Comments on: PHP: Relative URL to Absolute URL

By: Mohit

Mohit — Mon, 09 Jan 2012 04:17:57 +0000

Nice article I have used this PHP Script in my website and I have better performance in SEO also

By: NubbyNubkins

NubbyNubkins — Mon, 26 Sep 2011 20:51:40 +0000

My apologies. I forgot to clarify that the solution I used is not in the script itself but when the function is called.

By: NubbyNubkins

NubbyNubkins — Mon, 26 Sep 2011 20:48:16 +0000

I was experimenting with this script to see if my solution was slower and it definitely was. My solution was also far more complex. Thank you very much for offering this invaluable tool.
I would also like to mention that this script sometimes has a problem dealing with empty values. While this may not often occur for many people, the way I wrote my link spider would result in empty array values when duplicates or non-page links were found. Because of the location within my script where I needed the absolute url conversion, empty values would appear directly prior to this function. The e-warning would occur on line 327: “return $parts”. and directly under the “if (!$decode)”. The simple solution I used was to include:

if(($var != “”) || (!empty($var))) {url_to_absolute( $baseUrl, $relativeUrl);}

I just thought you might want to know that there is a potential issue when it concerns empty values.

By: Nitin

Nitin — Sun, 17 Apr 2011 14:34:33 +0000

The complexity in the Dave’s solution has been added intentionally and well explained. If you read the tutorial, you would know why he avoided using the

parse_url

function and wrote his custom functions.

Although I am not sure how many of his assumptions are still valid, it has been long when I looked at the code (it works so well!!)

By: Matt

Matt — Sun, 17 Apr 2011 12:41:25 +0000

This one is way more simple, and even works for parent directories (..) and recognizes base direcories (.): http://web-o-blog.blogspot.com/2011/04/function-absoluteurlbaseurl-relativeurl.html

By: Dinar Q.

Dinar Q. — Wed, 30 Jun 2010 14:20:18 +0000

i have tried to modify this way : function join_url( $parts, $encode=TRUE) – does not help – also does not work.
it returns empty string, as i remember.

By: Dinar Q.

Dinar Q. — Tue, 29 Jun 2010 15:57:30 +0000

address like http://kk.wiktionary.org/wiki/Басты_бет does not work, example of such link, in this page: http://tor.kz/index.php/%D0%91%D0%B0%D1%81%D1%82%D1%8B_%D0%B1%D0%B5%D1%82 .

By: Nitin

Nitin — Wed, 19 May 2010 21:11:30 +0000

split_url() over parse_url(), explained here: http://bit.ly/diaXxL

Cheers,

By: Tyler Akins

Tyler Akins — Wed, 19 May 2010 17:49:16 +0000

Oh, I’m sorry that I didn’t mention this in my last post, but why don’t you use parse_url() instead of your split_url()? I’m a fan of using built-in functions when possible for speed and because a larger developer base will be continually reviewing its functionality.

By: Tyler Akins

Tyler Akins — Wed, 19 May 2010 17:44:57 +0000

This function doesn’t really work well for file:/// URLs. For instance, url_to_absolute(‘file:///path/file.html’, ‘image.gif’) returns false. Would you be able to change line 82 a bit so that if the scheme is ‘file’, it won’t require ‘host’?

(Yeah, I can make the change locally, but I’d rather the change gets back into the main distribution.)