php sockets?

[acidglitter]

So I want to make a page on my site, that can (on its own) go to another site I tell it to go to and come back with a certain HTML code. How could I do that?? I looked up some things about php sockets and parsers but it confused me..

[Glyde]

Here's a simple example that'll pull information off of a web page

<?php
function getHTML($website) {
    $url = parse_url($website);
    $handle = fsockopen($url['host'], 80);
    $request = "GET {$url['path']} HTTP/1.1\r\n";
    $request .= "Host: {$url['host']}\r\n\r\n";
    fputs($handle, $request, strlen($request));
    $response = "";
    while (!feof($handle)) $response .= fgets($handle, 1024);
    print $response;
}
?>

That should work, then you can manipulate it as you feel needed

[acidglitter]

Thanks 

 

I'm not really sure how to use it though. I put

getHTML("http://vanillaeyeliner.com");

 

but I got this error

 

Bad Request

Your browser sent a request that this server could not understand.

 

Invalid URI in request GET HTTP/1.1

 

Additionally, a 404 Not Found error was encountered while trying to use an ErrorDocument to handle the request.

Apache/1.3.37 Server at x.geminiserver.com Port 80

[rhyspaterson]

You need to give the variable something to work with.

 

$website = 'http://vanillaeyeliner.com';

[acidglitter]

You need to give the variable something to work with.

 

$website = 'http://vanillaeyeliner.com';

 

I tried that and it still had the same error message

[rhyspaterson]

You must call the function and give it the variable.

 

<?php
function getHTML($website) {
    $url = parse_url($website);
    $handle = fsockopen($url['host'], 80);
    $request = "GET {$url['path']} HTTP/1.1\r\n";
    $request .= "Host: {$url['host']}\r\n\r\n";
    fputs($handle, $request, strlen($request));
    $response = "";
    while (!feof($handle)) $response .= fgets($handle, 1024);
    print $response;
}
?>

 

$website = 'http://www.google.com'
getHTML($website);

[acidglitter]

Okay there aren't any error messages now, but the page is blank.. How do I show it?

[Glyde]

I forgot to add something in the code.  Try it now.

<?php
function getHTML($website) {
    $url = parse_url($website
    $url['path'] = substr($url['path'], 0, 1) == "/" ? $url['path'] : "/" . $url['path'];
    $handle = fsockopen($url['host'], 80);
    $request = "GET {$url['path']} HTTP/1.1\r\n";
    $request .= "Host: {$url['host']}\r\n\r\n";
    fputs($handle, $request, strlen($request));
    $response = "";
    while (!feof($handle)) $response .= fgets($handle, 1024);
    print $response;
}
?>

[acidglitter]

Thats awesome. Thanks ;D

 

So my code looks like this:

<?php
function getHTML($website) {
    $url = parse_url($website);
    $url['path'] = substr($url['path'], 0, 1) == "/" ? $url['path'] : "/" . $url['path'];
    $handle = fsockopen($url['host'], 80);
    $request = "GET {$url['path']} HTTP/1.1\r\n";
    $request .= "Host: {$url['host']}\r\n\r\n";
    fputs($handle, $request, strlen($request));
    $response = "";
    while (!feof($handle)) $response .= fgets($handle, 1024);
    print $response;
}

$website = 'http://www.google.com';
getHTML($website);

?>

 

How should I change it so it still gets the entire page, but then the only thing it displays is the Google logo image?

[taith]

function getHTML($website) {
    $url = parse_url($website);
    $url['path'] = substr($url['path'], 0, 1) == "/" ? $url['path'] : "/" . $url['path'];
    $handle = fsockopen($url['host'], 80);
    $request = "GET {$url['path']} HTTP/1.1\r\n";
    $request .= "Host: {$url['host']}\r\n\r\n";
    fputs($handle, $request, strlen($request));
    $response = "";
    while (!feof($handle)) $response .= fgets($handle, 1024);
    print strip_tags($response,'<img>');
}

[acidglitter]

So it showed this

<img alt='' height=1 id=gbari width=1>document.getElementById('gbarc').style.height=(document.getElementById('gbari').height=window.gbar.getPad())+'px';document.getElementById('gbar').innerHTML=window.gbar.getHtml([['w','Web','',1],['i','Images','http://images.google.com/imghp',1],['v','Video','http://video.google.com/',1],['n','News','http://news.google.com/nwshp',1],['l','Maps','http://maps.google.com/maps',1],['m','Gmail','http://mail.google.com/mail',1],['','more','#',3],['b','Blog Search','http://blogsearch.google.com/',2],['j','Blogger','http://www.blogger.com/',2],['p','Books','http://books.google.com/bkshp',2],['c','Calendar','http://www.google.com/calendar',2],['o','Documents','http://docs.google.com/',2],['e','Finance','http://finance.google.com/finance',2],['g','Groups','http://groups.google.com/grphp',2],['z','Labs','http://labs.google.com/',2],['0','Orkut','http://www.orkut.com/',2],['t','Patents','http://www.google.com/ptshp',2],['q','Photos','http://picasaweb.google.com/home',2],['f','Products','http://www.google.com/prdhp',2],['y','Reader','http://www.google.com/reader',2],['s','Scholar','http://scholar.google.com/schhp',2]]);iGoogle | Sign in<img alt="Google" height=110 src="/intl/en_ALL/images/logo.gif" width=276>   Advanced Search  Preferences  Language ToolsAdvertising Programs - Business Solutions - About Google©2007 Google

 

I meant more of how could I get it to get the exact address of the image, like only show this

intl/en_ALL/images/logo.gif

[Glyde]

<?php
function getImages($website) {
    $url = parse_url($website);
    $url['path'] = substr($url['path'], 0, 1) == "/" ? $url['path'] : "/" . $url['path'];
    $handle = fsockopen($url['host'], 80);
    $request = "GET {$url['path']} HTTP/1.1\r\n";
    $request .= "Host: {$url['host']}\r\n\r\n";
    fputs($handle, $request, strlen($request));
    $response = "";
    while (!feof($handle)) $response .= fgets($handle, 1024);
    $response = strip_tags($response,'<img>');
    preg_match_all("@<img[^>]+?src=\"([^\"]+)\"[^>]+?>@is", $response, $matches);
    return $matches;
}
print_r(getImages("http://www.google.com"));
?>

[acidglitter]

Thanks for all the help so far

This is probably my last question about it..

 

If I wanted to see what my myspace inbox looked like, using the original code it would look like this..

<?php
function getHTML($website) {
    $url = parse_url($website);
    $url['path'] = substr($url['path'], 0, 1) == "/" ? $url['path'] : "/" . $url['path'];
    $handle = fsockopen($url['host'], 80);
    $request = "GET {$url['path']} HTTP/1.1\r\n";
    $request .= "Host: {$url['host']}\r\n\r\n";
    fputs($handle, $request, strlen($request));
    $response = "";
    while (!feof($handle)) $response .= fgets($handle, 1024);
    print $response;
}

$website = 'http://messaging.myspace.com/index.cfm?fuseaction=mail.inbox';
getHTML($website);

?>

 

But the problem is I need to be signed in to see it.. Is there a way I could put my username/password in the socket so it could still get the page html?

[acidglitter]

*bump*

[trq]

This whole process would be much easier using curl.

[Glyde]

Yea, it would be, but at first I didn't know he needed to be signed in.  Plus, I don't like coding for 3rd party extensions to the public because a lot of people don't have them installed.