如何在java中使用elasticsearch进行分组

本篇文章给大家分享的是有关如何在java中使用elasticsearch进行分组，小编觉得挺实用的，因此分享给大家学习，希望大家阅读完这篇文章后可以有所收获，话不多说，跟着小编一起来看看吧。

java连接elasticsearch 进行聚合查询进行相应操作

一：对单个字段进行分组求和

1、表结构图片：

根据任务id分组，分别统计出每个任务id下有多少个文字标题

1.SQL：select id, count(*) as sum from task group by taskid;

java ES连接工具类

publicclassESClientConnectionUtil{
publicstaticTransportClientclient=null;
publicfinalstaticStringHOST="192.168.200.211";//服务器部署
publicfinalstaticIntegerPORT=9301;//端口

publicstaticTransportClientgetESClient(){
System.setProperty("es.set.netty.runtime.available.processors","false");
if(client==null){
synchronized(ESClientConnectionUtil.class){
try{
//设置集群名称
Settingssettings=Settings.builder().put("cluster.name","es5").put("client.transport.sniff",true).build();
//创建client
client=newPreBuiltTransportClient(settings).addTransportAddress(newInetSocketTransportAddress(InetAddress.getByName(HOST),PORT));
}catch(Exceptionex){
ex.printStackTrace();

System.out.println(ex.getMessage());
}
}
}
returnclient;
}
publicstaticTransportClientgetESClientConnection(){
if(client==null){
System.setProperty("es.set.netty.runtime.available.processors","false");
try{
//设置集群名称
Settingssettings=Settings.builder().put("cluster.name","es5").put("client.transport.sniff",true).build();
//创建client
client=newPreBuiltTransportClient(settings).addTransportAddress(newInetSocketTransportAddress(InetAddress.getByName(HOST),PORT));
}catch(Exceptionex){
ex.printStackTrace();
System.out.println(ex.getMessage());
}
}
returnclient;
}

//判断索引是否存在
publicstaticbooleanjudgeIndex(Stringindex){
client=getESClientConnection();
IndicesAdminClientadminClient;
//查询索引是否存在
adminClient=client.admin().indices();
IndicesExistsRequestrequest=newIndicesExistsRequest(index);
IndicesExistsResponseresponses=adminClient.exists(request).actionGet();

if(responses.isExists()){
returntrue;
}
returnfalse;
}
}

java ES语句（根据单列进行分组求和）

//根据任务id分组进行求和
SearchRequestBuildersbuilder=client.prepareSearch("hottopic").setTypes("hot");
//根据taskid进行分组统计，统计出的列别名叫sum
TermsAggregationBuildertermsBuilder=AggregationBuilders.terms("sum").field("taskid");

sbuilder.addAggregation(termsBuilder);
SearchResponseresponses=sbuilder.execute().actionGet();
//得到这个分组的数据集合
Termsterms=responses.getAggregations().get("sum");
List<BsKnowledgeInfoDTO>lists=newArrayList<>();
for(inti=0;i<terms.getBuckets().size();i++){
//statistics
Stringid=terms.getBuckets().get(i).getKey().toString();//id
Longsum=terms.getBuckets().get(i).getDocCount();//数量
System.out.println("=="+terms.getBuckets().get(i).getDocCount()+"------"+terms.getBuckets().get(i).getKey());
}
//分别打印出统计的数量和id值

根据多列进行分组求和

//根据任务id分组进行求和
SearchRequestBuildersbuilder=client.prepareSearch("hottopic").setTypes("hot");
//根据taskid进行分组统计，统计出的列别名叫sum
TermsAggregationBuildertermsBuilder=AggregationBuilders.terms("sum").field("taskid");
//根据第二个字段进行分组
TermsAggregationBuilderaAggregationBuilder2=AggregationBuilders.terms("region_count").field("birthplace");
//如果存在第三个，以此类推；
sbuilder.addAggregation(termsBuilder.subAggregation(aAggregationBuilder2));
SearchResponseresponses=sbuilder.execute().actionGet();
//得到这个分组的数据集合
Termsterms=responses.getAggregations().get("sum");
List<BsKnowledgeInfoDTO>lists=newArrayList<>();
for(inti=0;i<terms.getBuckets().size();i++){
//statistics
Stringid=terms.getBuckets().get(i).getKey().toString();//id
Longsum=terms.getBuckets().get(i).getDocCount();//数量
System.out.println("=="+terms.getBuckets().get(i).getDocCount()+"------"+terms.getBuckets().get(i).getKey());
}
//分别打印出统计的数量和id值

对多个field求max/min/sum/avg

SearchRequestBuilderrequestBuilder=client.prepareSearch("hottopic").setTypes("hot");
//根据taskid进行分组统计，统计别名为sum
TermsAggregationBuilderaggregationBuilder1=AggregationBuilders.terms("sum").field("taskid")
//根据tasktatileid进行升序排列
.order(Order.aggregation("tasktatileid",true));
//求tasktitleid进行求平均数别名为avg_title

AggregationBuilderaggregationBuilder2=AggregationBuilders.avg("avg_title").field("tasktitleid");
//
AggregationBuilderaggregationBuilder3=AggregationBuilders.sum("sum_taskid").field("taskid");
requestBuilder.addAggregation(aggregationBuilder1.subAggregation(aggregationBuilder2).subAggregation(aggregationBuilder3));
SearchResponseresponse=requestBuilder.execute().actionGet();

Termsaggregation=response.getAggregations().get("sum");
Avgterms2=null;
Sumterm3=null;
for(Terms.Bucketbucket:aggregation.getBuckets()){
terms2=bucket.getAggregations().get("avg_title");//org.elasticsearch.search.aggregations.metrics.avg.InternalAvg
term3=bucket.getAggregations().get("sum_taskid");//org.elasticsearch.search.aggregations.metrics.sum.InternalSum
System.out.println("编号="+bucket.getKey()+";平均="+terms2.getValue()+";总="+term3.getValue());
}

以上就是如何在java中使用elasticsearch进行分组，小编相信有部分知识点可能是我们日常工作会见到或用到的。希望你能通过这篇文章学到更多知识。更多详情敬请关注恰卡编程网行业资讯频道。