IMPORTANT: No additional bug fixes or documentation updates will be released for this version. For the latest information, see the current release documentation.
« Joining queries Has child query »
Elastic Docs ›Elasticsearch Guide [8.3] ›Query DSL ›Joining queries

Nested query

edit

Nested query

edit

Wraps another query to search nested fields.

The nested query searches nested field objects as if they were indexed as separate documents. If an object matches the search, the nested query returns the root parent document.

Example request

edit

Index setup

edit

To use the nested query, your index must include a nested field mapping. For example:

response = client.indices.create(
  index: 'my-index-000001',
  body: {
    mappings: {
      properties: {
        "obj1": {
          type: 'nested'
        }
      }
    }
  }
)
puts response
res, err := es.Indices.Create(
	"my-index-000001",
	es.Indices.Create.WithBody(strings.NewReader(`{
	  "mappings": {
	    "properties": {
	      "obj1": {
	        "type": "nested"
	      }
	    }
	  }
	}`)),
)
fmt.Println(res, err)
PUT /my-index-000001
{
  "mappings": {
    "properties": {
      "obj1": {
        "type": "nested"
      }
    }
  }
}

Example query

edit
response = client.search(
  index: 'my-index-000001',
  body: {
    query: {
      nested: {
        path: 'obj1',
        query: {
          bool: {
            must: [
              {
                match: {
                  "obj1.name": 'blue'
                }
              },
              {
                range: {
                  "obj1.count": {
                    gt: 5
                  }
                }
              }
            ]
          }
        },
        score_mode: 'avg'
      }
    }
  }
)
puts response
res, err := es.Search(
	es.Search.WithIndex("my-index-000001"),
	es.Search.WithBody(strings.NewReader(`{
	  "query": {
	    "nested": {
	      "path": "obj1",
	      "query": {
	        "bool": {
	          "must": [
	            {
	              "match": {
	                "obj1.name": "blue"
	              }
	            },
	            {
	              "range": {
	                "obj1.count": {
	                  "gt": 5
	                }
	              }
	            }
	          ]
	        }
	      },
	      "score_mode": "avg"
	    }
	  }
	}`)),
	es.Search.WithPretty(),
)
fmt.Println(res, err)
GET /my-index-000001/_search
{
  "query": {
    "nested": {
      "path": "obj1",
      "query": {
        "bool": {
          "must": [
            { "match": { "obj1.name": "blue" } },
            { "range": { "obj1.count": { "gt": 5 } } }
          ]
        }
      },
      "score_mode": "avg"
    }
  }
}

Top-level parameters for nested

edit
path
(Required, string) Path to the nested object you wish to search.
query

(Required, query object) Query you wish to run on nested objects in the path. If an object matches the search, the nested query returns the root parent document.

You can search nested fields using dot notation that includes the complete path, such as obj1.name.

Multi-level nesting is automatically supported, and detected, resulting in an inner nested query to automatically match the relevant nesting level, rather than root, if it exists within another nested query.

See Multi-level nested queries for an example.

score_mode

(Optional, string) Indicates how scores for matching child objects affect the root parent document’s relevance score. Valid values are:

avg (Default)
Use the mean relevance score of all matching child objects.
max
Uses the highest relevance score of all matching child objects.
min
Uses the lowest relevance score of all matching child objects.
none
Do not use the relevance scores of matching child objects. The query assigns parent documents a score of 0.
sum
Add together the relevance scores of all matching child objects.
ignore_unmapped

(Optional, Boolean) Indicates whether to ignore an unmapped path and not return any documents instead of an error. Defaults to false.

If false, Elasticsearch returns an error if the path is an unmapped field.

You can use this parameter to query multiple indices that may not contain the field path.

Notes

edit

Context of script queries

edit

If you run a script query within a nested query, you can only access doc values from the nested document, not the parent or root document.

Multi-level nested queries

edit

To see how multi-level nested queries work, first you need an index that has nested fields. The following request defines mappings for the drivers index with nested make and model fields.

response = client.indices.create(
  index: 'drivers',
  body: {
    mappings: {
      properties: {
        driver: {
          type: 'nested',
          properties: {
            last_name: {
              type: 'text'
            },
            vehicle: {
              type: 'nested',
              properties: {
                make: {
                  type: 'text'
                },
                model: {
                  type: 'text'
                }
              }
            }
          }
        }
      }
    }
  }
)
puts response
res, err := es.Indices.Create(
	"drivers",
	es.Indices.Create.WithBody(strings.NewReader(`{
	  "mappings": {
	    "properties": {
	      "driver": {
	        "type": "nested",
	        "properties": {
	          "last_name": {
	            "type": "text"
	          },
	          "vehicle": {
	            "type": "nested",
	            "properties": {
	              "make": {
	                "type": "text"
	              },
	              "model": {
	                "type": "text"
	              }
	            }
	          }
	        }
	      }
	    }
	  }
	}`)),
)
fmt.Println(res, err)
PUT /drivers
{
  "mappings": {
    "properties": {
      "driver": {
        "type": "nested",
        "properties": {
          "last_name": {
            "type": "text"
          },
          "vehicle": {
            "type": "nested",
            "properties": {
              "make": {
                "type": "text"
              },
              "model": {
                "type": "text"
              }
            }
          }
        }
      }
    }
  }
}

Next, index some documents to the drivers index.

