filter_path parameter

Hi I was wondering if the filter_path parameter could be specified when building query, to avoid parsing huge elasticsearch responses

Here is my use case: I have the following query where

• I execute a first term aggregation that returns a large number of buckets
• then execute a metric aggregation on this bucket (avg in this example)
• and finally a sum_bucket aggregation
• When I run this query, the output contains all the intermediary my_huge_bucket.my_huge_bucket_metric results but I am only interested by the sum_bucket metric.

Is there a way to strip the intermediary aggregation from the result?

{
  "size": 0,
  "aggs": {
    "my_sum_bucket": {
      "sum_bucket": {
        "buckets_path": "my_huge_bucket.my_huge_bucket_metric"
      }
    },
    "my_huge_bucket": {
      "terms": {
        "field": "some_field",
        "size": 0
      },
      "aggs": {
        "my_huge_bucket_metric": {
          "avg": {
            "field": "some_field"
          }
        }
      }
    }
  }
}

[spalger]

@ilovett the q parameter is actually the lucene query string, not a url query string.

To specify arbitrary query string parameters all you need to do is add them to the root of the object passed to any api method. So @ilovett 's example would become:

client.search({
  index: 'myindex',
  filter_path: 'aggs.**.sum_bucket',
  body: {
    size: 0,
    aggs: { ... }
  }
})

---

That said, I would also like to add actual support for this and get it listed in the api conventions, so I’ll leave this open

[spalger]

Yeah, that’s a fair point. Thanks for the tip. I’ll probably update the api docs to have the api conventions at the top of each page, rather than on a new page. Unfortunately they are already so big 😕