ES 聚合查询Java Api

分页查询，按照时间进行分组查询求平均(多平均)，按照某一字段分组求聚合等等

分页查询

        SysUser loginUser = SecurityUtils.getLoginUser().getUser();Long userId = loginUser.getUserId();LambdaQueryChainWrapper<Warning> query = warningService.lambdaQuery();if (req.getStatus()!=null && req.getStatus().size() > 0){if (req.getStatus().contains(null)){throw new BaseException("错误的'处理状态'字段信息！");}else {query.in(Warning::getProcessStatus, req.getStatus());}}//query.in(req.getStatus()!=null && req.getStatus().size() > 0,Warning::getProcessStatus, req.getStatus());query.eq(StringUtils.isNotEmpty(req.getSiteId()), Warning::getDeptId, req.getSiteId());query.like(StringUtils.isNotEmpty(req.getDevCode()), Warning::getDevCode, req.getDevCode());query.like(StringUtils.isNotEmpty(req.getDevName()), Warning::getDevName, req.getDevName());List<WarningProcess> processes = warningProcessService.getWarningProcess(userId);if (processes == null || processes.size() == 0)// return new TableDataInfo();//没有处理人的话，就按照当前登录人的站点权限过滤{List<Long> userDepts = userService.queryUserDepts(userId);query.in(Warning::getDeptId, userDepts);}else{List<Long> warnIds = processes.stream().map(m -> m.getWarnId()).collect(Collectors.toList());query.in(CollectionUtils.isNotEmpty(warnIds), Warning::getId, warnIds);}query.orderByDesc(Warning::getWarnTime);TableDataInfo data = warningService.page(req, query);List<SysDept> depts = sysDeptService.list();data.getRows().stream().forEach(item -> {Warning warnItem = (Warning) item;SysDept dept = depts.stream().filter(m -> m.getDeptId().equals(warnItem.getDeptId())).findFirst().orElse(null);if (null != dept) {warnItem.setDeptName(dept.getDeptName());}});return data;

mybatisPlus 转换成es查询的JavaApi

SearchSourceBuilder searchSourceBuilder = new SearchSourceBuilder()//从哪到哪.from((req.getPageNum() - 1) * req.getPageSize())//数量.size(req.getPageSize())//按照warn_time 倒序排序.sort("warn_time", SortOrder.DESC).trackTotalHits(true);BoolQueryBuilder boolQueryBuilder = QueryBuilders.boolQuery();//相当于sql 的 in 操作if (req.getStatus()!=null && req.getStatus().size() > 0){boolQueryBuilder.filter(QueryBuilders.termsQuery("process_status",req.getStatus()));}//term 精确匹配if (StringUtils.isNotEmpty(req.getSiteId())){boolQueryBuilder.filter(QueryBuilders.termQuery("dept_id",req.getSiteId()));}//模糊查询if (StringUtils.isNotEmpty(req.getDevCode())){boolQueryBuilder.filter(QueryBuilders.wildcardQuery("dev_code.keyword","*"+req.getDevCode()+"*"));}if (StringUtils.isNotEmpty(req.getDevName())){boolQueryBuilder.filter(QueryBuilders.wildcardQuery("dev_name.keyword","*"+req.getDevName()+"*"));}List<WarningProcess> processes = warningProcessService.getWarningProcess(userId);if (processes == null || processes.size() == 0)// return new TableDataInfo();//没有处理人的话，就按照当前登录人的站点权限过滤{List<Long> userDepts = userService.queryUserDepts(userId);boolQueryBuilder.filter(QueryBuilders.termsQuery("dept_id.keyword",userDepts));}else{List<Long> warnIds = processes.stream().map(m -> m.getWarnId()).collect(Collectors.toList());if (CollectionUtils.isNotEmpty(warnIds)){//ES in操作 termsQueryboolQueryBuilder.filter(QueryBuilders.termsQuery("id",warnIds));}}//进行查询searchSourceBuilder.query(boolQueryBuilder);//返回结果SearchResponse searchResponse = EsClintUtil.execSearch(restHighLevelClient,searchSourceBuilder, index);//返回总数long total = searchResponse.getHits().getTotalHits();List<Warning> list = new ArrayList<>();final SearchHit[] searchHits = searchResponse.getHits().getHits();for (SearchHit searchHit : searchHits) {list.add(JSON.parseObject(searchHit.getSourceAsString(), Warning.class));}//返回自定义表格信息TableDataInfo info = new TableDataInfo(list, (int) total);

