Created
May 13, 2011 09:02
-
-
Save brucealdridge/970235 to your computer and use it in GitHub Desktop.
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
<?php | |
function doubleclick_adplanner() { | |
$data = file_get_contents('http://www.google.com/adplanner/static/top1000/'); | |
preg_match_all('/href=["\']([^"\']+)?["\']target=["\']_blank/im',$data,$matches); | |
return implode("\n",$matches[1]); | |
} | |
function alexa_global($num = 500) { | |
static $topsites; | |
if($topsites) return implode("\n", array_slice($topsites,0,$num)); | |
$topsites = array(); | |
if (! file_put_contents('top-1m.csv.zip', | |
file_get_contents('http://s3.amazonaws.com/alexa-static/top-1m.csv.zip') | |
) | |
) { | |
throw new Exception('Unable to write file check permissions'); | |
return; | |
} | |
$res = exec('gunzip -S .zip top-1m.csv.zip'); | |
if (file_exists('top-1m.csv')) { | |
$data = file_get_contents('top-1m.csv'); | |
$lines = explode("\n",$data); | |
foreach($lines as $line) { | |
$parts = explode(',',$line); | |
if (count($parts) == 2) | |
$topsites[$parts[0]] = 'http://'.$parts[1]; | |
} | |
}else{ | |
throw new Exception('Unable to extract zip or filename has changed'); | |
} | |
unlink('top-1m.csv'); | |
return implode("\n", array_slice($topsites,0,$num)); | |
} | |
function quantcast($num = 500) { | |
static $topsites; | |
if($topsites) return implode("\n", array_slice($topsites,0,$num)); | |
$topsites = array(); | |
if (! file_put_contents('Quantcast-Top-Million.txt.zip', | |
file_get_contents('http://www.quantcast.com/quantcast-top-million.zip') | |
) | |
) { | |
throw new Exception('Unable to write file check permissions'); | |
return; | |
} | |
exec('gunzip -S .zip Quantcast-Top-Million.txt.zip'); | |
if (file_exists('Quantcast-Top-Million.txt')) { | |
$data = file_get_contents('Quantcast-Top-Million.txt'); | |
$lines = explode("\n",$data); | |
foreach($lines as $line) { | |
preg_match('/(?:\d+)(?:\t)+?([^\s]+)/',$line,$matches); | |
if (count($matches)) | |
$topsites[] = 'http://'.$matches[1]; | |
} | |
}else{ | |
throw new Exception('Unable to extract zip or filename has changed'); | |
} | |
unlink('Quantcast-Top-Million.txt'); | |
return implode("\n", array_slice($topsites,0,$num)); | |
} | |
function alexa_us($num = 500) { | |
static $topsites; | |
if ($topsites) return implode("\n", array_slice($topsites,0,$num)); | |
$topsites = array(); | |
for ($i = 0; $i < 20; $i++) | |
{ | |
$url = 'http://www.alexa.com/topsites/countries'.($i ? ';'.$i : '').'/US'; | |
$data = file_get_contents($url); | |
preg_match_all('/topsites-label["\']\>([^<]+)?/im',$data,$matches); | |
$topsites = array_merge($topsites,$matches[1]); | |
} | |
foreach($topsites as &$site) { | |
$site = 'http://'.$site; | |
} | |
return implode("\n", array_slice($topsites,0,$num)); | |
} | |
file_put_contents('lists/DoubleClick Ad Planner.txt',doubleclick_adplanner()); | |
file_put_contents('lists/Alexa 500.txt',alexa_global(500)); | |
file_put_contents('lists/Alexa Global 1000.txt',alexa_global(1000)); | |
file_put_contents('lists/Alexa10k.txt',alexa_global(10000)); | |
file_put_contents('lists/Alexa US 500.txt',alexa_us(500)); | |
file_put_contents('lists/Alexa US 100.txt',alexa_us(100)); | |
file_put_contents('lists/Quantcast10K.txt',quantcast(10000)); |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment