ele->nodeValue = 'Sensei Ninja';    $doc->appendChild( $ele );    print $doc->saveHTML( );?>

<?PHPfunction dom2array_full($node){     $result = array();     if($node->nodeType == XML_TEXT_NODE) {         $result = $node->nodeValue;     }     else {         if($node->hasAttributes()) {             $attributes = $node->attributes;             if(!is_null($attributes))                  foreach ($attributes as $index=>$attr)                      $result[$attr->name] = $attr->value;         }         if($node->hasChildNodes()){             $children = $node->childNodes;             for($i=0;$i<$children->length;$i++) {                 $child = $children->item($i);                 if($child->nodeName != '#text')                 if(!isset($result[$child->nodeName]))                     $result[$child->nodeName] = dom2array($child);                 else {                     $aux = $result[$child->nodeName];                     $result[$child->nodeName] = array( $aux );                     $result[$child->nodeName][] = dom2array($child);                 }             }         }     }     return $result; } ?>

This is a couple of classes to deal with yahoo yui menu./*  $menubar = new MenuBar();  $file = new Menu("File");  $file->setAttribute("href", "http://file.com");  $quit = new Menu("Quit");  $quit->setAttribute("href", "http://quit.com");  $file->appendChild($quit);  $menubar->appendChild($file);  echo $menubar->grab();*///// Author: Daniel Queirolo.// LGPL///** ---------------------------------/**  Class MenuBar()/**  Creates a the menubar and appends/**  yuimenubaritems to it./** ---------------------------------*/class MenuBar extends DOMDocument{  public $menuID = "nav_menu";  // holds the css id that javascript yui menu code should have to recognize  private $UL;                  // This node holds every menu, This is THE node.  /** ---------------------------------  /**  Constructor  /**  Generates a menubar skeleton and the UL node  /** ---------------------------------*/  public function __construct() {    parent::__construct();    $rootdiv = parent::createElement("div");    $rootdiv->setAttribute("class", "yui-skin-sam");    parent::appendChild($rootdiv);    $yui_menubar = parent::createElement("div");    $yui_menubar->setAttribute("id", $this->menuID);    $yui_menubar->setAttribute("class", "yuimenubar");    $rootdiv->appendChild($yui_menubar);    $bd = parent::createElement("div");    $bd->setAttribute("class", "bd");    $yui_menubar->appendChild($bd);    $ul = parent::createElement("ul");    $ul->setAttribute("class", "first-of-type");    // ALL Menu() instances ocurr inside an <ul> tag.    $this->UL = $bd->appendChild($ul);  }  /** ---------------------------------  /**  appendChild()  /**  Appends a new yuimenubaritem to the menubar UL node.  /**  This function changes <li> and <a> classes to yuiMENUBARsomething  /** ---------------------------------*/  public function appendChild($child) {    $li = parent::importNode($child->LI, true);    $li->setAttribute("class", "yuimenubaritem");    $li->getElementsByTagName("a")->item(0)->setAttribute("class", "yuimenubaritemlabel");    $this->UL->appendChild($li);  }  public function grab() {    return parent::saveHTML();  }}/** ---------------------------------/**  Class Menu()/**  Creates a yuimenuitem li node/** ---------------------------------*/class Menu extends DOMDocument {  public $LI; // stores the <li> node (THE link) that will be exported to MenuBar() or used on appendChild()  /** ---------------------------------  /**  Constructor  /**  Generates a yuimenuitem li node  /**  No yuimenubar items are created here. MenuBar handles that.  /** ---------------------------------*/  public function __construct($link_name) {    parent::__construct();    $li = parent::createElement("li");    $li->setAttribute("class", "yuimenuitem");    // LI node stores THE link.    // if appendChild is used, the new (sub) Menu() would be LI node child.    $this->LI = parent::appendChild($li);    $a = parent::createElement("a", $link_name);    $a->setAttribute("class", "yuimenuitemlabel");    $li->appendChild($a);    $this->li = $li;    $this->a = $a;  }  /** ---------------------------------  /**  appendChild  /**  Appends a (sub) Menu() to current Menu() in LI  /** ---------------------------------*/  public function appendChild($child) {    $yuimenu = parent::createElement("div");    $yuimenu->setAttribute("class", "yuimenu");    $this->LI->appendChild($yuimenu);    $bd = parent::createElement("div");    $bd->setAttribute("class", "bd");    $yuimenu->appendChild($bd);    $ul = parent::createElement("ul");    $bd->appendChild($ul);    // child->NODE holds THE link from the new child (from child's __construct())    $ul->appendChild(parent::importNode($child->LI, true));  }  public function setAttribute($name, $value, $node="a") {      if ($node == "a") {        $this->a->setAttribute($name, $value);      }      else {        $this->li->setAttribute($name, $value);      }  }}

