asked on

php validate URL

Hi E´s,
I need to validate a URL in PHP.
I search in Internet for try get solution for the problem and I don't see any good script that work perfect.
The most perfect that I found was this code:

<?php
$url = "http://www.example.com/index2.php";
        if (!preg_match("#^http://www\.[a-z0-9-_.]+\.[a-z]{2,4}$#i",$url)) {
        echo "wrong url";
        } else {
        echo "ok";
        }
?>

Open in new window

The code above work partial fine for the domain names, like http://www.example.com validate, but http://example.com (without www) not validate!
Also not validate for this kind of URL's:
http://www.example.com/index.php
http://www.example.com/friendlyurl/

Any idea to improve the regular expression or other way to validate the URL?

The best regards, JC

SOLUTION

Ray Paseur

membership

This solution is only available to members.

To access this solution, you must be a member of Experts Exchange.

Start Free Trial

Gary

This seems to work didn't do much testing;

<?php

function checkurl($url){

return(preg_match("#^(https?://)?[^/.]+(\.[^/.]+)+/?$#i",$url));
}

echo "http://www.example.com: " . checkurl("http://www.example.com")."<br>";
echo "http://www.example.com/index.php: " . checkurl("http://www.example.com/index.php")."<br>";
echo "http://www.example.com/friendlyurl/: " . checkurl("http://www.example.com/friendlyurl/")."<br>";
echo "http://example.com: " . checkurl("http://example.com")."<br>";

Open in new window

Pedro Chagas

ASKER

Hi @Gary,
I increase this line in your code:

echo "http://example: " . checkurl("http://example.com")."<br>";

and the output is:

http://www.example.com: 1
http://www.example.com/index.php: 0
http://www.example.com/friendlyurl/: 0
http://example.com: 1
http://example: 1

The number 5 should be "0", and 2 and 3 "1".
Can you improve the RE?

Hi @Ray: I will read the article!

~JC

Gary

Yes but the url you are passing has .com - if you remove that it doesn't pass.

Gary

Maybe I misunderstood I thought you only wanted the domain and nothing else.

Gary

Changed the pattern, allowed bad characters in the first attempt, who'd thought a url would so hard to validate.

return(preg_match("#^(https?://)?([\da-zA-Z\.-]+)\.([a-z\.]{2,6})([\da-zA-Z/\.-]*)*/?$#i",$url));

Pedro Chagas

ASKER

Hi @Garry,
Is possible you improve your solution for check also GET variables in the URL, like this one:

http://example.com/some.php?hhh=10&dddd=20: 0

In line above the return is "0", not validate.

Thanks.

~JC

SOLUTION

Gary

membership

This solution is only available to members.

To access this solution, you must be a member of Experts Exchange.

Start Free Trial

Gary

Check that

return(preg_match("#^(https?://)?([\da-zA-Z_\.-]+)\.([a-z\.])([\d\w/\.=\\?]*)*/?$#i",$url));

SOLUTION

Terry Woods

membership

This solution is only available to members.

To access this solution, you must be a member of Experts Exchange.

Start Free Trial

Pedro Chagas

ASKER

I forget that kind of domains exist. now domains can be lot's of things.
For example I test this URL based in new or future domains:

http://example.games/jjj.php?uu=kjkjk

and validated and well.
But for example:

http://example.c/jjj.php?uu=kjkjk

Open in new window

, the domain is ".c", and the script validate, do not know if it good or bad. It is possible there are domains with only one character?
If not, can you please improve the RE, for not accept domains with one character?

~JC

Gary

Add it back in but use {2}
There is no extensions less than 2 characters

ASKER CERTIFIED SOLUTION

Gary

membership

This solution is only available to members.

To access this solution, you must be a member of Experts Exchange.

Start Free Trial

SOLUTION

Gary

membership

This solution is only available to members.

To access this solution, you must be a member of Experts Exchange.

Start Free Trial

Ray Paseur

... plus I'm heading to the bar!

Yes. Whenever the problem can only be solved with REGEX the truth is to be found here. http://xkcd.com/1171/

Pedro Chagas

ASKER

Based on the idea of @Ray, other way to check URL:

<?
$file = 'http://stackoverflow.com/questions/2280394/how-can-i-check-if-a-url-exists-via-php';
$file_headers = @get_headers($file);
if($file_headers[0] == 'HTTP/1.1 404 Not Found') {
    echo "não encontrado";
}
else {  
    echo "encontrado";
}
?>

Open in new window