$params = [
    'index' => 'drivers',
    'id' => '1',
    'body' => [
        'driver' => [
            'last_name' => 'McQueen',
            'vehicle' => [
                [
                    'make' => 'Powell Motors',
                    'model' => 'Canyonero',
                ],
                [
                    'make' => 'Miller-Meteor',
                    'model' => 'Ecto-1',
                ],
            ],
        ],
    ],
];
$response = $client->index($params);
$params = [
    'index' => 'drivers',
    'id' => '2',
    'body' => [
        'driver' => [
            'last_name' => 'Hudson',
            'vehicle' => [
                [
                    'make' => 'Mifune',
                    'model' => 'Mach Five',
                ],
                [
                    'make' => 'Miller-Meteor',
                    'model' => 'Ecto-1',
                ],
            ],
        ],
    ],
];
$response = $client->index($params);
resp = client.index(
    index="drivers",
    id="1",
    body={
        "driver": {
            "last_name": "McQueen",
            "vehicle": [
                {"make": "Powell Motors", "model": "Canyonero"},
                {"make": "Miller-Meteor", "model": "Ecto-1"},
            ],
        }
    },
)
print(resp)

resp = client.index(
    index="drivers",
    id="2",
    refresh=True,
    body={
        "driver": {
            "last_name": "Hudson",
            "vehicle": [
                {"make": "Mifune", "model": "Mach Five"},
                {"make": "Miller-Meteor", "model": "Ecto-1"},
            ],
        }
    },
)
print(resp)
response = client.index(
  index: 'drivers',
  id: 1,
  body: {
    driver: {
      last_name: 'McQueen',
      vehicle: [
        {
          make: 'Powell Motors',
          model: 'Canyonero'
        },
        {
          make: 'Miller-Meteor',
          model: 'Ecto-1'
        }
      ]
    }
  }
)
puts response

response = client.index(
  index: 'drivers',
  id: 2,
  refresh: true,
  body: {
    driver: {
      last_name: 'Hudson',
      vehicle: [
        {
          make: 'Mifune',
          model: 'Mach Five'
        },
        {
          make: 'Miller-Meteor',
          model: 'Ecto-1'
        }
      ]
    }
  }
)
puts response
{
	res, err := es.Index(
		"drivers",
		strings.NewReader(`{
	  "driver": {
	    "last_name": "McQueen",
	    "vehicle": [
	      {
	        "make": "Powell Motors",
	        "model": "Canyonero"
	      },
	      {
	        "make": "Miller-Meteor",
	        "model": "Ecto-1"
	      }
	    ]
	  }
	}`),
		es.Index.WithDocumentID("1"),
		es.Index.WithPretty(),
	)
	fmt.Println(res, err)
}

{
	res, err := es.Index(
		"drivers",
		strings.NewReader(`{
	  "driver": {
	    "last_name": "Hudson",
	    "vehicle": [
	      {
	        "make": "Mifune",
	        "model": "Mach Five"
	      },
	      {
	        "make": "Miller-Meteor",
	        "model": "Ecto-1"
	      }
	    ]
	  }
	}`),
		es.Index.WithDocumentID("2"),
		es.Index.WithRefresh("true"),
		es.Index.WithPretty(),
	)
	fmt.Println(res, err)
}
const response0 = await client.index({
  index: 'drivers',
  id: '1',
  body: {
    driver: {
      last_name: 'McQueen',
      vehicle: [
        {
          make: 'Powell Motors',
          model: 'Canyonero'
        },
        {
          make: 'Miller-Meteor',
          model: 'Ecto-1'
        }
      ]
    }
  }
})
console.log(response0)

const response1 = await client.index({
  index: 'drivers',
  id: '2',
  refresh: true,
  body: {
    driver: {
      last_name: 'Hudson',
      vehicle: [
        {
          make: 'Mifune',
          model: 'Mach Five'
        },
        {
          make: 'Miller-Meteor',
          model: 'Ecto-1'
        }
      ]
    }
  }
})
console.log(response1)
PUT /drivers/_doc/1
{
  "driver" : {
        "last_name" : "McQueen",
        "vehicle" : [
            {
                "make" : "Powell Motors",
                "model" : "Canyonero"
            },
            {
                "make" : "Miller-Meteor",
                "model" : "Ecto-1"
            }
        ]
    }
}

PUT /drivers/_doc/2?refresh
{
  "driver" : {
        "last_name" : "Hudson",
        "vehicle" : [
            {
                "make" : "Mifune",
                "model" : "Mach Five"
            },
            {
                "make" : "Miller-Meteor",
                "model" : "Ecto-1"
            }
        ]
    }
}

