Elasticsearch sub-aggregation on bucket script - Categorize bucket script result

[nicolas-guerrier]

I would like to categorize a bucket script result.
If I take the elasticsearch example :

POST /sales/_search
{
    "size": 0,
    "aggs" : {
        "sales_per_month" : {
            "date_histogram" : {
                "field" : "date",
                "interval" : "month"
            },
            "aggs": {
                "t-shirts": {
                  "filter": {
                    "term": {
                      "type": "t-shirt"
                    }
                  },
                  "aggs": {
                    "sales": {
                      "sum": {
                        "field": "price"
                      }
                    }
                  }
                },
                "t-shirt-percentage": {
                    "bucket_script": {
                        "buckets_path": {
                          "tShirtSales": "t-shirts>sales"
                        },
                        "script": "if (params.tShirtSales < 10) { 1 } else if(params.tShirtSales >= 10 && params.tShirtSales < 50) { 2 } else { 3 }"
                    }
                }
            }
        }
    }
}

I'm trying to break out three categories according to the number of tshirt sales.
The categories if the number of tshirt sales are:

 1. <10 => Category 1 
 2. 10-50 => Category 2
 3. >50 => Category 3

I would like to know the number of tshirt sales for each category.

I thougt to use a terms aggregations on bucket script to display this category but the bucket script not accept sub aggregation.

Is-it possible ? Have you an idea please ?

Thank you in advance for your response.