<?php
$file = false;
if(isset($_GET['load'] && is_string($_GET['load'])) {
$tmp = stripclashes($_GET['load']);
$tmp = str_replace(".","",$tmp);
$file = $tmp . '.php';
}
if($file != false && file_exists($file) && is_readable($file)) {
require_once $file;
}
?>
called via file.php?load=test
That process the PHP file, and as long as you spit out HTML from the file simply
target = document.getElementById('page');
target.innerHTML = response.responseText;
Now, i'm fairly certain parts of that are insecure, you could have a whitelist of allowable requires. It should ideally be looking in a specific directory for the files also. I'm honestly not all too sure about directly dumping the responseText back into a DIV either, security wise as it's ripe for XSS. But it's the end of the day and I haven't looked up anything on that one. Be aware, without any kind of checking on this, you could have a user being directed to a third party site using file_get_contents, which would be a Very Bad Thing. You could eval in PHP a file_get_contents request, which... is well, Very Very Bad. For example try
<?php
echo file_get_contents("http://www.google.com");
?>
But I fear I must ask here, why are you doing it this way? This seems a very roundabout way to achieve a Hyperlink.
Is this AJAX for AJAXs sake?