ES的聚合操作

构建数据：

​
    @Test
    public void createIndex(){
        /**
         * 创建索引
         * */
        client.admin().indices().prepareCreate("player").get();
    }
​
​
​
    /**
     * 创建映射
     */
    @Test
    public void testCreateIndexMapping_boost() throws Exception{
        /**
         * 格式：
         "mappings": {
            "player": {
                "properties": {
                     "name": {"index": "not_analyzed","type": "string"},
                    "age": {"type": "integer"},
                    "salary": {"type": "integer"},
                    "team": {"index": "not_analyzed","type": "string"},
                    "position": {"index": "not_analyzed","type": "string"}
                }
            }
         }
​
         */
        //构建json的数据格式，创建映射
        XContentBuilder mappingBuilder = XContentFactory.jsonBuilder()
                .startObject()
                .startObject("player")
                .startObject("properties")
                .startObject("name").field("type","string").field("index", "not_analyzed").endObject()
                .startObject("age").field("type","integer").endObject()
                .startObject("salary").field("type","integer").endObject()
                .startObject("team").field("type","string").field("index", "not_analyzed").endObject()
                .startObject("position").field("type","string").field("index", "not_analyzed").endObject()
                .endObject()
                .endObject()
                .endObject();
        PutMappingRequest request = Requests.putMappingRequest("player")
                .type("player")
                .source(mappingBuilder);
        client.admin().indices().putMapping(request).get();
    }
​
    @Test
    public void BulkInsertDocument() throws IOException {
        BulkRequestBuilder bulkRequest = client.prepareBulk();
​
// either use client#prepare, or use Requests# to directly build index/delete requests
        bulkRequest.add(client.prepareIndex("player", "player", "1")
                .setSource(jsonBuilder()
                        .startObject()
                        .field("name", "郭德纲")
                        .field("age", 33)
                        .field("salary",3000)
                        .field("team" , "cav")
                        .field("position" , "sf")
                        .endObject()
                )
        );
        bulkRequest.add(client.prepareIndex("player", "player", "2")
                .setSource(jsonBuilder()
                        .startObject()
                        .field("name", "于谦")
                        .field("age", 25)
                        .field("salary",2000)
                        .field("team" , "cav")
                        .field("position" , "pg")
                        .endObject()
                )
        );
        bulkRequest.add(client.prepareIndex("player", "player", "3")
                .setSource(jsonBuilder()
                        .startObject()
                        .field("name", "岳云鹏")
                        .field("age", 29)
                        .field("salary",1000)
                        .field("team" , "war")
                        .field("position" , "pg")
                        .endObject()
                )
        );
        bulkRequest.add(client.prepareIndex("player", "player", "4")
                .setSource(jsonBuilder()
                        .startObject()
                        .field("name", "孙越")
                        .field("age", 26)
                        .field("salary",2000)
                        .field("team" , "war")
                        .field("position" , "sg")
                        .endObject()
                )
        );
        bulkRequest.add(client.prepareIndex("player", "player", "5")
                .setSource(jsonBuilder()
                        .startObject()
                        .field("name", "张云雷")
                        .field("age", 26)
                        .field("salary",2000)
                        .field("team" , "war")
                        .field("position" , "pf")
                        .endObject()
                )
        );
        bulkRequest.add(client.prepareIndex("player", "player", "6")
                .setSource(jsonBuilder()
                        .startObject()
                        .field("name", "爱迪生")
                        .field("age", 40)
                        .field("salary",1000)
                        .field("team" , "tim")
                        .field("position" , "pf")
                        .endObject()
                )
        );
        bulkRequest.add(client.prepareIndex("player", "player", "7")
                .setSource(jsonBuilder()
                        .startObject()
                        .field("name", "牛顿")
                        .field("age", 21)
                        .field("salary",500)
                        .field("team" , "tim")
                        .field("position" , "c")
                        .endObject()
                )
        );
        bulkRequest.add(client.prepareIndex("player", "player", "4")
                .setSource(jsonBuilder()
                        .startObject()
                        .field("name", "爱因斯坦")
                        .field("age", 21)
                        .field("salary",300)
                        .field("team" , "tim")
                        .field("position" , "sg")
                        .endObject()
                )
        );
        bulkRequest.add(client.prepareIndex("player", "player", "8")
                .setSource(jsonBuilder()
                        .startObject()
                        .field("name", "特斯拉")
                        .field("age", 20)
                        .field("salary",500)
                        .field("team" , "tim")
                        .field("position" , "sf")
                        .endObject()
                )
        );
​
​
        BulkResponse bulkResponse = bulkRequest.get();
        if (bulkResponse.hasFailures()) {
            // process failures by iterating through each bulk response item
        }
    }
​
​

