按嵌套字段对文档进行排序

Question

我正在尝试按嵌套字段 sections.name 对 ElasticSearch 返回的结果进行排序，如下所示：

映射：

PUT /staff
{
    "mappings": {
        "list": {
            "properties": {
                "id": {"type": "text" },
                "name": {
                  "type":"text",
                  "fields" : {
                    "keyword" : {
                      "type" : "keyword",
                      "ignore_above" : 256
                    }
                  }
                },
                "sections" : {
                  "type":"nested",
                  "properties": {
                    "id": {"type":"text", "fielddata" : true},
                    "name": {
                    "fielddata" : true,
                    "type": "text",
                    "fields": {
                    "keyword" : {
                      "type" : "keyword",
                      "ignore_above" : 256
                      }
                    }
                  }
                }
              }
            }
        }
    }
}

文件：

POST /staff/list
{
    "id": 10,
    "name": "abc def",
    "sections":
    [
      {
        "id":"1",
        "name" : "zamphire"
      },{
        "id":"2",
        "name" : "warden"
      }
    ]
}

POST /staff/list
{
    "id": 9,
    "name": "abc def",
    "sections":
    [
      {
        "id":"1",
        "name" : "shaggi"
      },{
        "id":"2",
        "name" : "robert"
      }
    ]
}

POST /staff/list
{
    "id": 8,
    "name": "abc def",
    "sections":
    [
      {
        "id":"3",
        "name" : "zamphire"
      },{
        "id":"2",
        "name" : "abi"
      }
    ]
}

我正在执行以下查询：

GET /staff/_search
{
  "from": 0,
  "query": {
    "nested": {
      "path": "sections",
      "query": {
        "match": {
          "sections.id": {
            "query": "1"
          }
        }
      }
    }
  },
  "size": 25,
  "sort": [
    {
      "sections.name": {
        "nested": {
          "filter": {
            "nested": {
              "path": "sections",
              "query": {
                 "term" : { "sections.id" : "1" }
              }
            }
          }
        },
        "order": "asc"
      }
    }
  ],
  "_source": {
    "includes": [
      "id",
      "name",
      "sections"
    ]
  }
}

我得到这些结果：

{
  "took" : 2,
  "timed_out" : false,
  "_shards" : {
    "total" : 5,
    "successful" : 5,
    "skipped" : 0,
    "failed" : 0
  },
  "hits" : {
    "total" : 2,
    "max_score" : null,
    "hits" : [
      {
        "_index" : "staff",
        "_type" : "list",
        "_id" : "rJtyyGwBNB-cdBRb5XGR",
        "_score" : null,
        "_source" : {
          "name" : "abc def",
          "id" : 10,
          "sections" : [
            {
              "name" : "zamphire",
              "id" : "1"
            },
            {
              "name" : "warden",
              "id" : "2"
            }
          ]
        },
        "sort" : [
          null
        ]
      },
      {
        "_index" : "staff",
        "_type" : "list",
        "_id" : "rZtyyGwBNB-cdBRb6nHU",
        "_score" : null,
        "_source" : {
          "name" : "abc def",
          "id" : 9,
          "sections" : [
            {
              "name" : "shaggi",
              "id" : "1"
            },
            {
              "name" : "robert",
              "id" : "2"
            }
          ]
        },
        "sort" : [
          null
        ]
      }
    ]
  }
}

我希望 shaggi 部分出现在 zamphire 之前，因此两个文档的顺序应该颠倒。

我在结果中注意到这一点：

"sort" : [
  null
]

有关系吗？我在这里错过了什么？

Answer 1

将 sort 部分更改为此应该根据 docs

完成工作

  "sort": [
    {
      "sections.name": {
        "order": "asc", 
        "nested": {
          "path": "sections",
          "filter": {
            "term" : { "sections.id" : "1" }
          }
        }
      }
    }
  ]

Returns

{
  "took" : 7,
  "timed_out" : false,
  "_shards" : {
    "total" : 1,
    "successful" : 1,
    "skipped" : 0,
    "failed" : 0
  },
  "hits" : {
    "total" : {
      "value" : 2,
      "relation" : "eq"
    },
    "max_score" : null,
    "hits" : [
      {
        "_index" : "staff",
        "_type" : "_doc",
        "_id" : "8hSJyWwBHfpsFyAs9f_8",
        "_score" : null,
        "_source" : {
          "name" : "abc def",
          "id" : 9,
          "sections" : [
            {
              "name" : "shaggi",
              "id" : "1"
            },
            {
              "name" : "robert",
              "id" : "2"
            }
          ]
        },
        "sort" : [
          "shaggi"
        ]
      },
      {
        "_index" : "staff",
        "_type" : "_doc",
        "_id" : "8RSJyWwBHfpsFyAs5v98",
        "_score" : null,
        "_source" : {
          "name" : "abc def",
          "id" : 10,
          "sections" : [
            {
              "name" : "zamphire",
              "id" : "1"
            },
            {
              "name" : "warden",
              "id" : "2"
            }
          ]
        },
        "sort" : [
          "zamphire"
        ]
      }
    ]
  }
}

使用 elasticsearch 7.2.0 测试。

希望对您有所帮助。

按嵌套字段对文档进行排序

Sorting documents by a nested field

elasticsearch

nested

field

elasticsearch-6