Elasticsearch: Is there a way to divide _count value from bucket within a sampler aggregation with p-CodePudding

I'm struggling to do basic operation with some aggregations. The code below is something i imagined it could works, but it does not. What i wanted to accomplish is to divide each "value_number" which is _count per category bucket with some absolute number (it can be parameter but it does not matter ). So, is there a way to accomplish this? Thanks in advance!

"aggs": {
    "test": {
      "sampler": {
        "shard_size": 1000
      },
      "aggs": {
        "categories": {
          "terms": {
            "field": "category",
            "size": 150
          }
        },
        "bucketresults": {
          "bucket_script": {
            "buckets_path": {
              "value_number": "categories>_count"
            },
            "script": "value_number/100"
          }
        }
      }
    }
  }

CodePudding user response：

You can do using below query. In your current query, you need to move bucketresults as inner aggregation for categories. Also, when you access value_number then you need to access using params.value_number.

{
  "aggs": {
    "test": {
      "sampler": {
        "shard_size": 1000
      },
      "aggs": {
        "categories": {
          "terms": {
            "field": "category",
            "size": 150
          },
          "aggs": {
            "bucketresults": {
              "bucket_script": {
                "buckets_path": {
                  "value_number": "_count"
                },
                "script": "params.value_number/100"
              }
            }
          }
        }
      }
    }
  }
}

Below will be response:

"aggregations": {
    "test": {
      "doc_count": 5,
      "categories": {
        "doc_count_error_upper_bound": 0,
        "sum_other_doc_count": 0,
        "buckets": [
          {
            "key": "apple",
            "doc_count": 5,
            "bucketresults": {
              "value": 0.05
            }
          },
          {
            "key": "iphone",
            "doc_count": 4,
            "bucketresults": {
              "value": 0.04
            }
          },
          {
            "key": "mobile",
            "doc_count": 4,
            "bucketresults": {
              "value": 0.04
            }
          }
        ]
      }
    }
  }