1、背景

我们知道在sql中是可以实现 group by 字段a,字段b，那么这种效果在elasticsearch中该如何实现呢？此处我们记录在elasticsearch中的3种方式来实现这个效果。

2、实现多字段聚合的思路

图片来源：https://www.elastic.co/guide/en/elasticsearch/reference/current/search-aggregations-bucket-terms-aggregation.html
从上图中，我们可以知道，可以通过3种方式来实现 多字段的聚合操作。

3、需求

根据省(province)和性别(sex)来进行聚合，然后根据聚合后的每个桶的数据，在根据每个桶中的最大年龄(age)来进行倒序排序。

4、数据准备

4.1 创建索引

PUT /index_person
{
  "settings": {
    "number_of_shards": 1
  },
  "mappings": {
    "properties": {
      "id": {
        "type": "long"
      },
      "name": {
        "type": "keyword"
      },
      "province": {
        "type": "keyword"
      },
      "sex": {
        "type": "keyword"
      },
      "age": {
        "type": "integer"
      },
      "address": {
        "type": "text",
        "analyzer": "ik_max_word",
        "fields": {
          "keyword": {
            "type": "keyword",
            "ignore_above": 256
          }
        }
      }
    }
  }
}

4.2 准备数据

PUT /_bulk
{"create":{"_index":"index_person","_id":1}}
{"id":1,"name":"张三","sex":"男","age":20,"province":"湖北","address":"湖北省黄冈市罗田县匡河镇"}
{"create":{"_index":"index_person","_id":2}}
{"id":2,"name":"李四","sex":"男","age":19,"province":"江苏","address":"江苏省南京市"}
{"create":{"_index":"index_person","_id":3}}
{"id":3,"name":"王武","sex":"女","age":25,"province":"湖北","address":"湖北省武汉市江汉区"}
{"create":{"_index":"index_person","_id":4}}
{"id":4,"name":"赵六","sex":"女","age":30,"province":"北京","address":"北京市东城区"}
{"create":{"_index":"index_person","_id":5}}
{"id":5,"name":"钱七","sex":"女","age":16,"province":"北京","address":"北京市西城区"}
{"create":{"_index":"index_person","_id":6}}
{"id":6,"name":"王八","sex":"女","age":45,"province":"北京","address":"北京市朝阳区"}

5、实现方式

5.1 multi_terms实现

5.1.1 dsl

GET /index_person/_search
{
  "size": 0,
  "aggs": {
    "agg_province_sex": {
      "multi_terms": {
        "size": 10,
        "shard_size": 25,
        "order":{
          "max_age": "desc"    
        },
        "terms": [
          {
            "field": "province",
            "missing": "defaultProvince"
          },
          {
            "field": "sex"
          }
        ]
      },
      "aggs": {
        "max_age": {
          "max": {
            "field": "age"
          }
        }
      }
    }
  }
}

5.1.2 java 代码

    @Test
    @DisplayName("多term聚合-根据省和性别聚合，然后根据最大年龄倒序")
    public void agg01() throws IOException {

        SearchRequest searchRequest = new SearchRequest.Builder()
                .size(0)
                .index("index_person")
                .aggregations("agg_province_sex", agg ->
                        agg.multiTerms(multiTerms ->
                                        multiTerms.terms(term -> term.field("province"))
                                                .terms(term -> term.field("sex"))
                                                .order(new NamedValue<>("max_age", SortOrder.Desc))
                                )
                                .aggregations("max_age", ageAgg ->
                                        ageAgg.max(max -> max.field("age")))

                )
                .build();
        System.out.println(searchRequest);
        SearchResponse<Object> response = client.search(searchRequest, Object.class);
        System.out.println(response);
    }

5.1.3 运行结果

5.2 script实现

5.2.1 dsl

GET /index_person/_search
{
  "size": 0,
  "runtime_mappings": {
    "runtime_province_sex": {
      "type": "keyword",
      "script": """
          String province = doc['province'].value;
          String sex = doc['sex'].value;
          emit(province + '|' + sex);
      """
    }
  },
  "aggs": {
    "agg_province_sex": {
      "terms": {
        "field": "runtime_province_sex",
        "size": 10,
        "shard_size": 25,
        "order": {
          "max_age": "desc"
        }
      },
      "aggs": {
        "max_age": {
          "max": {
            "field": "age"
          }
        }
      }
    }
  }
}

5.2.2 java代码

@Test
    @DisplayName("多term聚合-根据省和性别聚合，然后根据最大年龄倒序")
    public void agg02() throws IOException {

