实现代码:
package com.git.edismax;
import java.io.IOException;
import org.apache.solr.client.solrj.SolrClient;
import org.apache.solr.client.solrj.SolrQuery;
import org.apache.solr.client.solrj.SolrServerException;
import org.apache.solr.client.solrj.SolrQuery.ORDER;
import org.apache.solr.client.solrj.impl.HttpSolrClient;
import org.apache.solr.client.solrj.response.QueryResponse;
import org.apache.solr.common.SolrDocument;
import org.apache.solr.common.SolrDocumentList;
import org.apache.solr.common.params.CommonParams;
/**
* 测试edismax的代码实现自定义权重排序股则
* @author songqinghu
*
*/
public class ScoreByEdismax {
private final static String baseURL = "http://localhost:8983/solr/dismax";
public static void main(String[] args) throws Exception{
scoreBySum() ;
System.out.println("=====分割线==========");
scoreByProportion();
}
/**
*
* @描述:按照各个数值的比重来进行权重计算 count 10% point 45% hot 45% --假设总分为100分 那么 分别最大分为 10 45 45
* @return void
* @exception
* @createTime:2016年4月18日
* @author: songqinghu
* @throws IOException
* @throws SolrServerException
*/
public static void scoreByProportion() throws SolrServerException, IOException{
//先查询出来三个字段中每个字段的最大值--编写好计算权重的公式---实际项目中应该还要加入是否存在的判断
long countMax = getMaxForField("count");
long pointMax = getMaxForField("point");
long hotMax = getMaxForField("hot");
String scoreMethod = "sum(product(div(count,"+countMax+"),10),product(div(point,"+pointMax+"),45),product(div(hot,"+hotMax+"),45))^1000000000";
SolrQuery query = new SolrQuery();
query.set(CommonParams.Q, "国美*");
query.set(CommonParams.FL,"id","name","count","point","hot","score");
query.set("defType","edismax");
query.set("bf", scoreMethod);
QueryResponse response = getClient().query(query);
resultShow(response);
}
/**
* @描述:XXXXXXX
* @return
* @return long
* @exception
* @createTime:2016年4月18日
* @author: songqinghu
* @throws IOException
* @throws SolrServerException
*/
private static long getMaxForField(String fieldName) throws SolrServerException, IOException{
SolrQuery query = new SolrQuery();
query.set(CommonParams.Q, "*:*");
query.set(CommonParams.FL, fieldName);
query.setSort(fieldName, ORDER.desc);
query.setRows(1);
QueryResponse countResponse = getClient().query(query);
SolrDocument maxCount = countResponse.getResults().get(0);
long result = (long) maxCount.getFieldValue(fieldName);
System.out.println(fieldName + ":" + result);
return result;
}
/**
*
* @描述:按照 count point 和 hot 的 和的数量来进行权重计算
* @return void
* @exception
* @createTime:2016年4月18日
* @author: songqinghu
* @throws IOException
* @throws SolrServerException
*/
public static void scoreBySum() throws SolrServerException, IOException{
SolrQuery query = new SolrQuery();
query.set(CommonParams.Q, "国美*");
query.set(CommonParams.FL,"id","name","count","point","hot","score");
//开启edismax方式来进行自定义权重算法
query.set("defType", "edismax");
query.set("bf","sum(count,point,hot)^1000000000");
QueryResponse response = getClient().query(query);
resultShow(response);
}
/**
*
* @描述:查询结果显示类
* @param response
* @return void
* @exception
* @createTime:2016年4月18日
* @author: songqinghu
*/
private static void resultShow(QueryResponse response){
int time = response.getQTime();
System.out.println("响应时间:"+ time+"ms");
SolrDocumentList results = response.getResults();
long numFound = results.getNumFound();
System.out.println("总数量:"+numFound);
for (SolrDocument doc : results) {
System.out.println("id:"+ doc.getFieldValue("id").toString());
System.out.println("name:"+ doc.getFieldValue("name").toString());
System.out.println("count:"+ doc.getFieldValue("count").toString());
System.out.println("point:"+ doc.getFieldValue("point").toString());
System.out.println("hot:"+ doc.getFieldValue("hot").toString());
System.out.println("score:"+ doc.getFieldValue("score").toString());
System.out.println();
}
}
/**
*
* @描述:获取单机版本的连接信息
* @return
* @return SolrClient
* @exception
* @createTime:2016年4月18日
* @author: songqinghu
*/
public static SolrClient getClient(){
SolrClient solrClient = new HttpSolrClient(baseURL);
return solrClient;
}
}
执行结果: 响应时间:1ms 总数量:6 id:6 name:国美手机 count:2314 point:1123 hot:717 score:4154.0
id:1 name:国美集团 count:5 point:4 hot:3 score:12.0
id:5 name:国美美信 count:1 point:5 hot:4 score:10.0
id:2 name:国美电器 count:4 point:3 hot:2 score:9.0
id:4 name:国美金融 count:2 point:1 hot:5 score:8.0
id:3 name:国美在线 count:3 point:2 hot:1 score:6.0
=====分割线========== count:2314 point:1123 hot:717 响应时间:2ms 总数量:6 id:6 name:国美手机 count:2314 point:1123 hot:717 score:100.0
id:5 name:国美美信 count:1 point:5 hot:4 score:0.45572376
id:1 name:国美集团 count:5 point:4 hot:3 score:0.3701771
id:4 name:国美金融 count:2 point:1 hot:5 score:0.36252183
id:2 name:国美电器 count:4 point:3 hot:2 score:0.26302284
id:3 name:国美在线 count:3 point:2 hot:1 score:0.15586855
权重计算分析:
查询主条件比例 | 权重公式比例 | solr计算 | 自己计算 |
^1 | ^100000000 |
0.45572376 | 0.45572373506010969850068170504362 |
^1 | ^1000000000000 | 0.45572376 | |
^0.0000000001 | ^1000000000000 |
0.45572376 |