You can now use a multi-level nested query to match documents based on the make and model fields.

response = client.search(
  index: 'drivers',
  body: {
    query: {
      nested: {
        path: 'driver',
        query: {
          nested: {
            path: 'driver.vehicle',
            query: {
              bool: {
                must: [
                  {
                    match: {
                      "driver.vehicle.make": 'Powell Motors'
                    }
                  },
                  {
                    match: {
                      "driver.vehicle.model": 'Canyonero'
                    }
                  }
                ]
              }
            }
          }
        }
      }
    }
  }
)
puts response
res, err := es.Search(
	es.Search.WithIndex("drivers"),
	es.Search.WithBody(strings.NewReader(`{
	  "query": {
	    "nested": {
	      "path": "driver",
	      "query": {
	        "nested": {
	          "path": "driver.vehicle",
	          "query": {
	            "bool": {
	              "must": [
	                {
	                  "match": {
	                    "driver.vehicle.make": "Powell Motors"
	                  }
	                },
	                {
	                  "match": {
	                    "driver.vehicle.model": "Canyonero"
	                  }
	                }
	              ]
	            }
	          }
	        }
	      }
	    }
	  }
	}`)),
	es.Search.WithPretty(),
)
fmt.Println(res, err)
GET /drivers/_search
{
  "query": {
    "nested": {
      "path": "driver",
      "query": {
        "nested": {
          "path": "driver.vehicle",
          "query": {
            "bool": {
              "must": [
                { "match": { "driver.vehicle.make": "Powell Motors" } },
                { "match": { "driver.vehicle.model": "Canyonero" } }
              ]
            }
          }
        }
      }
    }
  }
}

The search request returns the following response:

{
  "took" : 5,
  "timed_out" : false,
  "_shards" : {
    "total" : 1,
    "successful" : 1,
    "skipped" : 0,
    "failed" : 0
  },
  "hits" : {
    "total" : {
      "value" : 1,
      "relation" : "eq"
    },
    "max_score" : 3.7349272,
    "hits" : [
      {
        "_index" : "drivers",
        "_id" : "1",
        "_score" : 3.7349272,
        "_source" : {
          "driver" : {
            "last_name" : "McQueen",
            "vehicle" : [
              {
                "make" : "Powell Motors",
                "model" : "Canyonero"
              },
              {
                "make" : "Miller-Meteor",
                "model" : "Ecto-1"
              }
            ]
          }
        }
      }
    ]
  }
}

must_not clauses and nested queries

edit

If a nested query matches one or more nested objects in a document, it returns the document as a hit. This applies even if other nested objects in the document don’t match the query. Keep this in mind when using a nested query that contains an inner must_not clause.

Use the inner_hits parameter to see which nested objects matched a nested query.

For example, the following search uses an outer nested query with an inner must_not clause.

PUT my-index
{
  "mappings": {
    "properties": {
      "comments": {
        "type": "nested"
      }
    }
  }
}

PUT my-index/_doc/1?refresh
{
  "comments": [
    {
      "author": "kimchy"
    }
  ]
}

PUT my-index/_doc/2?refresh
{
  "comments": [
    {
      "author": "kimchy"
    },
    {
      "author": "nik9000"
    }
  ]
}

PUT my-index/_doc/3?refresh
{
  "comments": [
    {
      "author": "nik9000"
    }
  ]
}

POST my-index/_search
{
  "query": {
    "nested": {
      "path": "comments",
      "query": {
        "bool": {
          "must_not": [
            {
              "term": {
                "comments.author": "nik9000"
              }
            }
          ]
        }
      }
    }
  }
}

The search returns:

{
  ...
  "hits" : {
    ...
    "hits" : [
      {
        "_index" : "my-index",
        "_id" : "1",
        "_score" : 0.0,
        "_source" : {
          "comments" : [
            {
              "author" : "kimchy"
            }
          ]
        }
      },
      {
        "_index" : "my-index",
        "_id" : "2",
        "_score" : 0.0,
        "_source" : {
          "comments" : [
            {
              "author" : "kimchy"              
            },
            {
              "author" : "nik9000"             
            }
          ]
        }
      }
    ]
  }
}

This nested object matches the query. As a result, the search returns the object’s parent document as a hit.

This nested object doesn’t match the query. Since another nested object in the same document does match the query, the search still returns the parent document as a hit.

To exclude documents with any nested objects that match the nested query, use an outer must_not clause.

POST my-index/_search
{
  "query": {
    "bool": {
      "must_not": [
        {
          "nested": {
            "path": "comments",
            "query": {
              "term": {
                "comments.author": "nik9000"
              }
            }
          }
        }
      ]
    }
  }
}

The search returns:

{
  ...
  "hits" : {
    ...
    "hits" : [
      {
        "_index" : "my-index",
        "_id" : "1",
        "_score" : 0.0,
        "_source" : {
          "comments" : [
            {
              "author" : "kimchy"
            }
          ]
        }
      }
    ]
  }
}
« Joining queries Has child query »