cURL Help

[killah]

I am currently stuck on a solution. Let me explain, i have built a script which will take my test site & register the account, after that it then log's the account in.

 

After the account has logged in. I need it to send it to another script.. Ok i got the script working to this part. The script basicly work's like this.

 

Post form data -> register.php -> login.php -> mailbox.php?action=send

 

so far that work's and send's the mail. My problem stand's that the user were sending to, can be anything. This is why i have came here.

 

I need my curl script to send to usersonline.php and get a name inbetween a url link (<a href='viewuser.php?u=xx'>YY</a> xx => id, YY => name), i need to grab YY and put it in an array, so i can use it later on.

 

Is there anyway to grab a name inbetween a html tag using cURL?

 

I have the rest covered by going to usersonline.php and all.

 

Thank's.

[jxrd]

You could use DOM to do it. Or regex...but apparently that's less reliable.

[killah]

Care to post an example?

[jxrd]

Idk much about DOM, I've only heard it's the better option.

 

Regex would probably be something like this

 

preg_match($curl_return, '/<a href=\'viewuser\.php?u=[a-z]\'>([a-z])<\/a>/isS', $matches);
print_r($matches);

[killah]

Returns empty:

 

Array ( )

[jxrd]

Crap, I'm thinking of strstr() syntax

 

preg_match('/<a href=\'viewuser\.php?u=[a-z]\'>([a-z])<\/a>/isS', $curl_return, $matches);
print_r($matches);

 

Replace $curl_return with the curl return btw.

[killah]

This is what i got:

 

$mat = htmlentities(curl_exec($log));
preg_match('~<a href=\'viewuser.php?u=(.?)\'>(.?)</a>~isS', $mat, $matches);
print_r($matches);

 

except it doesnt pick up the \'viewuser.php?u=xx\'

 

if i change my pattern to '~<a>' it picks up every link on the site.

[jxrd]

Try changing (.?) to (.*?)

[killah]

Nope.

[jxrd]

This worked for me

<?php

$curl_return = '<a href=\'viewuser.php?u=something\'>username</a>';


preg_match('/<a href=\'viewuser\.php\?u=[a-z]+\'>([a-z]+)<\/a>/isS', $curl_return, $matches);
print_r($matches);

[killah]

Ok, let me explain this bit.

 

Basicly it log's into a full layout site, thus is why im using htmlentities/htmlspecialchars..

 

I need to search then for &lta href='viewuser.php?u=[0-9]'>[a-z0-9 _-!]</a>

 

Hope you can maybe figure a solution to that

[killah]

Here's an update.

 

curl_setopt($log, CURLOPT_URL, $url.'/usersonline.php');
$mat = str_replace(array('<','>'),array('[',']'), curl_exec($log));
echo $mat;
$mat = preg_replace('~\[a href=\'viewuser.php?u=(.*)\'\](.*)\[/a\]~is','[pid]$2[/pid]', $mat);
preg_match('~\[pid\](.*)\[/pid\]~is', $mat, $matches);
print_r($matches);

 

It displays:

 

[a href=viewuser.php?u=2]Kyle[/a]

 

But it simply won't grab the bit inside ]YY[

 

 

I just noticed your using preg_match when it's supposed to be preg_match_all, but now my array returns:

 

Array ( [0] => Array ( ) [1] => Array ( ) )

[jxrd]

<?php

$curl_return = '<a href=\'viewuser.php?u=something\'>username</a>';

$curl_return = str_replace(array('<', '>'), array('[', ']'), $curl_return);


preg_match('/\[a href=\'viewuser\.php\?u=[a-z]+\'\]([a-z]+)\[\/a\]/i', $curl_return, $matches);
print_r($matches);

Come on..surely you can adapt this.

[killah]

Your example work's perfect, it just doesnt work with my script for some reason :S

[jxrd]

What's the url of the page you are trying to scrape?

[killah]

I PM'ed it to you. Recheck the link if you got the disabled bit's.

[jxrd]

This script is to powerful for you. Your ip is not in the wanted list, your unwanted fool! Current IP: 82.28

Lol.

[killah]

Reload the page, as i just marked that out..

[killah]

Bump