jq: avoid nested array mapping - jq

Question

Here's my document:
{
"_id": "a6c8004c-efdc-4462-b474-31df2d417be3",
"address": [
{
"city": {
"value": "2505300"
},
"state": {
"value": "25"
},
"postalCode": {
"value": "25132"
}
}
]
}
{
"_id": "aa4b4ba9-32e3-4384-8cd8-0852bae0029b",
"address": [
{
"city": {
"value": "2521700"
},
"state": {
"value": "25"
},
"postalCode": {
"value": "25300"
}
}
]
}
Currently, I'm applying this JQ filter:
def pick_address:
{
city: .city | .value,
postalCode: .postalCode | .value,
state: .state | .value
};
map(.address | map(pick_address))
I'm getting this:
[
[
{
"city": "2505300",
"postalCode": "25132",
"state": "25"
}
],
[
{
"city": "2521700",
"postalCode": "25300",
"state": "25"
}
]
]
As you can see, I'm getting above objects with nested array.
I need to get it as:
[
{
"city": "2505300",
"postalCode": "25132",
"state": "25"
},
{
"city": "2521700",
"postalCode": "25300",
"state": "25"
}
]
I don't quite figure out what's wrong.
Any ideas?

Answer 1

Instead of mapping pick_address onto the address array
map(.address | map(pick_address))
just apply it directly onto its elements:
map(.address[] | pick_address)
Demo
On a side note: If within pick_address the operation .value is being applied to all fields, then def pick_address: map_values(.value); would be equivalent. Demo

Answer 2

More dynamic solution without any hardcoded keys, using
with_entries()
--slurp
jq --slurp 'map(.address[] | with_entries(.value |= .value))'
Will produce
[
{
"city": "2505300",
"state": "25",
"postalCode": "25132"
},
{
"city": "2521700",
"state": "25",
"postalCode": "25300"
}
]
As you can test in this online demo

Answer 3

Isn't it simpler to just do this?
jq '.[].address[] | [{city: .city.value, state: .state.value, postalCode: .postalCode.value}]'