        SearchRequest searchRequest = new SearchRequest.Builder()
                .size(0)
                .index("index_person")
                .runtimeMappings("runtime_province_sex", field -> {
                    field.type(RuntimeFieldType.Keyword);
                    field.script(script -> script.inline(new InlineScript.Builder()
                            .lang(ScriptLanguage.Painless)
                            .source("String province = doc['province'].value;\n" +
                                    "          String sex = doc['sex'].value;\n" +
                                    "          emit(province + '|' + sex);")
                            .build()));
                    return field;
                })
                .aggregations("agg_province_sex", agg ->
                        agg.terms(terms ->
                                        terms.field("runtime_province_sex")
                                                .size(10)
                                                .shardSize(25)
                                                .order(new NamedValue<>("max_age", SortOrder.Desc))
                                )
                                .aggregations("max_age", minAgg ->
                                        minAgg.max(max -> max.field("age")))
                )
                .build();
        System.out.println(searchRequest);
        SearchResponse<Object> response = client.search(searchRequest, Object.class);
        System.out.println(response);
    }

5.2.3 运行结果

5.3 通过copyto实现

我本地测试过，通过copyto没实现，此处故先不考虑

5.5 通过pipeline来实现

实现思路：
创建mapping时，多创建一个字段pipeline_province_sex，该字段的值由创建数据时指定pipeline来生产。

5.4.1 创建mapping

PUT /index_person
{
  "settings": {
    "number_of_shards": 1
  },
  "mappings": {
    "properties": {
      "id": {
        "type": "long"
      },
      "name": {
        "type": "keyword"
      },
      "province": {
        "type": "keyword"
      },
      "sex": {
        "type": "keyword"
      },
      "age": {
        "type": "integer"
      },
      "pipeline_province_sex":{
        "type": "keyword"
      },
      "address": {
        "type": "text",
        "analyzer": "ik_max_word",
        "fields": {
          "keyword": {
            "type": "keyword",
            "ignore_above": 256
          }
        }
      }
    }
  }
}

此处指定了一个字段pipeline_province_sex，该字段的值会由pipeline来处理。

5.4.2 创建pipeline

PUT _ingest/pipeline/pipeline_index_person_provice_sex
{
  "description": "将provice和sex的值拼接起来",
  "processors": [
    {
      "set": {
        "field": "pipeline_province_sex",
        "value": ["{{province}}", "{{sex}}"]
      }, 
      "join": {
        "field": "pipeline_province_sex",
        "separator": "|"
      }
    }
  ]
}

5.4.3 插入数据

PUT /_bulk?pipeline=pipeline_index_person_provice_sex
{"create":{"_index":"index_person","_id":1}}
{"id":1,"name":"张三","sex":"男","age":20,"province":"湖北","address":"湖北省黄冈市罗田县匡河镇"}
{"create":{"_index":"index_person","_id":2}}
{"id":2,"name":"李四","sex":"男","age":19,"province":"江苏","address":"江苏省南京市"}
{"create":{"_index":"index_person","_id":3}}
{"id":3,"name":"王武","sex":"女","age":25,"province":"湖北","address":"湖北省武汉市江汉区"}
{"create":{"_index":"index_person","_id":4}}
{"id":4,"name":"赵六","sex":"女","age":30,"province":"北京","address":"北京市东城区"}
{"create":{"_index":"index_person","_id":5}}
{"id":5,"name":"钱七","sex":"女","age":16,"province":"北京","address":"北京市西城区"}
{"create":{"_index":"index_person","_id":6}}
{"id":6,"name":"王八","sex":"女","age":45,"province":"北京","address":"北京市朝阳区"}

注意： 此处的插入需要指定上一步的pipeline
PUT /_bulk?pipeline=pipeline_index_person_provice_sex

5.4.4 聚合dsl

GET /index_person/_search
{
  "size": 0,
  "aggs": {
    "agg_province_sex": {
      "terms": {
        "field": "pipeline_province_sex",
        "size": 10,
        "shard_size": 25,
        "order": {
          "max_age": "desc"   
        }
      }, 
      "aggs": {
        "max_age": {
          "max": {
            "field": "age"
          }
        }
      }
    }
  }
}

5.4.5 运行结果

6、实现代码

https://gitee.com/huan1993/spring-cloud-parent/blob/master/es/es8-api/src/main/java/com/huan/es8/aggregations/bucket/MultiTermsAggs.java

7、参考文档

1. https://www.elastic.co/guide/en/elasticsearch/reference/current/search-aggregations-bucket-terms-aggregation.html