appended to brian dot reynolds at risaris dot com20-Feb-2007 10:09when you got variable nodes at start you array fails and looses nodes  beneath.solution that counts occurance though eats up performance:function xmlToArray($n){    $xml_array = array();    $occurance = array();     foreach($n->childNodes as $nc)     {         $occurance[$nc->nodeName]++;     }         foreach($n->childNodes as $nc){        if( $nc->hasChildNodes() )        {            if($occurance[$nc->nodeName] > 1)            {                $xml_array[$nc->nodeName][] = xmlToArray($nc);            }            else            {                $xml_array[$nc->nodeName] = xmlToArray($nc);            }        }        else        {            return utf8_decode($nc->nodeValue);        }    }    return $xml_array;}

I wrote a couple of functions to: - create a DOMDocument from a file - parse the namespaces in it - create a XPath object with all the namespaces registered - load the schemalocations - validate the file on the main schema (the one without prefix)It is useful for me, see if it is also for someone else!!Giuliofunction decodeNode($node){    $out = $node->ownerDocument->saveXML($node);    $re = "{^<((?:\\w*:)?\\w*)". //the tag name    "[\\s\n\r]*((?:[\\s\n\r]*".     "(?:\\w*:)?\\w+[\\s\n\r]*=[\\s\n\r]*". //possible attribute name    "(?:\"[^\"]*\"|\'[^\']*\'))*)". //attribute value    "[\\s\n\r]*>[\r\n]*".    "((?:.*[\r\n]*)*)". //content    "[\r\n]*</\\1>$}"; //closing tag    preg_match($re, $out, $mat);    return $mat;}function innerXml($node){    $mat = decodeNode($node);    return $mat[3];}function getnodeAttributes($node){    $mat = decodeNode($node);    $txt = $mat[2];    $re = "{((?:\\w*:)?\\w+)[\\s\n\r]*=[\\s\n\r]*(\"[^\"]*\"|\'[^\']*\')}";    preg_match_all($re, $txt, $mat);    $att = array();    for ($i=0; $i<count($mat[0]); $i++)    {        $value = $mat[2][$i];        if ($value[0] == "\'" || $value[0] == "\"")        {            $len = strlen($value);            $value = substr($value, 1, strlen($value)-2);        }        $att[ $mat[1][$i] ] = $value;    }    return $att;}function loadXml($file){    $doc = new DOMDocument();    $doc->load($file);    //cerca l'attributo xmlns    $xsi = false;    $doc->namespaces = array();    $doc->xpath = new DOMXPath($doc);        $attr = getnodeAttributes($doc->documentElement);    foreach ($attr as $name => $value)    {        if (substr($name,0,5) == "xmlns")        {            $uri = $value;            $pre = $doc->documentElement->lookupPrefix($uri);            if ($uri == "http://www.w3.org/2001/XMLSchema-instance")                $xsi = $pre;            $doc->namespaces[$pre] = $uri;            if ($pre == "")                $pre = "noname";            $doc->xpath->registerNamespace($pre, $uri);        }    }        if ($xsi)    {        $doc->schemaLocations = array();        $lst = $doc->xpath->query("//@$xsi:schemaLocation");        foreach($lst as $el)        {            $re = "{[\\s\n\r]*([^\\s\n\r]+)[\\s\n\r]*([^\\s\n\r]+)}";            preg_match_all($re, $el->nodeValue, $mat);            for ($i=0; $i<count($mat[0]); $i++)            {                $value = $mat[2][$i];                $doc->schemaLocations[ $mat[1][$i] ] = $value;            }        }        $olddir = getcwd();        chdir(dirname($file));        $schema = $doc->schemaLocations[$doc->namespaces[""]];        if (substr($schema,0,7) == "file://")        {            $schema = substr($value,7);        }        if (!$doc->schemaValidate($schema))            dbg()->err("Invalid file");        chdir($olddir);    }        return $doc;}

