Elastic Search join datatype 数据类型的使用

定义：join 数据类型是一个特殊字段，用于在相同索引的文档中创建一对多的关系。

使用场景：支持文档一对多关系的映射，并且多的一方写多读少

实践：

需求：设计一个用户购买了 多个商品 的数据模型

定义索引：

// 创建 index
PUT join_test
{
  "settings": {
    "number_of_replicas": 1, // 副分片数 初始化数据 设置成 0
    "number_of_shards" : 5, // 主分片数 
  }
}

// 设置index 的字段  父子文档共享
POST join_test/_doc/_mapping
{
  "properties": {
    "user_id": {
      "type": "keyword"
    },

    "bought_products_pid": {
      "type": "keyword"
    },
    "bought_products_price": {
      "type": "long"
    },
    "bought_products_tm": {
      "type": "long"
    },
    "my_join_field": { // 定义 购买商品 与用户文档的映射关系
      "type": "join",
      "relations": {
        "user": [ // user 为用户信息文档标识 父文档类型
          "bought_products" // 购买商品文档标识
        ]
      }
    }
  }
}

插入用户数据：

PUT join_test/_doc/1
{
  "user_id":"1",
  "my_join_field":{
    "name":"user"
  }
}

PUT join_test/_doc/2
{
  "user_id":"2",
  "my_join_field":{
    "name":"user"
  }
}

插入购买商品数据

// 用户1 购买 商品 1和2
// doc id 规则  子文档类型 + bought_products_pid 
// 子文档一定要更父文档放在一块 routing = 父文档 doc id
PUT join_test/_doc/bp_1?routing=1
{
  "bought_products_pid":"1",
  "bought_products_tm":1,
  "bought_products_price":1,
  "user_id":"1",
  "my_join_field":{
    "name":"bought_products", // 子文档类型
    "parent": "1" // 父文档 doc id
  }
}

PUT join_test/_doc/bp_2?routing=1
{
  "bought_products_pid":"2",
  "bought_products_tm":1,
  "bought_products_price":2,
  "user_id":"1",
  "my_join_field":{
    "name":"bought_products",
    "parent": "1"
  }
}

// 用户2购买商品2
PUT join_test/_doc/bp_2?routing=2
{
  "bought_products_pid":"2",
  "bought_products_tm":1,
  "bought_products_price":2,
  "user_id":"2",
  "my_join_field":{
    "name":"bought_products",
    "parent": "2"
  }
}

// 用户2购买商品3
PUT join_test/_doc/bp_3?routing=2
{
  "bought_products_pid":"3",
  "bought_products_price":3,
  "bought_products_tm":1,
  "user_id":"2",
  "my_join_field":{
    "name":"bought_products",
    "parent": "2"
  }
}

搜索：

1、查出购买商品2 的所有用户

# 查询父文档 使用 has_child 语法
GET join_test/_search
{
  "query": {
    "bool": {
      "must": [
        {
          "has_child": {
            "type": "bought_products",
            "query": {
              "match": {
                "bought_products_pid": "2"
              }
            }
          }
        }
      ]
    }
  }
}

# 响应结果
{
  "took" : 3,
  "timed_out" : false,
  "_shards" : {
    "total" : 5,
    "successful" : 5,
    "skipped" : 0,
    "failed" : 0
  },
  "hits" : {
    "total" : 2,
    "max_score" : 1.0,
    "hits" : [
      {
        "_index" : "join_test",
        "_type" : "_doc",
        "_id" : "2",
        "_score" : 1.0,
        "_source" : {
          "user_id" : "2",
          "my_join_field" : {
            "name" : "user"
          }
        }
      },
      {
        "_index" : "join_test",
        "_type" : "_doc",
        "_id" : "1",
        "_score" : 1.0,
        "_source" : {
          "user_id" : "1",
          "my_join_field" : {
            "name" : "user"
          }
        }
      }
    ]
  }
}

2、查询用户1购买的所有商品

# 查询 子文档 使用 has_parent 并且 需要 指定 需要查询出的子文档类型
GET join_test/_search
{
  "query": {
    "bool": {
      "must": [
        {
          "match": {
            "my_join_field": "bought_products" // 需要查出来的文档类型
          }
        },
        {
          "has_parent": { // 父文档条件
            "parent_type": "user",
            "query": {
              "match": {
                "user_id": "1"
              }
            }
          }
        }
      ]
    }
  }
}

# 响应结果
{
  "took" : 2,
  "timed_out" : false,
  "_shards" : {
    "total" : 5,
    "successful" : 5,
    "skipped" : 0,
    "failed" : 0
  },
  "hits" : {
    "total" : 2,
    "max_score" : 1.3566749,
    "hits" : [
      {
        "_index" : "join_test",
        "_type" : "_doc",
        "_id" : "bp_1",
        "_score" : 1.3566749,
        "_routing" : "1",
        "_source" : {
          "bought_products_pid" : "1",
          "bought_products_tm" : 1,
          "user_id" : "1",
          "my_join_field" : {
            "name" : "bought_products",
            "parent" : "1"
          }
        }
      },
      {
        "_index" : "join_test",
        "_type" : "_doc",
        "_id" : "bp_2",
        "_score" : 1.3566749,
        "_routing" : "1",
        "_source" : {
          "bought_products_pid" : "2",
          "bought_products_tm" : 1,
          "bought_products_price" : 2,
          "user_id" : "1",
          "my_join_field" : {
            "name" : "bought_products",
            "parent" : "1"
          }
        }
      }
    ]
  }
}

聚合：

1、统计支付金额 Top 10的用户

GET join_test/_search
{
  "size": 0,
  "query": {
    "bool": {
      "must": [
        {
          "match": {
            "my_join_field": "user"
          }
        }
      ]
    }
  },
  "aggs": {
    "group_by_user_id": {
      "terms": {
        "field": "user_id",
        "size": 10,
        "execution_hint": "map" // 查询结果 放在 内存中 进行 聚合，结果集小的可以加快 聚合速度
      },
      "aggs": {
        "sum_price": {
          "children": {
            "type": "bought_products"
          },
          "aggs": {
            "sum_price": {
              "sum": {
                "field": "bought_products_price"
              }
            }
          }
        },
        "sort": {
          "bucket_sort": {
            "sort": [
              {
                "sum_price>sum_price": {
                  "order": "desc"
                }
              }
            ]
          }
        }
      }
    }
  }
}


{
  "took" : 2,
  "timed_out" : false,
  "_shards" : {
    "total" : 5,
    "successful" : 5,
    "skipped" : 0,
    "failed" : 0
  },
  "hits" : {
    "total" : 2,
    "max_score" : 0.0,
    "hits" : [ ]
  },
  "aggregations" : {
    "group_by_user_id" : {
      "doc_count_error_upper_bound" : 0,
      "sum_other_doc_count" : 0,
      "buckets" : [
        {
          "key" : "2",
          "doc_count" : 1,
          "sum_price" : {
            "doc_count" : 2,
            "sum_price" : {
              "value" : 5.0
            }
          }
        },
        {
          "key" : "1",
          "doc_count" : 1,
          "sum_price" : {
            "doc_count" : 2,
            "sum_price" : {
              "value" : 3.0
            }
          }
        }
      ]
    }
  }
}