How to match Array elements in Sub-Documents by Criteria - node.js

Question

I have following Mongoose schemas :
EmployeeSchema :
var EmployeeSchema = new Schema({
name : String,
employeeDetailsId: {
type: Schema.Types.ObjectId,
ref: 'employeedetails'
}
});
EmployeeDetailSchema :
var EmployeeDetailSchema = new Schema({
employeeId: {
type: Schema.Types.ObjectId,
ref: 'employee'
},
primarySkills: [
{
type: Schema.Types.ObjectId,
ref: 'skills'
}
],
});
SkillsSchema :
var SkillsSchema = new Schema({
name: {
type: String,
required: true
}
});
EmployeeDetailSchema data gets saved on demand, like when a particular Skill is assigned to Employee. Once EmployeeDetail document is saved then corresponding EmployeeDetailID is saved back to EmployeeSchema as employeeDetailsId.
Now there is bi-directional relationship between EmployeeSchema and EmployeeDetailSchema.
NOTE :
Multiple Skills can be associated to an Employee and they are stored as an array of ObjectID's in EmployeeDetails Schema.
UseCase :
I want to fetch all Employees who have particular Skill associated with them, Skill will be input to the Mongoose / Mongo query.
Say input Skill ID is 1234 then i want to fetch all employees who have Skill id 1234 in EmployeeDetail > PrimarySkills array.
Following is the approach which i tried using Mongoose :
EmployeeModel.aggregate([
{
$lookup: {
from: 'employeedetails',
localField: 'employeeDetailsId',
foreignField: '_id',
as: 'details'
}
},
{
$match: {
$and: [
{ "details.primarySkills": { "$exists": true } },
{
"details.primarySkills": {
$in: [mongoose.Types.ObjectId(req.params.skillId)]
}
}
]
}
}
]).exec(function (err, result) {
if (err) return res.send('400', {
message: 'Unable to fetch employees data by status. Please try again later'
});
return res.jsonp(result);
});
Result : Empty array.
I have no clue where am going wrong, need some help.

Answer 1

My bad, original approach which i followed was all fine except a small mistake. I should have used req.query.skillId instead of req.params.skillId
For those wondering the difference b/w query and params, check this answer
This is the final solution, thought it may help others :
EmployeeModel.aggregate([
{
$lookup: {
from: 'employeedetails',
localField: 'employeeDetailsId',
foreignField: '_id',
as: 'details'
}
},
{
$match: {
$and: [
{ "details.primarySkills": { "$exists": true } },
{
"details.primarySkills": {
$in: [mongoose.Types.ObjectId(req.query.skillId)]
}
}
]
}
}
]).exec(function (err, result) {
if (err) return res.send('400', {
message: 'Unable to fetch employees data by status. Please try again later'
});
return res.jsonp(result);
});

Answer 2

One approach that you could take is apply the $lookup to the Skills model on the EmployeeDetails model first and then do another lookup to the Employee model
EmployeeDetails.aggregate([
{ "$match": { "primarySkills": req.query.skillId } },
{ "$unwind": "$primarySkills" }, // skip this pipeline step for MongoDB 3.4 and above
{
"$lookup": {
"from": "skills",// ensure this is the correct collection name
"foreignField": "_id",
"localField": "primarySkills",
"as": "skills"
}
},
{ "$unwind": "$skills" }
{
"$group": {
"_id": "$_id",
"employeeId": { "$first": "$employeeId" },
"primarySkills": { "$push": "$skills" }
}
},
{
"$lookup": {
"from": "employees",// ensure this is the correct collection name
"foreignField": "_id",
"localField": "employeeId",
"as": "employee"
}
}
]).exec(callback);