桶聚合

SELECT
dept_name,
count(d1.id) as imgcount,
count(distinct d1.device_code) as imgcount1,
sum(case d1.`status` when '1' then 1 ELSE 0 END) as warningCount
FROM   sys_dept s1
LEFT JOIN dev_shock_data d1 ON s1.dept_id = d1.dept_id
where s1.dept_id!=100
GROUP BY s1.dept_name

数据库的sql对应的Java代码

SearchSourceBuilder searchSourceBuilder = new SearchSourceBuilder();BoolQueryBuilder boolQueryBuilder = QueryBuilders.boolQuery();boolQueryBuilder.mustNot(QueryBuilders.termQuery("dept_id", 100));searchSourceBuilder.query(boolQueryBuilder);searchSourceBuilder.size(0);//不输出原始数据//按照站点分组TermsAggregationBuilder aggregationBuilder = AggregationBuilders.terms("deptgroup").field("dept_id")  //桶分组.subAggregation(AggregationBuilders.terms("statusgroup").field("status.keyword"))//设备个数去重数量.subAggregation(AggregationBuilders.cardinality("imgcount1").field("device_code.keyword")).size(5000);searchSourceBuilder.aggregation(aggregationBuilder);SearchResponse searchResponse = EsClintUtil.execSearch(restHighLevelClient,searchSourceBuilder, "dev_shock_data");// 遍历封装列表对象List<ImageStatisticRes> resList=new ArrayList<>();Terms parsedStringTerms =searchResponse.getAggregations().get("deptgroup");List<? extends Terms.Bucket> buckets = parsedStringTerms.getBuckets();Map<String,String> deptMap= EsClintUtil.getDepts(restHighLevelClient);List<String> list = new ArrayList<>();for (Terms.Bucket bucket : buckets) {//key的数据String key = bucket.getKey().toString();long docCount = bucket.getDocCount();//获取数据ImageStatisticRes res = new ImageStatisticRes();res.setSiteName(deptMap.get(key));res.setImageCount((int) docCount);//获取设备个数Cardinality imagecount1 = bucket.getAggregations().get("imgcount1");res.setImageCount1((int) imagecount1.getValue());Terms statusgroupTearms = bucket.getAggregations().get("statusgroup");for (Terms.Bucket statusBucket : statusgroupTearms.getBuckets()) {if (statusBucket.getKey().equals("1")) {res.setImageWarningCount((int) statusBucket.getDocCount());}}resList.add(res);list.add(key);}//获取到总的站台数据final List<String> deptIds = deptMap.keySet().stream().collect(Collectors.toList());//获取差集final List<String> collect = deptIds.stream().filter(item -> !list.contains(item)).collect(Collectors.toList());//遍历得到的差集就是不存在站台数据的for (String s : collect) {ImageStatisticRes res = new ImageStatisticRes();res.setSiteName(deptMap.get(s));res.setImageCount(0);res.setImageWarningCount(0);resList.add(res);}return resList;

时间聚合分组

SELECT HOUR (create_time) khour ,avg(angle_x) averageX,  avg(angle_y) averageY, avg(angle_z) averageZ  from dev_shock_datawhere device_id=#{deviceId} and create_time &gt;#{startTime} and  create_time &lt;#{endTime}group by HOUR (create_time)order by khour