1：分组求count

计算每个球队的球员数量：

select team, count(*) as player_count from player group by team;

@Test
public void groupAndCount() {
    //1：构建查询提交
    SearchRequestBuilder builder = client.prepareSearch("player").setTypes("player");
    //2：指定聚合条件
    TermsAggregationBuilder team = AggregationBuilders.terms("player_count").field("team");
    //3:将聚合条件放入查询条件中
    builder.addAggregation(team);
    //4:执行action，返回searchResponse
    SearchResponse searchResponse = builder.get();
    //5:将查询返回的searchResponse转换成map
    Map<String, Aggregation> aggregationMap = searchResponse.getAggregations().asMap();
    //6:取出聚合的字段
    StringTerms teamTerms = (StringTerms)aggregationMap.get("player_count");
    System.out.println(teamTerms);
    //7:对聚合的字段进行迭代
    Iterator<StringTerms.Bucket> iterator = teamTerms.getBuckets().iterator();
    while (iterator.hasNext()){
        StringTerms.Bucket bucket = iterator.next();
        System.out.println("每个球队 :" + bucket.getKey() + "有 【"+bucket.getDocCount()+"】 个人");
    }
​
}
​
​

2：Group by 多个字段

计算每个球队每个位置的球员数 select team, position, count(*) as pos_count from player group by team,position;

/**
 * group by 多个字段
 * 计算每个球队每个位置的球员数
 * select team, position, count(*) as pos_count from player group by team, position;
 * */
@Test
public void groupMutilFields() {
    //1：构建查询提交
    SearchRequestBuilder builder = client.prepareSearch("player").setTypes("player");
    //2：指定聚合条件
    TermsAggregationBuilder team = AggregationBuilders.terms("player_count").field("team");
    TermsAggregationBuilder potition = AggregationBuilders.terms("position_count").field("position");
    //注意父子关系
team.subAggregation(potition);
    //3:将聚合条件放入查询条件中
    builder.addAggregation(team).addAggregation(potition);
    //4:执行action，返回searchResponse
    SearchResponse searchResponse = builder.get();
    //5:将查询返回的searchResponse转换成map
    Map<String, Aggregation> aggregationMap = searchResponse.getAggregations().asMap();
    //6:取出聚合的字段
    StringTerms teamTerms = (StringTerms)aggregationMap.get("player_count");
    //7:对聚合的字段进行迭代
    Iterator<StringTerms.Bucket> iterator = teamTerms.getBuckets().iterator();
    while (iterator.hasNext()){
        StringTerms.Bucket bucket = iterator.next();
        //8:获取所有子聚合
        Map<String, Aggregation> position_map = bucket.getAggregations().asMap();
        StringTerms potitionTerms = (StringTerms)position_map.get("position_count");
        //9：对子集合下面的内容迭代  队名--位置--球员
        Iterator<StringTerms.Bucket> sub_iterator = potitionTerms.getBuckets().iterator();
        while (sub_iterator.hasNext()){
            StringTerms.Bucket sub_bucket = sub_iterator.next();
            System.out.println("球队 :" + bucket.getKey() + "下面的 "+sub_bucket.getKey()+"的位置 有"+sub_bucket.getDocCount());
        }
    }
}
​
​

3：分组求最大

计算每个球队年龄最大/最小/总/平均的球员年龄

select team, max(age) as max_age from player group by team;

