JSON parsing - get the average per group with JQ

Question

I have the following JSON data :

{
  "data": [  
    {"date": "2018-08-22","value": 3},
    {"date": "2018-08-30","value": 5},
    {"date": "2018-09-01","value": 5},
    {"date": "2018-09-22","value": 9},
    {"date": "2018-09-28","value": 4},
    {"date": "2018-10-18","value": 2},
    {"date": "2018-10-23","value": 1}
  ]
}

I would like to get the average per month. My expected output is the following :

{
  "data": [  
    {"month": "2018-08","avg": 4},
    {"month": "2018-09","avg": 6},
    {"month": "2018-10","avg": 3}
  ]
}

How can I do that ?

Answer 1

This question is almost identical to

JSON parsing - group by date range with JQ

and the solution given there can easily be adapted to the problem here:

def bucket: sub("-\\d+$"; "");

.data
| reduce .[] as $pair ({};
    .[$pair.date | bucket] += [$pair.value])
| {data: [to_entries[] | {month: .key, avg: (.value| add/length)}]}