Question

I have several types of <a> in my document. I’d like to select only the <a> whose title attribute begins with the text “more data …” like <a title="more data *">

For example given the markup below, I’d like to select the first and second <a>, but skip the third because title doesn’t begin with more data, and skip the 4th because <a> doesn’t even have a title attribute.

<a title="more data some text" href="http://mypage.com/page.html">More</a>
<a title="more data other text" href="http://mypage.com/page.html">More</a>
<a title="not needed" href="http://mypage.com/page.html">Not needed</a>
<a href="http://mypage.com/page.html">Not needed</a>

I’m using DOMXPath. How would my query look like?

$xpath = new DOMXPath($doc);
$q = $xpath->query('//a');

Answer 1

You could use a query like:

//a[starts-with(@title, "more data ")]

Which uses a predicate such that the title attribute’s value must start with the specified string.

---

An example

$doc = new DOMDocument;
$doc->loadXML('<example>
<a title="more data some text" href="http://mypage.com/page.html">More</a>
<a title="more data other text" href="http://mypage.com/page.html">More</a>
<a title="not needed" href="http://mypage.com/page.html">Not needed</a>
<a href="http://mypage.com/page.html">Not needed</a>
</example>');

$xpath = new DOMXPath($doc);
$links = $xpath->query('//a[starts-with(@title, "more data ")]');

echo "Found {$links->length} links", PHP_EOL;
foreach ($links as $link) {
    echo $link->getAttribute('href'), PHP_EOL;
}

And here’s the above example running online.