​
/**
 * 分组求：最大值、最小值、平均值
 * 计算每个球队年龄最大/最小/总/平均的球员年龄
 select team, max(age) as max_age from player group by team;
 *
 * */
@Test
public void groupMax(){
    //1：构建查询提交
    SearchRequestBuilder builder = client.prepareSearch("player").setTypes("player");
    //2：指定聚合条件
    TermsAggregationBuilder team = AggregationBuilders.terms("player_count").field("team");
    //3: 指定要查哪一个字段的最大值
    MaxAggregationBuilder ageFiled = AggregationBuilders.max("max_age").field("age");
    //: 找到聚合关系：父子关系
    team.subAggregation(ageFiled);
    //3:将聚合条件放入查询条件中
    builder.addAggregation(team);
    //4:执行action，返回searchResponse
    SearchResponse searchResponse = builder.get();
    //5:将查询返回的searchResponse转换成map
    Map<String, Aggregation> aggregationMap = searchResponse.getAggregations().asMap();
    //6:取出聚合的字段
    StringTerms teamTerms = (StringTerms)aggregationMap.get("player_count");
    //7:对聚合的字段进行迭代
    Iterator<StringTerms.Bucket> iterator = teamTerms.getBuckets().iterator();
    while (iterator.hasNext()){
        StringTerms.Bucket bucket = iterator.next();
        //8:获取所有子聚合
        Map age_map = bucket.getAggregations().asMap();
        int age = (int)((InternalMax) age_map.get("max_age")).getValue();
        System.out.println("球队 :" + bucket.getKey() + "  最大年龄： "+age);
    }
}
​
​

4：分组求最小

计算每个球队年龄最大/最小/总/平均的球员年龄 select team, min(age) as max_age from player group by team;

/**
 * 分组求：最大值、最小值、平均值
 * 计算每个球队年龄最大/最小/总/平均的球员年龄
 select team, min(age) as max_age from player group by team;
 *
 * */
@Test
public void groupMin(){
    //1：构建查询提交
    SearchRequestBuilder builder = client.prepareSearch("player").setTypes("player");
    //2：指定聚合条件
    TermsAggregationBuilder team = AggregationBuilders.terms("player_count").field("team");
    //3: 指定要查哪一个字段的最大值
    MinAggregationBuilder ageFiled = AggregationBuilders.min("min_age").field("age");
    //: 找到聚合关系：父子关系
    team.subAggregation(ageFiled);
    //3:将聚合条件放入查询条件中
    builder.addAggregation(team);
    //4:执行action，返回searchResponse
    SearchResponse searchResponse = builder.get();
    //5:将查询返回的searchResponse转换成map
    Map<String, Aggregation> aggregationMap = searchResponse.getAggregations().asMap();
    //6:取出聚合的字段
    StringTerms teamTerms = (StringTerms)aggregationMap.get("player_count");
    //7:对聚合的字段进行迭代
    Iterator<StringTerms.Bucket> iterator = teamTerms.getBuckets().iterator();
    while (iterator.hasNext()){
        StringTerms.Bucket bucket = iterator.next();
        //8:获取所有子聚合
        Map age_map = bucket.getAggregations().asMap();
        int age = (int)((InternalMin) age_map.get("max_age")).getValue();
        System.out.println("球队 :" + bucket.getKey() + "  最小年龄： "+age);
    }
}
​
​

5：分组求平均值

计算每个球队年龄最大/最小/总/平均的球员年龄 select team, min(age) as max_age from player group by team;

/**
 * 分组求：最大值、最小值、平均值
 * 计算每个球队年龄最大/最小/总/平均的球员年龄
 select team, min(age) as max_age from player group by team;
 *
 * */
@Test
public void groupAvg(){
    //1：构建查询提交
    SearchRequestBuilder builder = client.prepareSearch("player").setTypes("player");
    //2：指定聚合条件
    TermsAggregationBuilder team = AggregationBuilders.terms("player_count").field("team");
    //3: 指定要查哪一个字段的最大值
    AvgAggregationBuilder ageFiled = AggregationBuilders.avg("avg_age").field("age");
    //: 找到聚合关系：父子关系
    team.subAggregation(ageFiled);
    //3:将聚合条件放入查询条件中
    builder.addAggregation(team);
    //4:执行action，返回searchResponse
    SearchResponse searchResponse = builder.get();
    //5:将查询返回的searchResponse转换成map
    Map<String, Aggregation> aggregationMap = searchResponse.getAggregations().asMap();
    //6:取出聚合的字段
    StringTerms teamTerms = (StringTerms)aggregationMap.get("player_count");
    //7:对聚合的字段进行迭代
    Iterator<StringTerms.Bucket> iterator = teamTerms.getBuckets().iterator();
    while (iterator.hasNext()){
        StringTerms.Bucket bucket = iterator.next();
        //8:获取所有子聚合
        Map age_map = bucket.getAggregations().asMap();
        Double age = ((InternalAvg) age_map.get("avg_age")).getValue();
        System.out.println("球队 :" + bucket.getKey() + "  平均年龄： "+age);
    }
}
​
​

