php 操作sphinx

来源：互联网发布：行情分析软件下载编辑：程序博客网时间：2024/06/05 04:35

sphinx 安装完成之后

1、生成主数据索引

cd /usr/local/coreseek/bin./indexer ind_main

2、启动 searchd 进程,为php 程序连接准备

./searchd

接下来的事情就交给php了，在压缩包里面有个 sphinxapi.php 调用之

<?php
   header("Content-Type:text/html;charset=utf-8");

   $keywords = $_GET['keywords'];
   //分页
   $p        = $_GET['p'];

   //先在linux中启动sphinx: /usr/local/coreseek/bin/searchd

   //1.连sphinx(通过sphinx的api)
   //2.调用sphinx的方法得到文档ID，权重等信息
   //3.通过文档ID,去MySQL的p表中，查对应数据
   //4.对返回结果加工处理(关键字红色)

   include './sphinxapi.php';

   $limit = 10;
   if( $p ){
       $start = ($p-1)*$limit+1;
   }else {
       $start = 0;
   }
   $sp = new SphinxClient();
   //百度-日语包含百度，不包含日语的结果
   $sp->setServer('localhost', 9312);
   //设置权重值的范围
   $sp->setWeights(array(100,1));
   $sp->setLimits( $start , $limit );
   $sp->SetMatchMode ( SPH_MATCH_EXTENDED2 );//设置模式
   $sp->SetRankingMode ( SPH_RANK_PROXIMITY );//设置评分模式
   $sp->SetFieldWeights (array('title'=>2,'content'=>1));//设置字段的权重，如果area命中，那么权重算2
   $sp->SetSortMode (SPH_SORT_EXTENDED,'@weight DESC');//
   $result = $sp->query($keywords, 'ind_main');//第二个参数是索引名

   echo '<pre> 总共找到文档数';
   var_dump( $result['total'] );
   echo '<pre> 搜索用时';
   var_dump( $result['time'] );


   if($result['total']==0){
       die('暂无数据');
   }

   $ids = join(',', array_keys($result['matches']));
   //echo $ids;//1,2,4,5,6,7,8,12,14
   $sql = 'SELECT id,title,content, url FROM crawl_contents WHERE id IN ('. $ids . ') ORDER BY field(id,' .$ids . ')';
   //echo $sql;

   $lind = mysql_connect('localhost', 'root', 'liuhan');
   mysql_set_charset('UTF8');
   mysql_select_db('test');

   $res = mysql_query($sql);
   $posts = array();
   $weight = array();
   if($res!== false && mysql_num_rows($res)>0){
       while($row = mysql_fetch_assoc($res)){
           $posts[]        = $row;
           $id             = $row['id'];
           $weight[$id]     = $result['matches'][$id]['weight'];

       }
   }
   //var_dump( $posts );die();

   //对结果加颜色
   $options = array(
       'before_match'=>'<span style="color:red">',
       'after_match'=>'</span>',
   );


?>
<!DOCTYPE html>
<html>
<head>
   <meta charset="utf-8">
   <title></title>
</head>
<body>
   <h3>查询结果</h3>
   <table>
       <tr>
           <td>id</td>
           <td>weight</td>
           <td>职位名称</td>
           <td>职位详情</td>
           <td>链接</td>

       </tr>
       <?php foreach($posts as $item){
           $url       = $item['url'];
           unset( $item['url'] ); //去除链接，不做高亮
           $row        = $sp->buildExcerpts($item, 'ind_main', $keywords, $options);
           $row['url'] = $url;

           //var_dump($row);exit;
           //注意 $row是索引数组
       ?>
       <tr>

           <td><?php echo $item['id'];?></td>
           <td><?php echo $weight[$item['id']] ;?></td>
           <td><?php echo $row[1];?></td>
           <td><?php echo $row[2];?></td>
           <td><?php echo "<a target='_blank' href='".$row['url']."'> ".$row['url']."</a>";?></td>
       </tr>
       <?php } ?>
   </table>
</body>
</html>

0 0