Why does aggregation contain ID's that don't exist in the query results?

Elastic Stack Elasticsearch
1

To gather data on how many documents have been prepared by some employee in a given time period, I have written this query:

{
  "from": 0,
  "size": 0,
  "sort": [],
  "query": {
    "bool": {
      "must": [
        {
          "nested": {
            "path": "preparers",
            "query": {
              "terms": {
                "preparers.employee.id": [
                  -1097
                ]
              }
            }
          }
        },
        {
          "range": {
            "regDate": {
              "gte": "22.02.2023.",
              "format": "dd.MM.yyyy."
            }
          }
        },
        {
          "range": {
            "regDate": {
              "lte": "24.02.2023.",
              "format": "dd.MM.yyyy."
            }
          }
        }
      ]
    }
  },
  "aggs": {
    "body": {
      "nested": {
        "path": "preparers"
      },
      "aggs": {
        "body": {
          "terms": {
            "field": "preparers.employee.id",
            "size": 1000,
            "order": {
              "_count": "desc"
            },
            "min_doc_count": 0
          }
        }
      }
    }
  }
}

For the chosen employee the query returns correct data. However, it also returns aggregations for ID's that aren't even in the query results. A sample:

{
    "took": 6,
    "timed_out": false,
    "_shards": {
        "total": 3,
        "successful": 3,
        "skipped": 0,
        "failed": 0
    },
    "hits": {
        "total": 6,
        "max_score": 0,
        "hits": []
    },
    "aggregations": {
        "body": {
            "doc_count": 13,
            "body": {
                "doc_count_error_upper_bound": 0,
                "sum_other_doc_count": 0,
                "buckets": [
                    {
                        "key": -1097,
                        "doc_count": 6
                    },
                    {
                        "key": -1208,
                        "doc_count": 3
                    },
                    {
                        "key": -1210,
                        "doc_count": 0
                    },
                    {
                        "key": -1602,
                        "doc_count": 0
                    },
                    {
                        "key": -1233,
                        "doc_count": 0
                    },
                    {
                        "key": 204382,
                        "doc_count": 0
                    }
                ]
            }
        }
    }
}

The ID's with doc_count 0 don't exist in the results. Have I written this query wrong, or why are those non-existent ID's added in the aggregation?

2

Hi @A_K3

I suspected that document has inside "preparers" a list employee id with values -1097, -1208, -1210.

When you filter by the ID of employee x, the document is returned with all employee ids inside preparers and therefore in its aggregation the final result has several ids.