Question

I have an XML document from which I want to extract some data:

<tnt:results>
<tnt:result>
<Document id="id1">
<impact _blabla_ for="tree.def" name="Something has changed"
select="moreblabla">true</impact>
<impact _blabla_ for="plant.def" name="Something else has changed"
select="moreblabla">true</impact>
</Document>
</tnt:result>
</tnt:results>

in reality there is no new line — it’s one continuous string and and there can be multiple < Document > elements. I want to have a regular expression that extracts:

• id1
• tree.def / plant.def
• Something has changed / Something else has changed

I was able to come up with this code so far, but it only matches the first impact, rather than both of them:

preg_match_all('/<Document id="(.*)">(<impact.*for="(.*)".*name="(.*)".*<\/impact>)*<\/Document>/U', $response, $matches);

The other way to do it would be to match everything inside the Document element and pass it through a RegEx once more, but I thought I can do this with only one RegEx.

Thanks a lot in advance!

Answer 1

Just use DOM, it’s easy enough:

$dom = new DOMDocument;
$dom->loadXML($xml_string);

$documents = $dom->getElementsByTagName('Document');
foreach ($documents as $document) {
    echo $document->getAttribute('id');     // id1    

    $impacts = $document->getElementsByTagName('impact');
    foreach ($impacts as $impact) {
        echo $impact->getAttribute('for');  // tree.def
        echo $impact->getAttribute('name'); // Something has changed
    }
}