$project in $lookup mongodb

Question

I have a query, that use $lookup to "join" two models, after this i use $project to select olny the fields that i need, but my $project brings an arrray of objects (user_detail) that contains more data that i need. I want only two fields (scheduleStart and scheduleEnd) of my result.

My query:

 User.aggregate([{
      $match: {
        storeKey: req.body.store,      
      }
    },
    {
      $group: {
        _id: {
          id: "$_id",
          name: "$name",
          cpf: "$cpf",      
          phone: "$phone",
          email: "$email",
          birthday: "$birthday",
          lastName: "$lastname"      
        },
        totalServices: {
          $sum: "$services"
        },    
      }
    },
    {
      $lookup: {
        from: "schedules",
        localField: "_id.phone",
        foreignField: "customer.phone",
        as: "user_detail"
      }  
    },  
    {
      $project: {
        _id: 1,
        name: 1,
        name: 1,
        cpf: 1,      
        phone: 1,
        email: 1,
        birthday: 1,
        totalServices: 1,
        totalValue: { $sum : "$user_detail.value" },
        count: {
          $sum: 1
        },
        user_detail: 1
      }
    },

Result of query:

count: 1
totalServices: 0
totalValue: 73
user_detail: Array(2)
0:
...
paymentMethod: 0
paymentValue: "0"
scheduleDate: "2018-10-02"
scheduleEnd: "2018-10-02 08:40"
scheduleStart: "2018-10-02 08:20"
status: 3
store: "5b16cceb56a44e2f6cd0324b"
updated: "2018-11-27T13:30:21.116Z"
1:
...
paymentMethod: 0
paymentValue: "0"
scheduleDate: "2018-11-27"
scheduleEnd: "2018-11-27 00:13"
scheduleStart: "2018-11-27 00:03"
status: 2
store: "5b16cceb56a44e2f6cd0324b"
updated: "2018-11-27T19:33:39.498Z"
_id:
birthday: "1992-03-06"
email: "[email protected]"
id: "5bfed8bd70de7a383855f09e"
name: "Chris Santos G"
phone: "11969109995"
...

Result that i need:

count: 1
totalServices: 0
totalValue: 73
user_detail: Array(2)
0:
scheduleEnd: "2018-10-02 08:40"
scheduleStart: "2018-10-02 08:20"
1:
scheduleEnd: "2018-11-27 00:13"
scheduleStart: "2018-11-27 00:03"

_id:
birthday: "1992-03-06"
email: "[email protected]"
id: "5bfed8bd70de7a383855f09e"
name: "Chris Santos G"
phone: "11969109995"
...

How can i do this with my query?

Solution 1

You can use $lookup 3.6 syntax to $project the fields inside the $lookup pipeline

User.aggregate([
  { "$lookup": {
    "from": "schedules",
    "let": { "id": "$_id.phone" },
    "pipeline": [
      { "$match": { "$expr": { "$eq": ["$customer.phone", "$$id"] }}},
      { "$project": { "scheduleStart": 1, "scheduleEnd": 1 }}
    ],
    "as": "user_detail"
  }}
])

Solution 2

For version of mongo version > 3.6 this query should work for you:

 User.aggregate([{
      $match: {
        storeKey: req.body.store,      
      }
    },
    {
      $group: {
        _id: {
          id: "$_id",
          name: "$name",
          cpf: "$cpf",      
          phone: "$phone",
          email: "$email",
          birthday: "$birthday",
          lastName: "$lastname"      
        },
        totalServices: {
          $sum: "$services"
        },    
      }
    },
    {
      $lookup: {
        from: "schedules",
        localField: "_id.phone",
        foreignField: "customer.phone",
        as: "user_detail"
      }  
    },  
    {
      $project: {
        _id: 1,
        name: 1,
        name: 1,
        cpf: 1,      
        phone: 1,
        email: 1,
        birthday: 1,
        totalServices: 1,
        totalValue: { $sum : "$user_detail.value" },
        count: {
          $sum: 1
        },
        user_detail: {
            scheduleEnd: 1,
            scheduleStart: 1,
        }
      }
    },

Comments

[Adam Reis]

Having to add all the parent fields to the projection is such a pain though...