1.API聚合方法

@Slf4j
public class EsAggsTest {public static void testAggs() throws IOException {SearchSourceBuilder searchSourceBuilder = new SearchSourceBuilder();/*** 使用tag字段进行桶分组* 使用sum、avg进行指标聚合*/TermsAggregationBuilder aggregationBuilder = AggregationBuilders.terms("tag_tr").field("tag").                   //桶分组subAggregation(AggregationBuilders.sum("sum_id").field("id")).  //求和subAggregation(AggregationBuilders.avg("avg_id").field("id")); //求平均值searchSourceBuilder.aggregation(aggregationBuilder);/*** 不输出原始数据*/searchSourceBuilder.size(0);/*** 打印dsl语句*/log.info("dsl:" + searchSourceBuilder.toString());/*** 设置索引以及填充语句*/SearchRequest searchRequest = new SearchRequest();searchRequest.indices("test_demo");searchRequest.source(searchSourceBuilder);SearchResponse response = EsUtil.getRestHighLevelClient().search(searchRequest, RequestOptions.DEFAULT);/*** 解析数据，获取tag_tr的指标聚合参数。*/Aggregations aggregations = response.getAggregations();ParsedStringTerms parsedStringTerms = aggregations.get("tag_tr");List<? extends Terms.Bucket> buckets = parsedStringTerms.getBuckets();for (Terms.Bucket bucket : buckets) {//key的数据String key = bucket.getKey().toString();long docCount = bucket.getDocCount();//获取数据Aggregations bucketAggregations = bucket.getAggregations();ParsedSum sumId = bucketAggregations.get("sum_id");ParsedAvg avgId = bucketAggregations.get("avg_id");System.out.println(key + ":" + docCount + "-" + sumId.getValue() + "-" + avgId.getValue());}}public static void main(String[] args) throws IOException {testAggs();}
}

输出语句：

09:49:54.950 [main] INFO com.tellme.Test.es.EsAggsTest - dsl:{"size":0,"aggregations":{"tag_tr":{"terms":{"field":"tag","size":10,"min_doc_count":1,"shard_min_doc_count":0,"show_term_doc_count_error":false,"order":[{"_count":"desc"},{"_key":"asc"}]},"aggregations":{"sum_id":{"sum":{"field":"id"}},"avg_id":{"avg":{"field":"id"}}}}}}
疫情:5-24067.0-4813.4
军事:2-2010.0-1005.0
教育:2-2008.0-1004.0
婚姻:1-9001.0-9001.0

dsl语句：

{"size":0,"aggregations":{"tag_tr":{"terms":{"field":"tag","size":10,"min_doc_count":1,"shard_min_doc_count":0,"show_term_doc_count_error":false,"order":[{"_count":"desc"},{"_key":"asc"}]},"aggregations":{"sum_id":{"sum":{"field":"id"}},"avg_id":{"avg":{"field":"id"}}}}}
}

即先对tag字段进行分组聚合，而后对各组的数据进行指标聚合。

3. script聚合方式

    /*** 脚本聚合*/public static void testScript() throws IOException {Script script = new Script("doc['tag']+'-'+doc['publishTime']");
//        CardinalityAggregationBuilder aggregationBuilder = AggregationBuilders.cardinality("user").script(script);TermsAggregationBuilder aggregationBuilder = AggregationBuilders.terms("tag_tr").script(script);SearchSourceBuilder searchSourceBuilder = new SearchSourceBuilder();searchSourceBuilder.aggregation(aggregationBuilder);//不输出元素数据searchSourceBuilder.size(0);SearchRequest searchRequest=new SearchRequest();searchRequest.indices("test_demo");searchRequest.source(searchSourceBuilder);log.info("dsl:" + searchSourceBuilder.toString());SearchResponse response = EsUtil.getRestHighLevelClient().search(searchRequest, RequestOptions.DEFAULT);Aggregations aggregations = response.getAggregations();//获取配置ParsedStringTerms parsedStringTerms = aggregations.get("tag_tr");List<? extends Terms.Bucket> buckets = parsedStringTerms.getBuckets();for (Terms.Bucket bucket : buckets) {//key的数据String key = bucket.getKey().toString();long docCount = bucket.getDocCount();//获取数据System.out.println(key + ":" + docCount);}}

