2005年4月28日星期四

Google PageRank的计算源代码

  最近对google的PageRank(网页等级)比较感兴趣,一直想知道如何不用google toolbar来获取pr值。苦苦搜索之后,找到如下代码:

 

<?php
/**
    This code is released unto the public domain
*/
//header("Content-Type: text/plain; charset=utf-8");
define('GOOGLE_MAGIC', 0xE6359A60);

//unsigned shift right
function zeroFill($a, $b)
{
    $z = hexdec(80000000);
        if ($z & $a)
        {
            $a = ($a>>1);
            $a &= (~$z);
            $a |= 0x40000000;
            $a = ($a>>($b-1));
        }
        else
        {
            $a = ($a>>$b);
        }
        return $a;
}  

function mix($a,$b,$c) {
  $a -= $b; $a -= $c; $a ^= (zeroFill($c,13));
  $b -= $c; $b -= $a; $b ^= ($a<<8);
  $c -= $a; $c -= $b; $c ^= (zeroFill($b,13));
  $a -= $b; $a -= $c; $a ^= (zeroFill($c,12));
  $b -= $c; $b -= $a; $b ^= ($a<<16);
  $c -= $a; $c -= $b; $c ^= (zeroFill($b,5));
  $a -= $b; $a -= $c; $a ^= (zeroFill($c,3));
  $b -= $c; $b -= $a; $b ^= ($a<<10);
  $c -= $a; $c -= $b; $c ^= (zeroFill($b,15));

  return array($a,$b,$c);
}

function GoogleCH($url, $length=null, $init=GOOGLE_MAGIC) {
    if(is_null($length)) {
        $length = sizeof($url);
    }
    $a = $b = 0x9E3779B9;
    $c = $init;
    $k = 0;
    $len = $length;
    while($len >= 12) {
        $a += ($url[$k+0] +($url[$k+1]<<8) +($url[$k+2]<<16) +($url[$k+3]<<24));
        $b += ($url[$k+4] +($url[$k+5]<<8) +($url[$k+6]<<16) +($url[$k+7]<<24));
        $c += ($url[$k+8] +($url[$k+9]<<8) +($url[$k+10]<<16)+($url[$k+11]<<24));
        $mix = mix($a,$b,$c);
        $a = $mix[0]; $b = $mix[1]; $c = $mix[2];
        $k += 12;
        $len -= 12;
    }

    $c += $length;
    switch($len)              /* all the case statements fall through */
    {
        case 11: $c+=($url[$k+10]<<24);
        case 10: $c+=($url[$k+9]<<16);
        case 9 : $c+=($url[$k+8]<<8);
          /* the first byte of c is reserved for the length */
        case 8 : $b+=($url[$k+7]<<24);
        case 7 : $b+=($url[$k+6]<<16);
        case 6 : $b+=($url[$k+5]<<8);
        case 5 : $b+=($url[$k+4]);
        case 4 : $a+=($url[$k+3]<<24);
        case 3 : $a+=($url[$k+2]<<16);
        case 2 : $a+=($url[$k+1]<<8);
        case 1 : $a+=($url[$k+0]);
         /* case 0: nothing left to add */
    }
    $mix = mix($a,$b,$c);
    /*-------------------------------------------- report the result */
    return $mix[2];
}

//converts a string into an array of integers containing the numeric value of the char

function strord($string) {
    for($i=0;$i<strlen($string);$i++) {
        $result[$i] = ord($string{$i});
    }
    return $result;
}
// http://www.example.com/ - Checksum: 6540747202
$url = 'info:'.$_GET['url'];
$ch = GoogleCH(strord($url));
$url='info:'.urlencode($_GET['url']);
echo file_get_contents("http://www.google.com/search?client=navclient-auto&ch=6$ch&ie=UTF-8&oe=UTF-8&features=Rank&q=$url");
/* use curl send the user angent
$curl = curl_init("http://www.google.com/search?client=navclient-auto&ch=6$ch&ie=UTF-8&oe=UTF-8&features=Rank&q=$url");
curl_setopt ($curl, CURLOPT_USERAGENT, "Mozilla/4.0 (compatible; GoogleToolbar 2.0.110-big; Windows 2000 5.0)");
curl_exec($curl);
*/
?>
</></>