Make query in elasticsearch

Question

I'm learning elastic for a new project. I have a index with format:

{
    "content_id": "bbbbbb",
    "title": "content 2",
    "data": [
            {
              "id": "xx",
              "value": 3,
              "tags": ["a","b","c"]
            },
            {
              "id": "yy",
              "value": 1,
              "tags": ["e","d","c"]
            }
   ]
}

How can i make a query to search contents that have at least one element in data that include tags "a" and "b" ?

thanks so much !!

How can i make query or re design my format data to easy make new query ?

Answer 1

Working with lists and nested documents in Elasticsearch is a bit tricky.

When creating the index, you need to specify the mapping for the nested documents. You can do this by adding a mapping for the data field.

PUT /my_index

{
    "mappings": {
        "doc": {
            "properties": {
                "data": {
                    "type": "nested"
                }
            }
        }
    }
}

If you have already created the index, you should delete it first, then create it again with the mapping.

Before deleting the index, you can reindex (copy) the data to a new index.

Now you can query the data field using the nested query:

GET /my_index/_search

{
    "query": {
        "nested": {
            "path": "data",
            "query": {
                "terms": {
                    "data.tags": [
                        "j",
                        "b",
                        "c"
                    ]
                }
            }
        }
    }
}

Answer 2

If the requirement is to get the document if at least one object in data contains a and b both, then you need to specify the nested mapping as suggested by @ChamRun.

But for getting the results having a and b both, you need to use the below query :

{
    "query": {
        "nested": {
            "path": "data",
            "query": {
                "bool": {
                    "must": [
                        {
                            "term": {
                                "data.tags": "a"
                            }
                        },
                        {
                            "term": {
                                "data.tags": "b"
                            }
                        }
                    ]
                }
            }
        }
    }
}