$xmlDoc=<<<XML<?xml version="1.0"?><methodCall>   <methodName>examples.getStateName</methodName>   <params>      <param>         <value><i4>41</i4></value>         </param>      </params>   </methodCall>XML;$xml= new DOMDocument();$xml->preserveWhiteSpace=false;$xml->loadXML($xmlDoc);print_r(xml2array($xml));function xml2array($n){    $return=array();    foreach($n->childNodes as $nc)    ($nc->hasChildNodes())    ?($n->firstChild->nodeName== $n->lastChild->nodeName&&$n->childNodes->length>1)    ?$return[$nc->nodeName][]=xml2array($item)    :$return[$nc->nodeName]=xml2array($nc)    :$return=$nc->nodeValue;    return $return;}

Note that these DOM functions expect (and presumably return) all their data in UTF-8 character encoding, regardless of what PHP's current encoding is. This means that text nodes, attribute values etc, should be in utf8.This applies even if you're generating an XML document which is not ultimately in utf8.Mark

[Editor's Note: If you're using entities, then you have no choice. XML Catalogs can speed DTD resolution.]

Never use 

$dom->resolveExternals=true;

when parsing XHTML document that has the DOCTYPE declaration with DTD URL specified in it.

Otherwise parsing the XHTML with DOCTYPE like this one:

<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">

will result in PHP/DOM downloading the DTD file from W3C site when parsing your document. It will add extra delay to your script - I experienced that $dom->load()'s total time was from 1 to 16 seconds.

elixon

I wrote a framework to implement the StyleSheet interfaces as specified on the W3C website.  The code is written in PHP, and is NOT a complete implementation.  Use it how ya like.  I was planning on adding the CSSStyleSheet interfaces as well.  Feel free to ask.<?    class StyleSheetList {        public length;        private self;                function __construct ( ) {            $this->self = array();        }                function __get($property, $&ret) {            if($property == 'length')                $ret = count($this->self);            return true;        }                function __set($property, $val) {            if($property == 'length')                return true;        }                function item( $index ) {            return $this->self[$index];        }    }        class MediaList extends StyleSheetList {                function appendMedium ( $newMedium ) {            array_push($this->self, $newMedium);        }                function deleteMedium ( $oldMedium ) {            foreach($this->self as $item) {                if( $item == $oldMedium ) {                    $item = $this->self[ $this->length-1 ];                    array_pop($this->self);                    break;                }            }        }    }        class DocumentStyle {        public styleSheets;                function __construct ( ) {            $this->styleSheets = new StyleSheetList();        }                function __set($property, $val) {            if($property == 'styleSheets')                return true;        }    }        class LinkStyle {        public sheet;                function __construct () {            $this->sheet = new StyleSheet();        }                function __set($property, $val) {            if($property == 'sheet')                return true;        }    }        class StyleSheet {        public type;        public disabled;        public ownerNode;        public parentStyleSheet;        public href;        public title;        public media;                function __construct( $type, $disabled, $ownerNode, $parentStyleSheet, $href, $title){            $this->type = $type;            $this->disabled = $disabled;            $this->media = new MediaList();            $this->ownerNode = $ownerNode;            $this->parentStyleSheet = $parentStyleSheet;            $this->href = $href;            $this->title = $title;        }    }?>Only contactable via http://murpsoft.com/contact.html

The following can take a XML_TEXT_NODE node and return the contents in an array. Yanick's contribution rocks - butit overwrote with duplicates only keeping the last linein the returned array. All the other functions i tested from various sources failed to handle text nodes correctly. Hope this helps someone. It is adapted from code on this site.function myTextNode($n, &$a){ static $depth = 0; static $sz = ''; if ($cn = $n->firstChild)  {  while ($cn)  {   if ($cn->nodeType == XML_TEXT_NODE)    {    $sz .= $cn->nodeValue;   }    elseif ($cn->nodeType == XML_ELEMENT_NODE)    {    $b = 1;    if ($cn->hasChildNodes())     {     $depth++;     if ($this->myHeadings($cn, $a))      {      if ($sz){       array_push($a, $sz);       $sz = '';      }     }     $depth--;    }   }   $cn = $cn->nextSibling;  }  return $b; }} so you could use:   $nodes = $dom->getElementsByTagName("td"); if($nodes){ foreach ($nodes as $node){  $a = Array();  myTextNode($node, $a); }}

The Yanik's dom2array() function (added on 14-Mar-2007 08:40) does not handle multiple nodes with the same name, i.e.:<foo>  <name>aa</name>  <name>bb</name></foo>It will overwrite former and your array will contain just the last one ("bb")

