Created
July 21, 2016 14:56
-
-
Save daveajones/93c2401d263762b5327bb3be603db254 to your computer and use it in GitHub Desktop.
A php function to follow all redirects and return the final destination url.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
//Follow redirects to get to the final, good url | |
function get_final_url($url, $timeout = 5, $count = 0) | |
{ | |
$count++; | |
$url = clean_url($url); | |
$ua = 'Mozilla/5.0 (Windows NT 6.1; WOW64; rv:18.0) Gecko/20100101 Firefox/18.0'; | |
$cookie = tempnam("/tmp", "CURLCOOKIE"); | |
$curl = curl_init(); | |
curl_setopt($curl, CURLOPT_USERAGENT, $ua); | |
curl_setopt($curl, CURLOPT_URL, $url); | |
curl_setopt($curl, CURLOPT_COOKIEJAR, $cookie); | |
curl_setopt($curl, CURLOPT_ENCODING, ""); | |
curl_setopt($curl, CURLOPT_RETURNTRANSFER, true); | |
curl_setopt($curl, CURLOPT_AUTOREFERER, true); | |
curl_setopt($curl, CURLOPT_CONNECTTIMEOUT, $timeout); | |
curl_setopt($curl, CURLOPT_TIMEOUT, 30); | |
curl_setopt($curl, CURLOPT_SSL_VERIFYPEER, false); | |
curl_setopt($curl, CURLOPT_SSL_VERIFYHOST, 2); | |
$content = curl_exec($curl); | |
$response = curl_getinfo($curl); | |
curl_close($curl); | |
unlink($cookie); | |
//loggit(3, "DEBUG: get_final_url($url) returned: [".$response['http_code']."]"); | |
//Normal re-direct | |
if ($response['http_code'] == 301 || $response['http_code'] == 302 || $response['http_code'] == 303) { | |
//loggit(3, "DEBUG: ".print_r($response, TRUE)); | |
ini_set("user_agent", $ua); | |
$headers = get_headers($response['url']); | |
$location = ""; | |
foreach ($headers as $value) { | |
//loggit(3, "HEADER: [[".trim(substr($value, 9, strlen($value)))."]]"); | |
if (substr(strtolower($value), 0, 9) == "location:") { | |
//loggit(3, "DEBUG: This was a normal http redirect."); | |
loggit(3, "HEADER: [[".trim(substr($value, 9, strlen($value)))."]]"); | |
return get_final_url(trim(substr($value, 9, strlen($value))), 8, $count); | |
} | |
} | |
} | |
//Meta-refresh redirect | |
if (preg_match("/meta.*refresh.*URL=.*(http[^'\"]*)/i", $content, $value)) { | |
loggit(3, "DEBUG: This was a meta-refresh redirect."); | |
if (strpos($value[1], "http") !== FALSE) { | |
return get_final_url($value[1]); | |
} | |
} | |
//Javascript re-direct | |
if (preg_match("/window\.location\.replace\('(.*)'\)/i", $content, $value) || preg_match("/window\.location\=\"(.*)\"/i", $content, $value)) { | |
//loggit(3, "DEBUG: This was a javascript redirect."); | |
if (strpos($value[1], "http") !== FALSE) { | |
return get_final_url($value[1]); | |
} else { | |
return $response['url']; | |
} | |
} else { | |
//loggit(3, "DEBUG: No redirection."); | |
return $response['url']; | |
} | |
} |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment