通过kibana理解Elasticsearch中搜索，聚合的一些概念

1.在Java中对应的entity

public class Person {
    String name;
    int age;
}

2.kibana中的数据

      {
        "_index": "person_index",
        "_type": "person_type",
        "_id": "i674x3IBsOhWggMeikEo",
        "_score": 1,
        "_source": {
          "age": 9,
          "name": "李四"
        }
      },
      {
        "_index": "person_index",
        "_type": "person_type",
        "_id": "jq74x3IBsOhWggMeikGu",
        "_score": 1,
        "_source": {
          "age": 63,
          "name": "侯七"
        }
      },
      {
        "_index": "person_index",
        "_type": "person_type",
        "_id": "ja74x3IBsOhWggMeikGD",
        "_score": 1,
        "_source": {
          "age": 41,
          "name": "马六"
        }
      },
      {
        "_index": "person_index",
        "_type": "person_type",
        "_id": "j674x3IBsOhWggMeikHH",
        "_score": 1,
        "_source": {
          "age": 452,
          "name": "赵八"
        }
      },
      {
        "_index": "person_index",
        "_type": "person_type",
        "_id": "iq74x3IBsOhWggMeiUGi",
        "_score": 1,
        "_source": {
          "age": 16,
          "name": "张三"
        }
      },
      {
        "_index": "person_index",
        "_type": "person_type",
        "_id": "jK74x3IBsOhWggMeikFX",
        "_score": 1,
        "_source": {
          "age": 32,
          "name": "王五"
        }
      },
      {
        "_index": "person_index",
        "_type": "person_type",
        "_id": "kK74x3IBsOhWggMeikHU",
        "_score": 1,
        "_source": {
          "age": 32,
          "name": "孙九"
        }
      },
      {
        "_index": "person_index",
        "_type": "person_type",
        "_id": "ka74x3IBsOhWggMeikHg",
        "_score": 1,
        "_source": {
          "age": -8,
          "name": "钱二麻子"
        }
      }

3.kibana搜索

可以看出，第一层的hits表示根据请求条件命中的结果的参数，如total，max_score。

hits里的hits就是命中的结果集合，里面是每个结果的index索引，type类型等等，source即对象的值。

4.聚合结果

上面的json搜索条件是先搜出年龄小于60，然后再对搜出的结果进行term分组，类似于sql语句里的group by。

搜索结果里的aggregations下的group下的buckets即为符合条件的聚合结果的集合，即桶。

group是为整个分组结果起的名