mongodb query findById() and month and year - mongodb

Question

I've searched for an answer on how to solve my problem but found nothing, so I am very sorry if I am repeating a question that has been asked before.
i'm trying to find a results for a specific userId by month and year.
the dates has been stored in db in this format : yyyy-mm-dd.
I'm trying the following query with Mongo v3.6.8:
db.collection.find({
$and: [
{
$expr: {
$eq: [
{
$month: {
$dateFromString: {
dateString: "$Ntry.BookgDt",
format: "%Y-%m-%d"
}
}
},
12
]
}
},
{
$expr: {
$eq: [
{
$year: {
$dateFromString: {
dateString: "$Ntry.BookgDt",
format: "%Y-%m-%d"
}
}
},
2020
]
}
},
{
$expr: {
$eq: [
"id",
"5fab9a66c493dc4a3c49a7a3"
]
}
}
]
})
Sample data:
[
{
"userid": "5fab9a66c493dc4a3c49a7a3",
"name": "user name",
"acc": "admin",
"Blas": "00.00",
"Ntry": [
{
"Amt": "11.72",
"BookgDt": "2020-08-16",
},
{
"Amt": "16.72",
"BookgDt": "2020-06-23",
}
]
},
{
"userid": "5fab9a77c493dc4a3c49a7a3",
"name": "user name",
"acc": "user",
"Blas": "00.00",
"Ntry": [
{
"Amt": "11.72",
"BookgDt": "2020-08-23",
},
{
"Amt": "16.72",
"BookgDt": "2020-07-23",
}
]
}
]
so my query is to find all Ntry for UserId 5fab9a66c493dc4a3c49a7a3 in month 8 and year 2020, but I got this erorr:
query failed: (ConversionFailure) $dateFromString requires that 'dateString' be a string, found: array with value ["2020-08-16", "2020-06-23"]
can you please help me to find the suitable query, and thank you in advance.
here is also a mongo play ground link, It's the best for quick editing:

Answer 1

This pipeline should work:
db.foo.aggregate([
// Top level match
{$match: {userid: "5fab9a66c493dc4a3c49a7a3"}}
// Next, only keep entries in the Ntry array with month 8 and year 2020.
// addFields: {"Ntry": {$filter: {input: "$Ntry"}}} means overwrite the
// original Ntry array.
,{$addFields: {"Ntry": {$filter: {
input: "$Ntry",
as: "zz",
cond: { $and: [
{$eq: [{$month:{$dateFromString:{dateString:"$$zz.BookgDt",format: "%Y-%m-%d"}}}\
, 8] },
{$eq: [{$year: {$dateFromString:{dateString:"$$zz.BookgDt",format: "%Y-%m-%d"}}}\
, 2020] }
]}
}}
}}
// It is possible everything got filtered out of the Ntry array, leaving
// an empty (size 0) array. We likely do not want that, so further
// cut down the output material. You can comment this out to see what
// changes, especially if you change the month and year targets above.
,{$match: {$expr: {$ne: [ {$size: "$Ntry"}, 0] } }}
]);
It's probably simpler to call $dateFromString twice but if you are feeling adventurous, then use $let inside the cond to convert the date just once:
db.foo.aggregate([
{$match: {userid: "5fab9a66c493dc4a3c49a7a3"}}
,{$addFields: {"Ntry": {$filter: {
input: "$Ntry",
as: "zz",
cond: {
$let: {
vars: {dd: {$dateFromString:{dateString:"$$zz.BookgDt",format: "%Y-%m-%d"}}},
in: {
$and: [
{$eq: [{$month: "$$dd"}, 8] },
{$eq: [{$year: "$$dd"}, 2020] }
]
}
}
}
}}
}}
,{$match: {$expr: {$ne: [ {$size: "$Ntry"}, 0] } }}
]);