数据库的sql转换成es对应的JavaApi

final ArrayList<GroupByHour> groupByHours = new ArrayList<>();SearchSourceBuilder searchSourceBuilder = new SearchSourceBuilder().sort("create_time", SortOrder.ASC).trackTotalHits(true);BoolQueryBuilder boolQueryBuilder = QueryBuilders.boolQuery();boolQueryBuilder.must(QueryBuilders.termQuery("device_id",deviceId));boolQueryBuilder.filter(QueryBuilders.rangeQuery("create_time").gt(startTime).lt(endTime));searchSourceBuilder.size(0);searchSourceBuilder.query(boolQueryBuilder);AvgAggregationBuilder  averageX = AggregationBuilders.avg("averageX").field(x);AvgAggregationBuilder  averageY = AggregationBuilders.avg("averageY").field(y);AvgAggregationBuilder  averageZ = AggregationBuilders.avg("averageZ").field(z);AggregationBuilder  groupTime = AggregationBuilders.dateHistogram("khour").field("create_time") // 根据date字段值，对数据进行分组// 时间分组间隔：DateHistogramInterval.* 序列常量，支持每月，每年，每天等等时间间隔.dateHistogramInterval(DateHistogramInterval.HOUR)// 设置返回结果中桶key的时间格式.format("HH")
//              .timeZone(DateTimeZone.forTimeZone(TimeZone.getTimeZone("GMT%2B8")))
//              .timeZone(DateTimeZone.forTimeZone(TimeZone.getTimeZone("GMT+8:00"))).minDocCount(0L).subAggregation(averageX).subAggregation(averageY).subAggregation(averageZ);searchSourceBuilder.aggregation(groupTime);final SearchResponse dev_shock_data = EsClintUtil.execSearch(restHighLevelClient, searchSourceBuilder, "dev_shock_data");ParsedDateHistogram khour = dev_shock_data.getAggregations().get("khour");final List<? extends Histogram.Bucket> buckets1 = khour.getBuckets();for (Histogram.Bucket bucket : buckets1) {final GroupByHour groupByHour = new GroupByHour();if(bucket.getDocCount()>0){groupByHour.setHour(Integer.parseInt(bucket.getKeyAsString()));Avg xAvg = bucket.getAggregations().get("averageX");Avg yAvg = bucket.getAggregations().get("averageY");Avg zAvg = bucket.getAggregations().get("averageZ");groupByHour.setAverageX(xAvg.getValue());groupByHour.setAverageY(yAvg.getValue());groupByHour.setAverageZ(zAvg.getValue());groupByHours.add(groupByHour);}}return groupByHours;

欢迎大家前来提需求！

ElasticSearch 聚合查询 JavaApi相关推荐

Elasticsearch聚合查询案例分享
为什么80%的码农都做不了架构师?>>> Elasticsearch聚合查询案例分享 1.案例介绍本文包含三个案例: 案例1:统计特定时间范围内每个应用的总访问量.访问成功数 ...
php聚合查询,php elasticsearch 聚合查询(Aggregation)
Elasticsearch中的聚合查询,类似SQL的SUM/AVG/COUNT/GROUP BY分组查询,主要用于统计分析场景. 这里主要介绍PHP Elasticsearch 聚合查询的写法,如果不 ...
es7.1聚合查询 javaAPI
es7.1聚合查询 javaAPI(一) 文章目录 es7.1聚合查询 javaAPI(一) 1. 查询符合条件的数据量 2. 查询某字段去重后的数据量 3. 单字段分组聚合 4. 多字段聚合链接: ...
Elasticsearch聚合查询多字段设置权重
Elasticsearch聚合查询多字段设置权重背景环境说明 script设置权重小结背景实际应用中,可能会需要为为doc文档中某个字段的某些特定的值设置权重,影响排序.es提供了比较灵活的 ...
ElasticSearch聚合查询返回结果buckets取值
ElasticSearch聚合查询返回结果buckets取值 1.聚合查询如下: {"size":0,"query":{"bool":{&q ...
ElasticSearch聚合查询Restful语法和JavaApi详解（基于ES7.6）
本文收录于github和gitee ,里面有我完整的Java系列文章,学习或面试都可以看看 (一)概述在前面关于ES的一系列文章中,已经介绍了ES的概念.常用操作.JavaAPI以及实际的一个小de ...
es java api 查询_es聚合查询javaapi
查询节气详情 package net.apishop.www.controller; import import import import import import import import i ...
ES/ElasticSearch 聚合查询时报错：too_many_buckets_exception
环境:ElasticSearch6.7 问题描述: {"error": {"root_cause": [],"type": "se ...
（转）Elasticsearch 聚合查询、桶聚合、管道聚合及复合查询
转自: https://blog.csdn.net/zx711166/article/details/81906881 聚合查询聚合是一种基于查询条件对数据进行分桶.计算的方法. 聚合可以嵌套,由此 ...

ElasticSearch 聚合查询 JavaApi

ES 聚合查询Java Api

欢迎大家前来提需求！

ElasticSearch 聚合查询 JavaApi相关推荐

最新文章

热门文章