关联（join）字段类型

关联字段类型用于在同一索引内的文档之间建立父子关系。

示例

创建一个映射，以在产品与其品牌之间建立父子关系：

PUT testindex1
{
  "mappings": {
    "properties": {
      "product_to_brand": { 
        "type": "join",
        "relations": {
          "brand": "product" 
        }
      }
    }
  }
}

然后，索引一个包含关联字段类型的父文档：

PUT testindex1/_doc/1
{
  "name": "Brand 1",
  "product_to_brand": {
    "name": "brand" 
  }
}

您也可以使用不带对象表示法的简写形式来索引父文档：

PUT testindex1/_doc/1
{
  "name": "Brand 1",
  "product_to_brand" : "brand" 
}

索引子文档时，需要指定 routing 查询参数，因为同一父子层级结构中的父文档和子文档必须被索引到同一个分片上每个子文档在 parent 字段中引用其父文档的 ID。

索引两个子文档，每个父文档对应一个：

PUT testindex1/_doc/3?routing=1
{
  "name": "Product 1",
  "product_to_brand": {
    "name": "product", 
    "parent": "1" 
  }
}

PUT testindex1/_doc/4?routing=1
{
  "name": "Product 2",
  "product_to_brand": {
    "name": "product", 
    "parent": "1" 
  }
}

查询关联字段

查询关联字段时，响应会包含子字段，用于指明返回的文档是父文档还是子文档。对于子对象，也会返回父文档的 ID。

搜索所有文档

GET testindex1/_search
{
  "query": {
    "match_all": {}
  }
}

响应指明了文档是父文档还是子文档：

{
  "took" : 4,
  "timed_out" : false,
  "_shards" : {
    "total" : 1,
    "successful" : 1,
    "skipped" : 0,
    "failed" : 0
  },
  "hits" : {
    "total" : {
      "value" : 3,
      "relation" : "eq"
    },
    "max_score" : 1.0,
    "hits" : [
      {
        "_index" : "testindex1",
        "_type" : "_doc",
        "_id" : "1",
        "_score" : 1.0,
        "_source" : {
          "name" : "Brand 1",
          "product_to_brand" : {
            "name" : "brand"
          }
        }
      },
      {
        "_index" : "testindex1,
        "_type" : "_doc",
        "_id" : "3",
        "_score" : 1.0,
        "_routing" : "1",
        "_source" : {
          "name" : "Product 1",
          "product_to_brand" : {
            "name" : "product",
            "parent" : "1"
          }
        }
      },
      {
        "_index" : "testindex1",
        "_type" : "_doc",
        "_id" : "4",
        "_score" : 1.0,
        "_routing" : "1",
        "_source" : {
          "name" : "Product 2",
          "product_to_brand" : {
            "name" : "product",
            "parent" : "1"
          }
        }
      }
    ]
  }
}

搜索某个父文档的所有子文档

查找与品牌 1 关联的所有产品：

GET testindex1/_search
{
  "query" : {
    "has_parent": {
      "parent_type":"brand",
      "query": {
        "match" : {
          "name": "Brand 1"
        }
      }
    }
  }
}

响应包含了与品牌 1 关联的产品 1 和产品 2：

{
  "took" : 7,
  "timed_out" : false,
  "_shards" : {
    "total" : 1,
    "successful" : 1,
    "skipped" : 0,
    "failed" : 0
  },
  "hits" : {
    "total" : {
      "value" : 2,
      "relation" : "eq"
    },
    "max_score" : 1.0,
    "hits" : [
      {
        "_index" : "testindex1",
        "_type" : "_doc",
        "_id" : "3",
        "_score" : 1.0,
        "_routing" : "1",
        "_source" : {
          "name" : "Product 1",
          "product_to_brand" : {
            "name" : "product",
            "parent" : "1"
          }
        }
      },
      {
        "_index" : "testindex1",
        "_type" : "_doc",
        "_id" : "4",
        "_score" : 1.0,
        "_routing" : "1",
        "_source" : {
          "name" : "Product 2",
          "product_to_brand" : {
            "name" : "product",
            "parent" : "1"
          }
        }
      }
    ]
  }
}

搜索某个子文档的父文档

查找产品 1 的父文档：

GET testindex1/_search
{
  "query" : {
    "has_child": {
      "type":"product",
      "query": {
        "match" : {
            "name": "Product 1"
        }
      }
    }
  }
}

响应返回了品牌 1 作为产品 1 的父文档：

{
  "took" : 4,
  "timed_out" : false,
  "_shards" : {
    "total" : 1,
    "successful" : 1,
    "skipped" : 0,
    "failed" : 0
  },
  "hits" : {
    "total" : {
      "value" : 1,
      "relation" : "eq"
    },
    "max_score" : 1.0,
    "hits" : [
      {
        "_index" : "testindex1",
        "_type" : "_doc",
        "_id" : "1",
        "_score" : 1.0,
        "_source" : {
          "name" : "Brand 1",
          "product_to_brand" : {
            "name" : "brand"
          }
        }
      }
    ]
  }
}

具有多个子文档的父文档

一个父文档可以有多个子文档。创建包含多个子文档的映射：

PUT testindex1
{
  "mappings": {
    "properties": {
      "parent_to_child": {
        "type": "join",
        "relations": {
          "parent": ["child 1", "child 2"]  
        }
      }
    }
  }
}

关联字段类型注意事项

  • 一个索引中只能有一个关联字段映射。

  • 检索、更新或删除子文档时，需要提供路由参数。这是因为同一关系中的父文档和子文档必须被索引到同一个分片上。

  • 不支持多个父文档。

  • 只有当现有文档已被标记为父文档时，才能向其添加子文档。

  • 可以向现有的关联字段添加新的关系。

后续步骤