MongoDB: Add date comparison to arrayFilters - mongodb

Question

Objects of my collection have a field, that is an array of objects with one of the field being a string date
{
citizens: [{
name: 'John'
birthday: '1993/07/13'
},
{
name: 'Sarah'
birthday: '1996/07/13'
},
{
name: 'Natalia',
birthday: '2015/07/13'
}]
}
{
citizens: [{
name: 'Leo'
birthday: '1994/02/08'
},
{
name: 'Paul'
birthday: '1934/09/13'
},
{
name: 'Rego',
birthday: '2019/01/29'
}]
}
I want to set to all the users older than 18 status 'adult'
Here is what I try to do:
users.updateMany({}, {
$set: { 'citizens.$[elem].status': 'adult' },
},
{
arrayFilters: [
{ 'elem.status': { $exists: false } },
{ $lt: [{ $toDate: 'elem.$birthday' }, 18yearsaAgoDate] }, <-- 18years don't mean much here, I actually use $$NOW
],
multi: true,
});
But I get 'unknown top level operator: $lt' error when run this. How do I supposed to use $lt in arrayFilter?
Thanks in advance!

Answer 1

Here's how you could do it in a simple update using the aggregation pipelined updates:
db.collection.updateMany({},
[
{
$set: {
citizens: {
$map: {
input: "$citizens",
in: {
$mergeObjects: [
{
status: {
$cond: [
{
$gt: [
{
$dateDiff: {
startDate: {
$toDate: "$$this.birthday"
},
endDate: "$$NOW",
unit: "year"
}
},
18
]
},
"adult",
"$$REMOVE"
]
}
},
"$$this"
]
}
}
}
}
}
])
Mongo Playground
I've used some version 5+ operators like $dateDiff as it makes the code cleaner, but you could still achieve the same results without them using $subtract and a constant for 18 years, like so:
{
$lt: [
{
$toDate: "$$this.birthday"
},
{
$subtract: [
"$$NOW",
567648000000// 18 years in miliseconds
]
}
]
}
Mongo Playground

Answer 2

This is an update using the arrayFilters syntax.
db.collection.updateMany(
{ },
{
$set: { "citizens.$[elem].status": "adult" }
},
{
arrayFilters: [ { "elem.status": { $exists: false } , "elem.birthday": { $lt: "2004/07/27" } } ]
}
)
Note the date value "2004/07/27" is the day 18 years ago (very close approximate value). And using string values in date comparison requires that the value is formatted in "YYYY/mm/dd".

Answer 3

It would have worked like this if your date was already in the right format. Since you need to format it, I think you should use an aggregation pipeline with a $merge stage:
db.collection.aggregate([
{$set: {
citizens: {
$map: {
input: "$citizens",
in: {$mergeObjects: [
{status: {
$cond: [
{$lt: [{$toDate: "$$this.birthday"}, 18yearsaAgoDate]},
"adult",
"$$REMOVE"
]
}
},
"$$this"
]
}
}
}
}
},
{ $merge : { into : "collection" } }
])
See how it works on the playground example