How to delete all nodes from DOMDocument except custom ones?

Question

I have a DOMDocument in PHP and I'm trying to delete all nodes except of a container with a specific ID.

Lets say I have the following DOM Document:

<section>
  <div id="first-section">
    <ul>
      <li>Test</li>
      <li>Test</li>
    </ul>
  </div>
  <div id="second-section">
    <ul>
      <li>Test</li>
      <li>Test</li>
    </ul>
    <div id="sub-section">
      <h2>Hello World</h2>
    </div>
  </div>
  <div id="third-section">
    <ul>
      <li>Test</li>
      <li>Test</li>
    </ul>
  </div>
</section>

My PHP Code:

$domDocument = $this->domParser->loadHTML($markup);

$xpath = new \DOMXPath($domDocument);
$nlist = $xpath->query("//*[@id='sub-section']");

$domDocument->saveHTML();

With this code I query the correct container. But how could I remove all nodes except this node from my document, so that in the end I have the following nodes:

<div id="sub-section">
    <h2>Hello World</h2>
</div>

What I tried

I tried to go the reversed way with a query like this: "/*/*[not(@id='test')]" But it works not fine for nested HTML structures. Sometimes, depending on the structure, it removes all nodes.

Whats the way to go here?

So I could pick the one I need with `$element = $domDocument->getElementById($id)` and how could I create a new document? `$newDocument = new \DOMDocument();` and `$newDocument->appendChild($element);` gives me: `Uncaught DOMException: Wrong Document Error` — Vueer, Nov 19 '22 at 15:47
https://stackoverflow.com/questions/5735857/php-domdocument-move-nodes-from-a-document-to-another may help. — Nigel Ren, Nov 19 '22 at 15:51
you need to traverse the dom tree vertically. Get all nodes in the document. Traverse each node and mark the one where your element is present. Delete the rest — SoWhat, Nov 19 '22 at 16:04

Markus Zeller · Accepted Answer · 2022-11-19T16:16:11.400

That logic is strange. How do you know then what to keep? What in a nested case?

I would pick the ones I need and copy to a new document.

Clone a node to a new document

$xml = <<<'_XML'
<section>
  <div id="first-section">
    <ul>
      <li>Test</li>
      <li>Test</li>
    </ul>
  </div>
  <div id="second-section">
    <ul>
      <li>Test</li>
      <li>Test</li>
    </ul>
    <div id="sub-section">
      <h2>Hello World</h2>
    </div>
  </div>
  <div id="third-section">
    <ul>
      <li>Test</li>
      <li>Test</li>
    </ul>
  </div>
</section>
_XML;

libxml_use_internal_errors(true);
$doc = new DOMDocument();
$doc->loadHTML($xml);

$newDoc = new DOMDocument();
$newDoc->appendChild($newDoc->importNode($doc->getElementById('sub-section'), true));

echo $newDoc->saveHTML();

Extract only one node

When you only need just one node, you can even easier go with

libxml_use_internal_errors(true);
$doc = new DOMDocument();
$doc->loadHTML($xml);
echo $doc->saveHTML($doc->getElementById('sub-section'));

Output

The same output with both examples.

<div id="sub-section">
      <h2>Hello World</h2>
    </div>

Demo

https://3v4l.org/ttTS6

Thanks! `$doc->saveHTML($doc->getElementById('sub-section'));` seems not to be of type DOMDocument. How could I handle this last challenge? In the function I need to call it needs to be a DOMDocument.. — Vueer, Nov 19 '22 at 21:52
This will happen when `$doc->getElementById('sub-section')` will return null, namingly that element was not found. — Markus Zeller, Nov 19 '22 at 21:55

How to delete all nodes from DOMDocument except custom ones?

1 Answers1

Clone a node to a new document

Extract only one node

Output

Demo