Match and merge from multiple files with jq - json

Question

I have a directory with a bunch of "logins" json files like this:
[
{
"user_id": "5ce722b803b54f03f745cdf45d579920",
"time": "2019-10-29T20:03:18.894006Z"
},
{
"user_id": "5ce722b858f3e80e6e85aad3113a1665",
"time": "2019-10-29T20:11:32.4843Z"
}
]
In another directory I have a bunch of "users" json files like this:
[
{
"id": "5ce722b803b54f03f745cdf45d579920",
"email": "foo#gmail.com",
"first_name": "John",
"last_name": "Doe",
"enabled": true,
"created_at": "2019-06-13T17:07:17.2925Z",
"updated_at": "2019-06-13T17:15:20.903085Z",
"groups": {
"count": 1,
"shortlist": [
{
"id": "5d0282c5d5d6063286140e864a0c6506",
"name": "cool users",
"description": "cool users",
"locked": true
}
]
},
"avatar": "",
"role_id": "5d0282c488bba9ebc62df8b3c38571a9",
"company_uid": ""
},
{
"id": "5d0284fdec62d47039e7119013b0aa2c",
"email": "bar#gmail.com",
"first_name": "Jane",
"last_name": "Doe",
"enabled": true,
"created_at": "2019-06-13T17:16:45.210018Z",
"updated_at": "2019-06-13T17:16:45.210018Z",
"groups": {
"count": 1,
"shortlist": [
{
"id": "5d0282c5d5d6063286140e864a0c6506",
"name": "cool users",
"description": "cool users",
"locked": true
}
]
},
"avatar": "",
"role_id": "5d0282c488bba9ebc62df8b3c38571a9",
"company_uid": ""
}
]
What I'm trying to do with jq is:
For each user_id in the "Logins" files, I want to find a matching id in the "Users" files.
I want to merge those two object together.
The intended outcome is another json file(s), which contains login and corresponding user data. As a bonus, I only want the email first and last name from "Users".
End result would be something like this:
[
{
"user_id": "5ce722b803b54f03f745cdf45d579920",
"time": "2019-10-29T20:03:18.894006Z",
"email": "foo#gmail.com",
"first_name": "John",
"last_name": "Doe"
}
}
I've tried variations of the below, but end up with what looks like an infinite loop or something. I know my for loops are wrong, just not sure how to work with multiple files like this.
lastlogins="/last10/*.json"
users="/users/*.json"
for ll in $lastlogins; do
for user in $users; do
userid=$(jq -r '.[].user_id' $ll)
jq -c --arg userid "$userid" '.[] | select(.id == $userid)' $user
done
done

Answer 1

There is no need to use shell looping. That is, everything can be done using jq.
For example, using the invocation:
jq -f users-logins.jq --argfile users users.json logins.json
where users-logins.jq contains:
INDEX($users[]; .id) as $udict
| map( if $udict[.user_id] then . + $udict[.user_id] else empty end)
| map( {user_id, time, email, first_name, last_name} )
the output using the sample inputs would be:
[
{
"user_id": "5ce722b803b54f03f745cdf45d579920",
"time": "2019-10-29T20:03:18.894006Z",
"email": "foo#gmail.com",
"first_name": "John",
"last_name": "Doe"
}
]