How to reduce similar elements of array

[dparmar]

I am struggling to figure out a way to aggregate sample values based on name key. Appreciate your help!

Input:

- name: api
  sample:
    count: 20
    rate: 2
- name: api
  sample:
    count: 30
    rate: 3
- name: data
  sample:
    count: 40
    rate: 4
- name: data
  sample:
    count: 50
    rate: 5
- name: users
  sample:
    value: 2
- name: users
  sample:
    value: 3

Expected output:

- name: api
  sample:
    count: 50
    rate: 5
- name: data
  sample:
    count: 90
    rate: 9
- name: users
  sample:
    value: 5

[mikefarah]

This was a fun one - but it did require a nested reduce. I've got the answer in a .yq file so that I can comment on what's happening:

sum_samples.yq

#!./yq

.[] as $i ireduce({}; # reduce over each entry, into a new map
    .[$i.name].name = $i.name |  # set the name to the $i entry

    .[$i.name].sample |= ( # update the sample by looping over each $i sample and adding it
        ($i.sample | to_entries[]) as $k ireduce(.; .[$k.key] += $k.value)
    )

) | [.[]] # convert the map into an array of its values

Then you can run it:

./sum.yq data1.yaml

See reduce for more info.