coreseek的php实现例子

今天把coreseek和目前干的项目进行了整合了，前台用的是php，测试数据在mysql里面，共有11w条数据
如果采用普通的mysql查询，页面运行的时间是0.9s，索引以后搜索索引再根据索引得到的id查询数据库详细信息的页面执行时间是0.03s，速度快了30倍，我很满意，哈哈

以下是php搜索索引的核心部分代码

 <?php
require_once AN_ROOT."/inc/sphinxapi.php";
require_once AN_ROOT."/res/adodb/adodb.inc.php";
$keywords = $_REQUEST['keywords'];

$info    = getInfoCoreSeek($keywords,$_GET['page']);

function getInfoCoreSeek($keywords,$page=1) {
    $mode = SPH_MATCH_ANY;
    $host = "localhost";
    $port = 3312;
    $index = "*";
    $limit = 4;
    $ranker = SPH_RANK_PROXIMITY_BM25;

    if(!$page) $page=1;
    $page = intval($page);
    $off = ($page-1)*$limit;

    $cl = new SphinxClient ();
    $cl->SetServer ( $host, $port );
    $cl->SetConnectTimeout ( 1 );
    $cl->SetWeights ( array ( 100, 1 ) );
    $cl->SetMatchMode ( $mode );
    $cl->SetLimits ( $off, $limit, ( $limit>10000 ) ? $limit : 10000 );
    $cl->SetRankingMode ( $ranker );
    $cl->SetArrayResult ( true );
    $res    = $cl->Query ( join(" ", $keywords), $index );
    if($res) {
        $pageL    = PageQuery($res['total'],$page,4,"act=search&keywords=".$_REQUEST['keywords']);
        foreach($res['matches'] as $v) {
            $arrID[]    = $v['id'];    
        }
        $rs    = $GLOBALS['db']->GetAll("select id,title,content from info where id in(".join(',', $arrID).")");
        $GLOBALS['tpl']->assign('pageL', $pageL);
        //echo '<pre>';
        //print_r($res);
        //echo '</pre>';
        return $rs;
    }
}
?>