Grabbing HTTP Headers

jeofmoyster

Is it possible to grab the Http headers of a page? For example, I want to be able to detect the Last-Modified header of a webpage to archive / detect when a page has been modified. How would I go about detecting just the headers, not necessarily anything else about the page? Is Curl useful here?

Weedpacket

You could use cURL; but [man]fsockopen[/man] would be enough. See example one on that page, but use HEAD instead of GET.

Installer

apache_response_headers() is also a possibility (with "ini_set('last_modified', '1')").

edit: Or set "last_modified" in ".htaccess".

xgarreau

$fp = fsockopen($host, $port, $errno, $errstr, 30);
$http_code = 0;
$new_loc = "";
if ($fp) {
	$out = "HEAD $page HTTP/1.1\r\n";
	$out .= "Host: $host\r\n";
	$out .= "Connection: Close\r\n\r\n";

fwrite($fp, $out);
do {
	$header = fgets ($fp, 4096);
} while ($header && (strstr($header, "Last-Modified:") === false));
if ($header) {
	$theader = split(" ", $header);
	$lastmodified = $theader[1];
}

fclose($fp);
}

$lastmodified will the last modified header content, if it was present. Else, it will be unset.

Weedpacket

Originally posted by xgarreau
...
[/B]

I'd've thought a link to [man]fsockopen[/man] would have been enough, and provide more information.

xgarreau

Well ...

Let's say an example AND a link is better to fully understand ...

Regards

Weedpacket

Originally posted by xgarreau
Well ...

Let's say an example AND a link is better to fully understand ...

Regards

Nah, an example would be found by following the link, and just providing code that 90% of the time just gets cut-and-pasted does nothing to improve understanding.