GROUP BY performance

Hi,

Requests is the name of the bucket, and requests is the field of the
document.

Here is how our date field of the corresponding document looks like:
“date”: "06/07/2017 22-56-04.648"
With this formatting its simply easier to read the date and time, comparing
to the ISO 8601 timestamp format. If needed, we can definitely change it to
whatever will make CouchBase’s life easier.

Here is the (shortened) example structure of the document:

{
“date”: “06/07/2017 22-56-04.648”,
“country”: “…”,
“zipCode”: “…”,
“id”: “…”,
“name”: “…”,
“requests”: 1
}

So the question is - what index should we prepare in order to run
previously mentioned N1QL query?

Alex