In response to lutfi at smartconsultant dot us :(see my post onhttp://fr2.php.net/manual/en/function.dom-domdocument-getelementsbytagname.php)Use this class I wrote:class XPathableNode extends DOMNode{    protected $Node;    protected $DOMDocument_from_node;    protected $DOMXpath_for_node;    public function __construct(/* DOMNode */ $node)    {        $this->Node=$node;        $this->DOMDocument_from_node=new                                                           DomDocument();        $domNode=$this->DOMDocument_from_node                                  ->importNode($this->Node, true);        $this->DOMDocument_from_node                                 ->appendChild($domNode);        $this->DomXpath_for_node =                     new Domxpath($this->                                         DOMDocument_from_node);    }    public function convertHTML()    {    return $this->DOMDocument_from_node                                                             ->saveHTML();    }    public /*DomNodeList*/ function applyXpath($xpath)    {    return $this->DomXpath_for_node                                                       ->query($xpath);    }}(sorry for the display... What a terrible hinderance on thepart of php.net !)Then :Make a new XPathableNode out of your parent node.You may then retrieve a DOMNodeList from it by applying axpath (thus being able to specify the depth  and name of elements you want).Has got me around some (of the many) DOM awkwardnesses a few times.;o)

I found the xml2array function below very useful, but there seems to be a bug in it. The $item variable was never getting set. I've expanded this out to be a bit more readable, and the corrected code is : function xmlToArray($n){    $return=array();    foreach($n->childNodes as $nc){        if( $nc->hasChildNodes() ){            if( $n->firstChild->nodeName== $n->lastChild->nodeName&&$n->childNodes->length>1){                $item = $n->firstChild;                $return[$nc->nodeName][]=$this->xmlToArray($item);            }            else{                 $return[$nc->nodeName]=$this->xmlToArray($nc);            }       }       else{           $return=$nc->nodeValue;       }     }    return $return;}

In response to..."If you create your own custom element extending DOMElement and append him in place of the document element, you cannot access to any new members newly defined in your custom class via DOMDocument::$documentElement."... it is not a bug, it is a feature. The DOMDocument::$documentElement property name may be misleading but according to the DOM Level 2 Core specification it is a convenience attribute meant to access the root element of your DOMDocument. See here: http://www.w3.org/TR/DOM-Level-2-Core/core.html#i-Document

I use DOM to generate dynamically XHTML document.When trying to extend the DOMDocument and DOMElement classes, I found a very annoying bug concerning DOMDocument::$documentElement.If you create your own custom element extending DOMElement and append him in place of the document element, you cannot access to any new members newly defined in your custom class via DOMDocument::$documentElement.In my situation, I cannot use DOMDocument::registerNodeClass() because the document element is not necessarily the base class for all the elements in my document.*******problem*******See bellow for the repro step:<?php    class MyElement extends DOMElement{        public $myProp="myProp";        public function myMethod(){            return 'myMethod()';        }    }        $myDocument=new DOMDocument();    $myDocument->appendChild(new MyElement('myElement','myElement'));    echo ('$myElement->myProp :'.$myDocument->documentElement->myProp.'<br />');    echo ('$myElement->myMethod :'.$myDocument->documentElement->myMethod().'<br />');?>will output:Notice: Undefined property: DOMElement::$myProp in C:\Program Files\EasyPHP 2.0b1\www\testDOMBug\test2.php on line 11$myElement->myProp :Fatal error: Call to undefined method DOMElement::myMethod() in C:\Program Files\EasyPHP 2.0b1\www\testDOMBug\test2.php on line 12*******solution*******After searching around, I found a pretty odd way to fix this problem. It seems that you have to stock a reference to your appended document element in an user-defined (and persistent) variable (in other words, not only in DOMDocument::$documentElement). See below:<?php        class MyElement extends DOMElement{        public $myProp="myProp";        public function myMethod(){            return 'myMethod()';        }    }        $myDocument=new DOMDocument();    $mydocumentElement=$myDocument->appendChild(new MyElement('myElement','myElement')); //here is the hack    echo ('$myElement->myProp :'.$myDocument->documentElement->myProp.'<br />');    echo ('$myElement->myMethod :'.$myDocument->documentElement->myMethod().'<br />');?>will output:$myElement->myProp :myProp$myElement->myMethod :myMethod()Hope it will help.