转换为dsl：

{"size":0,"aggregations":{"tag_tr":{"terms":{"script":{"source":"doc['tag']+'-'+doc['publishTime']","lang":"painless"},"size":10,"min_doc_count":1,"shard_min_doc_count":0,"show_term_doc_count_error":false,"order":[{"_count":"desc"},{"_key":"asc"}]}}}
}

输出结果：

[军事]-[2021-05-13T17:33:14.000Z]:2
[教育]-[2021-05-13T17:33:14.000Z]:2
[疫情]-[2021-05-13T17:33:14.000Z]:2
[婚姻]-[2021-05-17T15:15:07.000Z]:1
[疫情]-[2021-05-14T11:40:06.000Z]:1
[疫情]-[2021-05-14T17:44:37.000Z]:1
[疫情]-[2021-05-14T17:50:42.000Z]:1

Es7.x使用RestHighLevelClient进行聚合操作相关推荐

2021年大数据常用语言Scala（二十七）：函数式编程聚合操作
目录聚合操作聚合 reduce 定义案例折叠 fold 定义案例聚合操作聚合操作,可以将一个列表中的数据合并为一个.这种操作经常用来统计分析中聚合 reduce reduce表示 ...
mongodb的聚合操作
mongodb的聚合操作 1 mongodb的聚合是什么聚合(aggregate)是基于数据处理的聚合管道,每个文档通过一个由多个阶段(stage)组成的管道,可以对每个阶段的管道进行分组.过滤等功 ...
python和R对dataframe的分组聚合操作：dplyr、groupby、agg、group_by、nunique、reset_index、rename、summarise、n_distinct
python和R对dataframe的分组聚合操作:dplyr.groupby.agg.group_by.nunique.reset_index.rename.summarise.n_distinct ...
elasticsearch聚合操作——本质就是针对搜索后的结果使用桶bucket（允许嵌套）进行group by，统计下分组结果，包括min/max/avg...
分析最后,我们还有一个需求需要完成:允许管理者在职员目录中进行一些分析. Elasticsearch有一个功能叫做聚合(aggregations),它允许你在数据上生成复杂的分析统计.它很像SQL中 ...
OpenStack/Gnocchi简介——时间序列数据聚合操作提前计算并存储起来，先算后取的理念...
先看下 http://www.cnblogs.com/bonelee/p/6236962.html 这里对于环形数据库的介绍,便于理解归档这个操作! 转自:http://blog.sina.com.c ...
MongoDB 聚合操作
MongoDB 聚合操作在MongoDB中,有两种方式计算聚合:Pipeline 和 MapReduce.Pipeline查询速度快于MapReduce,但是MapReduce的强大之处在于能够在多 ...
limit mongodb 聚合_MongoDB 聚合操作(转)
在MongoDB中,有两种方式计算聚合:Pipeline 和 MapReduce.Pipeline查询速度快于MapReduce,但是MapReduce的强大之处在于能够在多台Server上并行执行复 ...
Python中的Numpy(3.聚合操作)
1.ndarray的常用的聚合操作 # 1.最大值/最小值 :np.max / np.min n1 = np.random.randint(0,100,20) print(n1) # 打印出n1中的最 ...
Map端的聚合操作Combiner
需求词频统计,假设有一个文件,统计这个文件词频常见的解决办法例如原始文本是: 交给两个Map去处理,map1处理hello is hello is,map2处理hadoop is hadoop ...

Es7.x使用RestHighLevelClient进行聚合操作

1.API聚合方法

3. script聚合方式

Es7.x使用RestHighLevelClient进行聚合操作相关推荐

最新文章

热门文章