Best way to verify site exists?

How would you all recommend that a site exists?
Should I get_headers() and then search through the headers for "200" for the status code? Well I want to allow 301, or 302 also.
What should I do?

 

fsockopen on port 80, read the headers sent back to make sure your code isnt being tricked.

 

thanks, will check it out

 

Hmm, it doesn't seem to make much sense to me. How to I read the headers?

 

get_headers() should be easier to use.

function url_exists($url)
{
	$headers = get_headers($url);
	return preg_match('/^HTTP\/\d\.\d\s+(200|301|302)/', $headers[0]);
}

PHP:

 

nico is right if you have get headers, then maybe you should use that, although nicos function doesn't return the response code, just true or false.

<?
function httpCode( $url )
{
	if( preg_match("/^https?:\\/\\//si", $url ) )
		$url = preg_replace("/^https?:\\/\\//si", "", $url );
		
	$sock = fsockopen( $url, 80, $errno, $errstr );
	
	if( !$sock ) 
		return false;
	
	$out = "GET / HTTP/1.1\r\n";
    $out .= "Host: $url\r\n";
   	$out .= "Connection: Close\r\n\r\n";
	
	fwrite( $sock, $out );
   	
	while( !feof($sock) )
		if( preg_match("/HTTP\\/1.[0-9] ([0-9]+)/si", fgets( $sock, 128 ), $http ))
			return $http[1];
	
	fclose($sock);
}

printf("Server exited with response code %d", httpCode( "digitalpoint.com" ) );

PHP:

Thats how I would have done it before php5

 

thanks both, I will rep if I can

 

I thought he just wanted to verify if the site exists, and not the actual status.

 

Good point about not knowing if OP wants to check existence vs. status, so I figured I'd throw my own curl-based code into the mix (see bottom of post).

It's very well commented, easy to use. Run

$header_array = check_url('SOME_URL',true);

Code (markup):

To get all the headers for a site, INCLUDING redirects. So, for example, if you use 'http://google.com' as the URL in the call, you'll get back TWO headers: the first a '301 moved' and then a '200 ok'.

If you call

$header_array = check_url('SOME_URL',false);

Code (markup):

You get back only ONE header, because the false flag tells us not to follow redirects.

save the code below into a file and run it -- the output data is very clear.

<?
	function parse_HTTP_header($HTTPheader)
	{
		/* 
			First, see how *many* headers we have.  For example, in the case of
		 	a 3xx redirect, we'll have TWO headers: a 3xx redirect, and then
		 	the header (hopefully a 200) we got redirected to.
		*/
		$headers = preg_split('/\r\n\r\n/', $HTTPheader,-1,PREG_SPLIT_NO_EMPTY); 
		
		/*
			ok, now we have an array containing a number of headers.  For each one,
			create a map containing the header information.
			
			The first line of a header contains: 
					<server version> <response code> <description> \r\n
			
			Every other line contains:
			 		<atribute> : <value> \r\n
		
			So we need to split out each long header line into separate lines (which are terminated by \r\n)
			and then split each of THOSE lines into constituent data or attribute/value pairs
			as described above.
		*/

		$header_blocks = array();
		
		foreach($headers as $header) {
			$lines = preg_split('/\r\n/',$header,-1,PREG_SPLIT_NO_EMPTY);

			/* 
				now we have each line of the header in an array called $lines
			 	So now we split *that* up as follows:
			 	
				The *first* entry in the array contains response code data in this form:
					<server version> <response code> <description> (e.g. HTTP/1.1 200 OK)
				
				Every subsequent entry contains attribute/value pairs in this form:
					<attribute> : <value> (Content-Type : text/html)
			 		
				So we need to treat the first entry of the array differently from the rest.
			*/
			$data = array();

			$response = preg_split('/\s/',$lines[0],3,PREG_SPLIT_NO_EMPTY);
			$data['header_server'] = $response[0];
			$data['header_code'] = $response[1];
			$data['header_description'] = $response[2];
			
			// now do each attribute/value pair
			for ($i=1; $i < count($lines); $i++) {
				list($attr,$value) = preg_split('/:/',$lines[$i],2,PREG_SPLIT_NO_EMPTY);
				$data[$attr] = trim($value);
			}

			// now attach all the data to the master array of headers we return to the caller
			$header_blocks[] = $data; 
		}
		
		return $header_blocks;
}

/*
	The following 
*/

function check_url($url,$redirect=true)
{
	$ch = curl_init($url);
	curl_setopt($ch, CURLOPT_NOBODY, 1);
	curl_setopt($ch, CURLOPT_HEADER, 1);
	curl_setopt($ch, CURLOPT_FOLLOWLOCATION, $redirect === true ? 1 : 0);
	curl_setopt($ch, CURLOPT_RETURNTRANSFER, 1);
	$result=curl_exec ($ch);
	curl_close ($ch);
	return parse_HTTP_header($result);
}

/*------------------------------------------------------------------------------------
									TEST CALL
Rather than explain what comes back in the $headers array, just run the code below
and look at the output onscreen.  It's all quite clear.
------------------------------------------------------------------------------------*/

	echo '<pre>';
	$headers = check_url('http://google.com',true);
	print_r($headers);
	echo '</pre>';
?>

PHP:

 

You were right, I didn't need the status was just using that to make sure it exists lol

 

function url_exists($url)
{
$result = parse_url($url);
if(isset($result['host'])) {
$cmdResult = gethostbynamel($result['host']);

if($cmdResult) {
return true;
} else {
return false;
}
} else {
return false;
}
}

 

oops, sucker for not reading properly lol....