I needed a function that quickly converts an xml with key/value pairs to an array.for example:<?xml version="1.0" encoding="UTF-8"?><test>    <key id="Array key #1" value="Value #1"/>    <key id="Array key #2" value="Value #2"/>    <key id="Dupe" value="Duplicate keys"/>    <key id="Dupe" value="create numeric arrays"/>    <key id="And another key">        <![CDATA[            Multiline            data            works            just            as            well.        ]]>    </key>    <nested>        <key id="Nested key" value="Nested data works as well, but it still results in a 1 dimensional array."/>    </nested></test>Results in the following array:array(5) {  ["Array key #1"]=>  string(8) "Value #1"  ["Array key #2"]=>  string(8) "Value #2"  ["Dupe"]=>  array(2) {    [0]=>    string(14) "Duplicate keys"    [1]=>    string(21) "create numeric arrays"  }  ["And another key"]=>  string(49) "Multiline            data            works            just            as            well."  ["Nested key"]=>  string(73) "Nested data works as well, but it still results in a 1 dimensional array."}Here's the code:<?php    function xml2array($xml) {        $domDocument = new DOMDocument;        $domDocument->loadXML($xml);        $domXPath = new DOMXPath($domDocument);        $array = array();        foreach ($domXPath->query('//key') as $keyDOM) {            $id = $keyDOM->getAttribute('id');            $value = $keyDOM->hasAttribute('value') ? $keyDOM->getAttribute('value') : trim($keyDOM->textContent);            if (array_key_exists($id, $array)) {                if (is_array($array[$id])) {                    $array[$id][] = $value;                } else {                    $array[$id] = array($array[$id]);                    $array[$id][] = $value;                }            } else {                $array[$id] = $value;            }        }        return $array;    }?>

Array to DOMHere is a recursive function to turn a multidimensional array into an XML document.  It will handle multiple elements of the same tag, but only one per parent element. IE: Can't generate:         Can generate:<root>                  <root>  <sub1>data1</sub1>      <subs1>  <sub1>data2</sub1>         <value>data1</value>  <sub2>data1</sub2>         <value>data2</value>  <sub2>data2</sub2>      </subs1></root>                   <subs2>                            <value>data1</value>                            <value>data2</value>                          <subs2>                        </root>Also, the function performs no type of error checking on your array and will throw a DOMException if a key value you used in your array contains invalid characters for a proper DOM tag. This function is untested for "deep" multidimensional arrays.Complete code ready to run with example:<?PHP  function AtoX($array, $DOM=null, $root=null){        if($DOM  == null){$DOM  = new DOMDocument('1.0', 'iso-8859-1');}    if($root == null){$root = $DOM->appendChild($DOM->createElement('root'));}        $name = $array['#MULTIPLE_ELEMENT_NAME'];      foreach($array as $key => $value){            if(is_int($key) && $name != null){          if(is_array($value)){            $subroot = $root->appendChild($DOM->createElement($name));            AtoX($value, $DOM, $subroot);          }          else if(is_scalar($value)){            $root->appendChild($DOM->createElement($name, $value));          }        }        else if(is_string($key) && $key != '#MULTIPLE_ELEMENT_NAME'){          if(is_array($value)){            $subroot = $root->appendChild($DOM->createElement($key));            AtoX($value, $DOM, $subroot);          }          else if(is_scalar($value)){            $root->appendChild($DOM->createElement($key, $value));          }        }      }    return $DOM;    }    $array = array(                    '#MULTIPLE_ELEMENT_NAME' => 'GenericDatas',                    'Date'      => 'November 03, 2007',                    'Company'   => 'Facility One',                    'Field'     => 'Facility Management Software',                    'Employees' => array(                                      '#MULTIPLE_ELEMENT_NAME' => 'Employee',                                      'Cindy',                                      'Sean',                                      'Joe',                                      'Owen',                                      'Jim',                                      'Dale',                                      'Kelly',                                      'Ryan',                                      'Johnathan',                                      'Robin',                                      'William Marcus',                                      'NewCoops' => array(                                                          '#MULTIPLE_ELEMENT_NAME' => 'Coop',                                                          'John',                                                          'Tyler',                                                          'Ray',                                                          'Dawn'                                                         )                                         ),                    'Datas',                    'DATAS',                    'OtherDatas'                );    $DOM  = new DOMDocument('1.0', 'iso-8859-1');  $root = $DOM->appendChild($DOM->createElement('CompanyData'));   $DOM  = AtoX($array, $DOM, $root);  $DOM->save('C:\test.xml');?>