I have 2 methods that I came up with, I personally like the first, since its way more simple, and works with all TLD extensions, but I might not be seeing all sides of this.



function getDomainName($ful_url) {
$ref = parse_url($full_url);
if ($ref['host'] != "") {
$ref_check = substr($ref['host'], 0, 4);

// check for www
if ($ref_check == "www.") {
$ref['host'] = substr($ref['host'], 4);
}
} else {
$ref['host'] = "BAD_URL";
}

// Output filtered url
return $ref['host'];
}

vs.

function getDomainName($url) {
$url = strtolower($url);
if (substr($url, 0, 7) == 'http://')
$url = substr($url, 7);
if (strpos($url, '/'))
$url = substr($url, 0, strpos($url, '/'));
while (strpos($url, '.') != strrpos($url, '.'))
$url = substr($url, strpos($url, '.')+1);
return $url;
}

EDIT: Sorry, misread original post

URLs don't have to have "www" in them. Your code dies if you feed it something like "http://slurp.junk.co.jp".

My code doesn't need a www. It strips the www if its there. 2nd piece of code only works with normal tld, like .com/.net and doesnt with 2 part tlds like co.uk

First one should work with everything.

My code doesn't need a www. It strips the www if its there. 2nd piece of code only works with normal tld, like .com/.net and doesnt with 2 part tlds like co.uk

First one should work with everything.

Test it yourself. It clearly returns "BAD_URL" if it doesn't find "www." in the array.

Test it yourself. It clearly returns "BAD_URL" if it doesn't find "www." in the array.
No, it doesn't. It returns it if its not a valid URL.

I personally like this



<?php
// get host name from URL
preg_match('@^(?:http://)?([^/]+)@i',
"http://www.php.net/index.html", $matches);
$host = $matches[1];

// get last two segments of host name
preg_match('/[^.]+\.[^.]+$/', $host, $matches);
echo "domain name is: {$matches[0]}\n";
?>

The regex is pretty good, I like the "until slash" part, but it does not validate whether it contains any periods. For example, it would also match http://localhost It will also validate against http://192.168.0.1

The second part will fall apart if you have country tld's. For example, blah.co.uk

This is fine, until someone fat fingers something that gets past the regex. Users will do the most insane things to data input screens. And then there are my own coding errors :)

No, it doesn't. It returns it if its not a valid URL.

Well then I guess the PHP installations on my test machines must be horribly broken because that's what it does on every single one of them. I don't know what else to say other than "don't quit your day job".

Why don't you just use parse_url()?

Well then I guess the PHP installations on my test machines must be horribly broken because that's what it does on every single one of them. I don't know what else to say other than "don't quit your day job".

Right back at you, buddy. I also suggest you get a pair of eyeglasses. :rolleyes: There is a } you are not seeing.

Why don't you just use parse_url()?

I did, in the first example.... just wanted to see if its a sound code.

Well then I guess the PHP installations on my test machines must be horribly broken because that's what it does on every single one of them. I don't know what else to say other than "don't quit your day job".
Right back at you, buddy. I also suggest you get a pair of eyeglasses. There is a } you are not seeing.

This might shed some light in a test situation: http://www.1921681100.com/public/623.php