xlttj
xlttj

Reputation: 1158

Use XPath with PHP's SimpleXML to find nodes containing a String

I try to use SimpleXML in combination with XPath to find nodes which contain a certain string.

<?php
$xhtml = <<<EOC
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN"
    "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="de" lang="de">
    <head>
        <meta http-equiv="content-type" content="text/html; charset=utf-8" />
        <title>Test</title>
    </head>
    <body>
        <p>Find me!</p>
        <p>
            <br />
            Find me!
            <br />
        </p>
    </body>
</html>
EOC;

$xml = simplexml_load_string($xhtml);
$xml->registerXPathNamespace('xhtml', 'http://www.w3.org/1999/xhtml');

$nodes = $xml->xpath("//*[contains(text(), 'Find me')]");

echo count($nodes);

Expected output: 2 Actual output: 1

When I change the xhtml of the second paragraph to

<p>
    Find me!
    <br />
 </p>

then it works like expected. How has my XPath expression has to look like to match all nodes containing 'Find me' no matter where they are?

Using PHP's DOM-XML is an option, but not desired.

Thank's in advance!

Upvotes: 4

Views: 13568

Answers (4)

Willy
Willy

Reputation: 11

I was looking for a way to find whether a node with exact value "Find Me" exists and this seemed to work.

$node = $xml->xpath("//text()[.='Find Me']");

Upvotes: 1

xlttj
xlttj

Reputation: 1158

Err, umm? But thanks @Jordy for the quick answer.

First, that's DOM-XML, which is not desired, since everything else in my script is done with SimpleXML.

Second, why do you translate to uppercase and search for an unchanged string 'Find me'? 'Searching for 'FIND ME' would actually give a result.

But you pointed me towards the right direction:

$nodes = $xml->xpath("//text()[contains(., 'Find me')]");

does the trick!

Upvotes: 1

Josh Davis
Josh Davis

Reputation: 28730

It depends on what you want to do. You could select all the <p/> elements that contain "Find me" in any of their descendants with

//xhtml:p[contains(., 'Find me')]

This will return duplicates and so you don't specify the kind of nodes then it will return <body/> and <html/> as well.

Or perhaps you want any node which has a child (not a descendant) text node that contains "Find me"

//*[text()[contains(., 'Find me')]]

This one will not return <html/> or <body/>.


I forgot to mention that . represents the whole text content of a node. text() is used to retrieve [a nodeset of] text nodes. The problem with your expression contains(text(), 'Find me') is that contains() only works on strings, not nodesets and therefore it converts text() to the value of the first node, which is why removing the first <br/> makes it work.

Upvotes: 11

user443346
user443346

Reputation:

    $doc = new DOMDocument();
    $doc->loadHTML($xhtml);

    $xPath = new DOMXpath($doc);
    $xPathQuery = "//text()[contains(translate(.,'abcdefghijklmnopqrstuvwxyz', 'ABCDEFGHIJKLMNOPQRSTUVWXYZ'), 'Find me')]";
    $elements = $xPath->query($xPathQuery);

    if($elements->length > 0){

    foreach($elements as $element){
        print "Found: " .$element->nodeValue."<br />";
    }}

Upvotes: 0

Related Questions