快捷導(dǎo)航

Google PR查詢接口checksum新算法

更新時(shí)間：2009年01月10日 21:45:43 作者：

前些日子一篇N久之前的老文忽然成了被閱讀的熱點(diǎn)，檢查之后才發(fā)現(xiàn)自己使用那段代碼來(lái)做pr查詢的頁(yè)面已經(jīng)不能正常得到URL的Page Rank值了

Google本身提供了查詢指定的url的PageRank值的接口，知道了這個(gè)接口，就可以很容易編寫(xiě)腳本在頁(yè)面上實(shí)現(xiàn)這一功能，而無(wú)需再依賴google toolbar才能進(jìn)行查詢。本文提供了一個(gè)用PHP實(shí)現(xiàn)的pr查詢接口。同時(shí)修正了大部分版本中存在的 Linux 操作系統(tǒng)及64位操作系統(tǒng)下無(wú)法正常生成checksum的問(wèn)題。

使用很簡(jiǎn)單，只要在需要的地方

<?php
include_once("./pr.inc.php");
echo getPR($urlToQuery);
?>

即可顯示出指定url的PageRank的數(shù)值。知道了這個(gè)數(shù)值再在其基礎(chǔ)上模擬出Google Toolbar上面的圖形化的pr顯示也就不是難事了。實(shí)際上實(shí)現(xiàn)原理說(shuō)白了也很簡(jiǎn)單，就是傳遞特定的查詢參數(shù)到Google的搜索引擎，然后抓取返回的頁(yè)面內(nèi)容。

演示頁(yè)面請(qǐng)參見(jiàn) : Google PageRank Query

本文代碼素材來(lái)源: http://forums.seochat.com/archive/t-17286/Php-Pagerank-checker ；NewGCH方法實(shí)現(xiàn)參考于Firefox的工具欄擴(kuò)展插件SearchStatus的相關(guān)代碼實(shí)現(xiàn)。

網(wǎng)上還有一個(gè)開(kāi)源的pr狀態(tài)查詢的項(xiàng)目: http://pagerankstatus.mozdev.org/source.html , 可以從cvs上直接抓取完整的源代碼(cvs用戶密碼 guest):

cvs -d :pserver:guest@mozdev.org:/cvs login
cvs -d :pserver:guest@mozdev.org:/cvs co pagerankstatus

一個(gè)專(zhuān)門(mén)提供pr顯示接口的網(wǎng)站： http://www.prchecker.info/

------------------------------------------------------------------

pr.inc.php源文件如下（Updated 2008-05-04 14:29 -- Google修改了checksum的計(jì)算算法，需要在原有的GCH方法之后再套一層NewGCH方法來(lái)得到正確的checksum，同時(shí)引發(fā)的php int overflow及64位機(jī)器兼容性問(wèn)題請(qǐng)參照以下源代碼的變化部分）：

復(fù)制代碼代碼如下:

