curl?

[Dtonlinegames]

Hey guys I was wondering how youd do something like the W3C checker, like user supplied url and the script checks elements on the page.

 

I dont know where to start with this to be honest.

 

cheers!

[neogemima]

Wow, that's a REALLY broad question.  I would start by reading all information available on whatever code you would like to use to do that.  If it were me, and I've never even considered how to build one until two seconds ago, I would create a database that uses php to log all content in one's source file and do some checks that way.

 

You'd have to use scripts that check the header of the file and then check their file based on what type of header they are using.  Basically check their content against that header's library of acceptable scripts and see if they match.

[Ken2k7]

I don't know how W3C validates pages, but it would surprise me if they use a language other than PHP to do so. Frankly, I'm not sure how to do all of it with PHP.

[thebadbad]

Yes, you can use cURL to read a remote page's source code, and then 'check' it (whatever that involves) using e.g. regular expressions (using the preg_* functions).

 

A vague answer to a vague question

[Dtonlinegames]

'm not sure if fopen has the functionality I need it to have but I've never had a reason to use Curl so its a bit like argh!

 

But basically I want to open the supplied URL check it has certain elements and how many of each and give a score at the end of it. I wouldnt really want to store the users data cos of protection etc but I guess I'm going to need a copy of it somewhere to referance and read would probably be faster and sue less bandwidth too.

 

Dunno fopen? or curl? lol

[thebadbad]

An example for you:

 

<?php
ini_set('user_agent', 'Mozilla/5.0 (Windows; U; Windows NT 6.0; da; rv:1.9.0.10) Gecko/2009042316 Firefox/3.0.10');
$url = 'http://google.com/';
$source = file_get_contents($url);
$count = preg_match_all('~<div[^>]*>~i', $source);
echo "$count div elements were found on $url";
?>

[Axeia]

fopen wont do as you can't check reponse headers with it an if it really is similar to the w3c service you'll want to know the headers as well as sending an UTF-8 header eliminates the need for a meta tag (Unless you're using IE!).. or at least I think it was, been using the meta anyways for IE.

[Dtonlinegames]

I've used file_get_contents and preg_match_all seemed to give me what I needed just a lot of syntax errors with the wild cards in the preg matches! confusing. its getting there I'll post a link when its done

[JonnoTheDev]

file_get_contents() is slow compared to a curl request.