ElasticSearch中的关于对象数组查询，请注意要正确使用Nested类型

ElasticSearch中的关于对象数组查询，请注意要正确使用Nested类型，使用不当会导致检索时结果不正确。

直接上Demo演示

假设张三和李四分别喜欢喝某品牌的啤酒，青岛（黑啤和白啤）燕京（黑啤和白啤）

curl -X PUT 'http://localhost:9200/my-index-test001/_doc/1' -H 'Content-Type: application/json' -d '
{
  "id" : "1",
  "name": "张三",
  "favoriteBeers" : [ 
    {
      "brand" : "青岛",
      "type" :  "黑啤"
    },
    {
      "brand" : "燕京",
      "type" :  "白啤"
    }
  ]
}'

curl -X PUT 'http://localhost:9200/my-index-test001/_doc/2' -H 'Content-Type: application/json' -d '
{
  "id" : "2",
  "name": "李四",
  "favoriteBeers" : [ 
    {
      "brand" : "青岛",
      "type" :  "白啤"
    },
    {
      "brand" : "燕京",
      "type" :  "黑啤"
    }
  ]
}'

要查询喜欢喝青岛白啤的用户

curl -X GET 'http://localhost:9200/my-index-test001/_search?pretty=true' -H 'Content-Type: application/json' -d '
{
  "query" :{
    "bool":{
      "must":[
        {"match": {"favoriteBeers.brand": "青岛"}},
        {"match": {"favoriteBeers.type": "白啤"}}
      ]
    }
  }
}'

查询结果如下：

{
  "took" : 3,
  "timed_out" : false,
  "_shards" : {
    "total" : 1,
    "successful" : 1,
    "skipped" : 0,
    "failed" : 0
  },
  "hits" : {
    "total" : {
      "value" : 2,
      "relation" : "eq"
    },
    "max_score" : 0.7976569,
    "hits" : [
      {
        "_index" : "my-index-test001",
        "_type" : "_doc",
        "_id" : "1",
        "_score" : 0.7976569,
        "_source" : {
          "id" : "1",
          "name" : "张三",
          "favoriteBeers" : [
            {
              "brand" : "青岛",
              "type" : "黑啤"
            },
            {
              "brand" : "燕京",
              "type" : "白啤"
            }
          ]
        }
      },
      {
        "_index" : "my-index-test001",
        "_type" : "_doc",
        "_id" : "2",
        "_score" : 0.7976569,
        "_source" : {
          "id" : "2",
          "name" : "李四",
          "favoriteBeers" : [
            {
              "brand" : "青岛",
              "type" : "白啤"
            },
            {
              "brand" : "燕京",
              "type" : "黄啤"
            }
          ]
        }
      }
    ]
  }
}

我们预想中，这个检索应该只能检索出李四，但是结果是张三也被匹配到了？为什么呢？

前面张三的文档将在内部转换为看起来更像这样的文档：

{
  "name" :        "张三",
  "favoriteBeers.brand" : [ "青岛", "燕京" ],
  "favoriteBeers.type" :  [ "黑啤", "白啤" ]
}

那我们该怎么做呢？

很简单，ElasticSearch官方告诉我们，将nested字段用于对象数组即可。

如果需要索引对象数组并保持数组中每个对象的独立性，请使用nested数据类型而不是object数据类型。

我们先将这条索引删除，重新定义mapping映射文件：

curl -X DELETE 'http://localhost:9200/my-index-test001'

curl -X PUT 'http://localhost:9200/my-index-test001?pretty' -H 'Content-Type: application/json' -d '
{
  "mappings": {
    "properties": {
      "id": {
        "type": "keyword"
      },
      "name": {
        "type": "keyword"
      },
      "favoriteBeers": {
        "type": "nested",
        "properties": {
          "brand": {
            "type": "keyword"
          },
          "type": {
            "type": "keyword"
          }
        }
      }
    }
  }
}'

重新索引之前的张三、李四

请注意这个新查询，相应要增加nested path

curl -X GET 'http://localhost:9200/my-index-test001/_search?pretty=true' -H 'Content-Type: application/json' -d '
{
  "query" : {
    "bool": {
      "must":[
        {
          "nested":{
            "path":"favoriteBeers",
            "query":{
              "bool":{
                "must":[
                  {"match":{"favoriteBeers.brand":"青岛"}},
                  {"match":{"favoriteBeers.type":"白啤"}}
                ]
              }
            }
          }
        }
      ]
    }
  }
}'

得到正常结果：

{
  "took" : 3,
  "timed_out" : false,
  "_shards" : {
    "total" : 1,
    "successful" : 1,
    "skipped" : 0,
    "failed" : 0
  },
  "hits" : {
    "total" : {
      "value" : 1,
      "relation" : "eq"
    },
    "max_score" : 1.3862942,
    "hits" : [
      {
        "_index" : "my-index-test001",
        "_type" : "_doc",
        "_id" : "2",
        "_score" : 1.3862942,
        "_source" : {
          "id" : "2",
          "name" : "李四",
          "favoriteBeers" : [
            {
              "brand" : "青岛",
              "type" : "白啤"
            },
            {
              "brand" : "燕京",
              "type" : "黄啤"
            }
          ]
        }
      }
    ]
  }
}

参考文档：
https://www.elastic.co/guide/en/elasticsearch/reference/current/nested.html