<?php 
// url get method macro. 
define('G_PR_GET_TYPE_FILE', 1); // use fopen() function 
define('G_PR_GET_TYPE_SOCKET', 2); // use standard fsocketopen function 
// main function to be called 
function getPR($_url,$gettype=G_PR_GET_TYPE_SOCKET){ 
$url = 'info:'.$_url; 
$ch = GCH(strord($url)); 
$ch = NewGCH($ch); 
$url=str_replace("_","%5F",'info:'.urlencode($_url)); 
$googlePRUrl = 
"http://toolbarqueries.google.com/search?client=navclient-auto&ch=6" 
.$ch."&ie=UTF-8&oe=UTF-8&features=Rank&q=".$url; 
$pr_str = retrieveURLContent($googlePRUrl,$gettype); 
return substr($pr_str,strrpos($pr_str, ":")+1); 
} 
//unsigned shift right 
function zeroFill($a, $b){ 
$z = hexdec('8'.implode('',array_fill(0,PHP_INT_SIZE*2-1,'0'))); 
if ($z & $a){ 
$a = ($a>>1); 
$a &= (~$z); 
$a |= hexdec('4'.implode('',array_fill(0,PHP_INT_SIZE*2-1,'0'))); 
$a = ($a>>($b-1)); 
} 
else{ 
$a = ($a>>$b); 
} 
return $a; 
} 
// discard bits beyonds 32 bit. 
function trunkbitForce32bit($n){ 
if(PHP_INT_SIZE <= 4){ 
settype($n,'float'); 
if ( $n < 0 ) $n += 4294967296; 
return $n; 
} 
else{ 
$clearbit = ''; 
for($i=0;$i<PHP_INT_SIZE-4;$i++){ 
$clearbit .= '00'; 
} 
for($i=0;$i<4;$i++){ 
$clearbit .= 'ff'; 
} 
return ($n & hexdec($clearbit)); 
} 
} 
function bigxor($m,$n){ 
//if(function_exists('gmp_init')){ 
// return floatval(gmp_strval(gmp_xor($m,$n))); 
//} 
//else{ 
return $m ^ $n; 
//} 
} 
function mix($a,$b,$c){ 
$a = trunkbitForce32bit($a); 
$b = trunkbitForce32bit($b); 
$c = trunkbitForce32bit($c); 
$a -= $b; $a = trunkbitForce32bit($a); 
$a -= $c; $a = trunkbitForce32bit($a); 
$a = bigxor($a,(zeroFill($c,13))); $a = trunkbitForce32bit($a); 
$b -= $c; $b = trunkbitForce32bit($b); 
$b -= $a; $b = trunkbitForce32bit($b); 
$b = bigxor($b,trunkbitForce32bit($a<<8)); $b = trunkbitForce32bit($b); 
$c -= $a; $c = trunkbitForce32bit($c); 
$c -= $b; $c = trunkbitForce32bit($c); 
$c = bigxor($c,(zeroFill($b,13))); $c = trunkbitForce32bit($c); 
$a -= $b;$a = trunkbitForce32bit($a); 
$a -= $c;$a = trunkbitForce32bit($a); 
$a = bigxor($a,(zeroFill($c,12)));$a = trunkbitForce32bit($a); 
$b -= $c;$b = trunkbitForce32bit($b); 
$b -= $a;$b = trunkbitForce32bit($b); 
$b = bigxor($b,trunkbitForce32bit($a<<16)); 
$c -= $a; $c = trunkbitForce32bit($c); 
$c -= $b; $c = trunkbitForce32bit($c); 
$c = bigxor($c,(zeroFill($b,5))); $c = trunkbitForce32bit($c); 
$a -= $b;$a = trunkbitForce32bit($a); 
$a -= $c;$a = trunkbitForce32bit($a); 
$a = bigxor($a,(zeroFill($c,3)));$a = trunkbitForce32bit($a); 
$b -= $c;$b = trunkbitForce32bit($b); 
$b -= $a;$b = trunkbitForce32bit($b); 
$b = bigxor($b,trunkbitForce32bit($a<<10)); 
$c -= $a; $c = trunkbitForce32bit($c); 
$c -= $b; $c = trunkbitForce32bit($c); 
$c = bigxor($c,(zeroFill($b,15))); $c = trunkbitForce32bit($c); 
return array($a,$b,$c); 
} 
function NewGCH($ch){ 
$ch = ( trunkbitForce32bit( ( $ch / 7 ) << 2 ) | 
( ( myfmod( $ch,13 ) ) & 7 ) ); 
$prbuf = array(); 
$prbuf[0] = $ch; 
for( $i = 1; $i < 20; $i++ ) 
{ 
$prbuf[$i] = $prbuf[$i-1] - 9; 
} 
$ch = GCH( c32to8bit( $prbuf ) ); 
return $ch; 
} 
function myfmod($x,$y){ 
$i = floor( $x / $y ); 
return ( $x - $i * $y ); 
} 
function c32to8bit($arr32){ 
$arr8 = array(); 
