json - query with conditions - json

Question

I have following json:
{
"id": "1",
"name": "profile1",
"userId": "0",
"groupId": "3",
"attributes": [
{
"id": "104",
"name": "Enable",
"value": "1"
},
{
"id": "105",
"name": "TargetNode",
"value": "system1"
},
{
"id": "106",
"name": "Timeout",
"value": "30"
}
],
"xconns": [
{
"id": "1",
"locked": false,
"attributeList": [
{
"id": "101",
"name": "Lgrp",
"value": "1"
},
{
"id": "102",
"name": "IsRem",
"value": "1"
},
{
"id": "103",
"name": "Media",
"value": "1"
}
]
},
{
"id": "1",
"locked": false,
"attributeList": [
{
"id": "101",
"name": "Lgrp",
"value": "1"
},
{
"id": "102",
"name": "IsRem",
"value": "1"
},
{
"id": "103",
"name": "Media",
"value": "1"
}
]
},
{
"id": "1",
"locked": false,
"attributeList": [
{
"id": "101",
"name": "Lgrp",
"value": "1"
},
{
"id": "102",
"name": "IsRem",
"value": "1"
},
{
"id": "103",
"name": "Media",
"value": "1"
}
]
}
]
}
{
"id": "2",
"name": "profile2",
"userId": "7",
"groupId": "0",
"attributes": [
{
"id": "104",
"name": "Enable",
"value": "1"
},
{
"id": "105",
"name": "TargetNode",
"value": "system2"
},
{
"id": "106",
"name": "Timeout",
"value": "30"
}
],
"xconns": [
{
"id": "2",
"locked": false,
"attributeList": [
{
"id": "101",
"name": "Lgrp",
"value": "1"
},
{
"id": "102",
"name": "IsRem",
"value": "1"
},
{
"id": "103",
"name": "Media",
"value": "1"
}
]
},
{
"id": "2",
"locked": false,
"attributeList": [
{
"id": "101",
"name": "Lgrp",
"value": "1"
},
{
"id": "102",
"name": "IsRem",
"value": "1"
},
{
"id": "103",
"name": "Media",
"value": "1"
}
]
},
{
"id": "2",
"locked": false,
"attributeList": [
{
"id": "101",
"name": "Lgrp",
"value": "1"
},
{
"id": "102",
"name": "IsRem",
"value": "1"
},
{
"id": "103",
"name": "Media",
"value": "1"
}
]
}
]
}
I can filter following:
$ jq -r 'select([.attributes[] | .name == "TargetNode" ] | any ) | [{userId, groupId, id, name}] | .[] | if (.userId == "0") then del(.userId) else . end | if (.groupId == "0") then del(.groupId) else . end | to_entries | map("\(.key | ascii_upcase):\(.value)") | #tsv' file.json
GROUPID:3 ID:1 NAME:profile1
USERID:7 ID:2 NAME:profile2
I need to add also value of TargetNode:
GROUPID:3 ID:1 NAME:profile1 TARGETNODE:system1
USERID:7 ID:2 NAME:profile2 TARGETNODE:system2
is there a way to include it in
[{userId, groupId, id, name, TargetNode}]
to get the value of TargetNode and not null?
GROUPID:3 ID:1 NAME:profile1 TARGETNODE:null
USERID:7 ID:2 NAME:profile2 TARGETNODE:null
Update:
the solution provided by RomanPerekhrest is nearly ok, but there is one issue because the json file in real is much bigger, there are more attrobutes in "main secttion", for example:
{
"id": "1",
"name": "profile1",
"userId": "0",
"groupId": "3",
"attrib101": "A",
"attrib102": "B",
"attributes": [
...
...
it is cousing that RomanPerekhrest's jq filter returns too much...
how to rid of them too?
ID:1 NAME:profile1 GROUPID:3 ATTRIB101:A ATTRIB102:B TARGETNODE:system1
ID:2 NAME:profile2 USERID:7 ATTRIB101:C ATTRIB102:D TARGETNODE:system2

Answer 1

jq solution:
jq -r '.attributes |= map(select(.name == "TargetNode"))
| if (.attributes | length != 0) then .targetNode = .attributes[0].value else . end
| if (.userId == "0") then del(.userId) else . end
| if (.groupId == "0") then del(.groupId) else . end
| del(.attributes, .xconns) | to_entries
| map("\(.key | ascii_upcase):\(.value)") | #tsv' file.json
If an object with "name": "TargetNode" pair not exists - TARGETNODE won't be added into resulting structure
The output:
ID:1 NAME:profile1 GROUPID:3 TARGETNODE:system1
ID:2 NAME:profile2 USERID:7 TARGETNODE:system2