6：分组求和

计算每个球队球员的平均年龄，同时又要计算总年薪 select team, avg(age)as avg_age, sum(salary) as total_salary from player group by team;

/**
 * 分组求：最大值、最小值、平均值
 * 计算每个球队球员的平均年龄，同时又要计算总年薪
 select team, avg(age)as avg_age, sum(salary) as total_salary from player group by team;
 *
 * */
@Test
public void groupsum(){
    //1：构建查询提交
    SearchRequestBuilder builder = client.prepareSearch("player").setTypes("player");
    //2：指定聚合条件
    TermsAggregationBuilder team = AggregationBuilders.terms("player_count").field("team");
    //3: 指定要查哪一个字段
    AvgAggregationBuilder ageFiled = AggregationBuilders.avg("avg_age").field("age");
​
    SumAggregationBuilder salaryField= AggregationBuilders.sum("sum_salary").field("salary");
    //: 找到聚合关系：父子关系
    team.subAggregation(ageFiled).subAggregation(salaryField);
    //3:将聚合条件放入查询条件中
    builder.addAggregation(team);
    //4:执行action，返回searchResponse
    SearchResponse searchResponse = builder.get();
    //5:将查询返回的searchResponse转换成map
    Map<String, Aggregation> aggregationMap = searchResponse.getAggregations().asMap();
    //6:取出聚合的字段
    StringTerms teamTerms = (StringTerms)aggregationMap.get("player_count");
    //7:对聚合的字段进行迭代
    Iterator<StringTerms.Bucket> iterator = teamTerms.getBuckets().iterator();
    while (iterator.hasNext()){
        StringTerms.Bucket bucket = iterator.next();
        //8:获取所有子聚合
        Map age_map = bucket.getAggregations().asMap();
​
        Double age = ((InternalAvg) age_map.get("avg_age")).getValue();
        double sum_salary = ((InternalSum) age_map.get("sum_salary")).getValue();
        System.out.println("球队 :" + bucket.getKey() + "  平均年龄： "+age + "  球队总salary：" + sum_salary);
    }
}
​
​

7：聚合排序

计算每个球队总年薪，并按照总年薪倒序排列 select team, sum(salary) as total_salary from player group by team order by total_salary desc;

/**
 * 排序
 计算每个球队总年薪，并按照总年薪倒序排列
 select team, sum(salary) as total_salary from player group by team order by total_salary desc;
 * */
@Test
public void groupOrder(){
    /**
     *
     * TermsBuilder teamAgg= AggregationBuilders.terms("team").order(Order.aggregation("total_salary ", false);
     SumBuilder salaryAgg= AggregationBuilders.avg("total_salary ").field("salary");
     sbuilder.addAggregation(teamAgg.subAggregation(salaryAgg));
     * */
​
    //1：构建查询提交
    SearchRequestBuilder builder = client.prepareSearch("player").setTypes("player");
    //2：指定聚合条件
    TermsAggregationBuilder team = AggregationBuilders.terms("player_count").field("team")
            .order(Terms.Order.aggregation("total_salary", false));//false是降序，true是升序
    //3: 指定要查哪一个字段
    SumAggregationBuilder salaryField= AggregationBuilders.sum("total_salary").field("salary");
    //: 找到聚合关系：父子关系
    team.subAggregation(salaryField);
    //3:将聚合条件放入查询条件中
    builder.addAggregation(team);
    //4:执行action，返回searchResponse
    SearchResponse searchResponse = builder.get();
    //5:将查询返回的searchResponse转换成map
    Map<String, Aggregation> aggregationMap = searchResponse.getAggregations().asMap();
    //6:取出聚合的字段
    StringTerms teamTerms = (StringTerms)aggregationMap.get("player_count");
    //7:对聚合的字段进行迭代
    Iterator<StringTerms.Bucket> iterator = teamTerms.getBuckets().iterator();
    while (iterator.hasNext()){
        StringTerms.Bucket bucket = iterator.next();
        //8:获取所有子聚合
        Map age_map = bucket.getAggregations().asMap();
​
        double sum_salary = ((InternalSum) age_map.get("total_salary")).getValue();
        System.out.println("球队 :" + bucket.getKey() + "  球队总salary：" + sum_salary);
    }
}
​
​