for( $i = 0; $i < count($arr32); $i++ ) { 
for( $bitOrder = $i * 4; 
$bitOrder <= $i * 4 + 3; $bitOrder++ ) { 
$arr8[$bitOrder] = $arr32[$i] & 255; 
$arr32[$i] = zeroFill( $arr32[$i], 8 ); 
} 
} 
return $arr8; 
} 
function GCH($url, $length=null){ 
if(is_null($length)) { 
$length = sizeof($url); 
} 
$init = 0xE6359A60; 
$a = 0x9E3779B9; 
$b = 0x9E3779B9; 
$c = 0xE6359A60; 
$k = 0; 
$len = $length; 
$mixo = array(); 
while( $len >= 12 ){ 
$a += ($url[$k+0] +trunkbitForce32bit($url[$k+1]<<8) 
+trunkbitForce32bit($url[$k+2]<<16) 
+trunkbitForce32bit($url[$k+3]<<24)); 
$b += ($url[$k+4] +trunkbitForce32bit($url[$k+5]<<8) 
+trunkbitForce32bit($url[$k+6]<<16) 
+trunkbitForce32bit($url[$k+7]<<24)); 
$c += ($url[$k+8] +trunkbitForce32bit($url[$k+9]<<8) 
+trunkbitForce32bit($url[$k+10]<<16) 
+trunkbitForce32bit($url[$k+11]<<24)); 
$mixo = mix($a,$b,$c); 
$a = $mixo[0]; $b = $mixo[1]; $c = $mixo[2]; 
$k += 12; 
$len -= 12; 
} 
$c += $length; 
switch( $len ) { 
case 11: 
$c += trunkbitForce32bit($url[$k+10]<<24); 
case 10: 
$c+=trunkbitForce32bit($url[$k+9]<<16); 
case 9 : 
$c+=trunkbitForce32bit($url[$k+8]<<8); 
case 8 : 
$b+=trunkbitForce32bit($url[$k+7]<<24); 
case 7 : 
$b+=trunkbitForce32bit($url[$k+6]<<16); 
case 6 : 
$b+=trunkbitForce32bit($url[$k+5]<<8); 
case 5 : 
$b+=trunkbitForce32bit($url[$k+4]); 
case 4 : 
$a+=trunkbitForce32bit($url[$k+3]<<24); 
case 3 : 
$a+=trunkbitForce32bit($url[$k+2]<<16); 
case 2 : 
$a+=trunkbitForce32bit($url[$k+1]<<8); 
case 1 : 
$a+=trunkbitForce32bit($url[$k+0]); 
} 
$mixo = mix( $a, $b, $c ); 
$mixo[2] = trunkbitForce32bit($mixo[2]); 
if( $mixo[2] < 0 ){ 
return ( 
hexdec('1'. 
implode('', 
array_fill(0,PHP_INT_SIZE*2,'0'))) 
+ $mixo[2] ); 
} 
else{ 
return $mixo[2]; 
} 
} 
// converts a string into an array of integers 
// containing the numeric value of the char 
function strord($string){ 
for($i=0;$i<strlen($string);$i++){ 
$result[$i] = ord($string{$i}); 
} 
return $result; 
} 
// return url page content or false if failed. 
function retrieveURLContent($url,$gettype){ 
switch($gettype){ 
case G_PR_GET_TYPE_FILE: 
return retrieveURLContentByFile($url); 
break; 
default: 
return retrieveURLContentBySocket($url); 
break; 
} 
} 
function retrieveURLContentByFile($url){ 
$fd = @fopen($url,"r"); 
if(!$fd){ 
return false; 
} 
$result = ""; 
while($buffer = fgets($fd, 4096)) { 
$result .= $buffer; 
} 
fclose($fd); 
return $result; 
} 
function retrieveURLContentBySocket($url, 
$host="", 
$port=80, 
$timeout=30){ 
if($host == ""){ 
if(!($pos = strpos($url,'://'))){ 
return false; 
} 
$host = substr( $url, 
$pos+3, 
strpos($url,'/',$pos+3) - $pos - 3); 
$uri = substr($url,strpos($url,'/',$pos+3)); 
} 
else{ 
$uri = $url; 
} 
$request = "GET ".$uri." HTTP/1.0\r\n" 
."Host: ".$host."\r\n" 
."Accept: */*\r\n" 
."User-Agent: ZealGet\r\n" 
."\r\n"; 
$sHnd = @fsockopen ($host, $port, $errno, $errstr, $timeout); 
if(!$sHnd){ 
return false; 
} 
@fputs ($sHnd, $request); 
// Get source 
$result = ""; 
while (!feof($sHnd)){ 
$result .= fgets($sHnd,4096); 
} 
fclose($sHnd); 
$headerend = strpos($result,"\r\n\r\n"); 
if (is_bool($headerend)) 
{ 
return $result; 
} 
else{ 
return substr($result,$headerend+4); 
} 
} 