retrieve all images path from string with PHP - php

Question

How do I get all images path from a string?
Note I just want the path containing the word "media".
For example given this string (part of the DOM)
<div class="my-class">
<img src="http://my-website.com/cache/media/2017/10/img67.jpeg" class="" alt="test" width="120" height="100">
<img src="http://my-website.com/cache/2017/10/img68.png" class="" alt="test" width="120" height="100">
<img src="http://my-website.com/cache/media/2017/10/img69.jpg" class="" alt="test" width="120" height="100">
<h2 class="uk-margin-top-remove">About us</h2>
</div>
I want an array containing a similar result:
array(
[0] => "http://my-website.com/cache/media/2017/10/img67.png"
[1] => "http://my-website.com/cache/media/2017/10/img69.png"
);
I don't want the second img because src attribute doesn't contain the word "media".

Answer 1

You could use preg_match_all() to get URLs but it is even better to use a DOM reader.
$str = '<div class="my-class">
<img src="http://my-website.com/cache/media/2017/10/img67.jpeg" class="" alt="test" width="120" height="100">
<img src="http://my-website.com/cache/2017/10/img68.png" class="" alt="test" width="120" height="100">
<img src="http://my-website.com/cache/media/2017/10/img69.jpg" class="" alt="test" width="120" height="100">
<h2 class="uk-margin-top-remove">About us</h2>
</div>' ;
$matches = [] ;
preg_match_all('~(http\://my-website\.com/cache/media/(.*?))"~i', $str, $matches) ;
var_dump($matches[1]);
Will returns :
array(2) {
[0]=>
string(52) "http://my-website.com/cache/media/2017/10/img67.jpeg"
[1]=>
string(51) "http://my-website.com/cache/media/2017/10/img69.jpg"
}

Answer 2

Some boilerplate code to get you started:
<?php
$data = <<<DATA
<div class="my-class">
<img src="http://my-website.com/cache/media/2017/10/img67.jpeg" class="" alt="test" width="120" height="100">
<img src="http://my-website.com/cache/2017/10/img68.png" class="" alt="test" width="120" height="100">
<img src="http://my-website.com/cache/media/2017/10/img69.jpg" class="" alt="test" width="120" height="100">
<h2 class="uk-margin-top-remove">About us</h2>
</div>
DATA;
# set up the dom
$dom = new DOMDocument();
$dom->loadHTML($data, LIBXML_HTML_NOIMPLIED | LIBXML_HTML_NODEFDTD);# | LIBXML_COMPACT | LIBXML_NOENT );
# set up the xpath
$xpath = new DOMXPath($dom);
foreach ($xpath->query("//img[contains(#src, '/media/')]/#src") as $image) {
echo $image->nodeValue . "\n";
}
Which yields
http://my-website.com/cache/media/2017/10/img67.jpeg
http://my-website.com/cache/media/2017/10/img69.jpg
This loads the DOM and uses an xpath query for every image where we'll loop over afterwards.
If for some reasons (why?) you are unable to use a DOM parser, your could use the secondbest option:
<img
(?s:(?!>).)+?
src=(['"])
(?P<src>(?:(?!\1).)+?/media/.*?\1)
And use the src group, see a demo on regex101.com.