$unwind nested document and $match

Question

I have a nested document which looks like:

    var User = new Schema({
    id: String,
    position: [{
        title: String,
           applied:[{
                candidate_id: String,
                name: String
        }],
        }],

What I am looking to do is return all of the 'applied' subdocuments which match a certain 'candidate_id'

What I have so far:

  app.get('/applied', function(req, res){

  var position = "58dc2bd4e7208a3ea143959e";

  User.aggregate(
        {$unwind : "$position"},
        {$unwind : "$position.applied"},
        {$match:{'position.applied.candidate_id': position}}).exec(function (err, result) {
          console.log(result);
        });
        res.render('applied', { title: 'applied',layout:'candidate'});
});

I have another function which returns all the positions that match, and that code works:

app.post('/search', function (req, res) {

  var position = new RegExp(req.body.position, 'i');
  var location = new RegExp(req.body.location, 'i');

  User.aggregate(
        {$unwind : "$position"},
        {$match:{'position.title': position,'position.location':location}}).exec(function (err, result) {
      console.log(result);
      res.send({ results: result });
  });
});

So basically I am struggling with getting a sub-sub-document. Any idea where I'm going wrong?

Sample data:

{
"_id" : ObjectId("58c2871414cd3d209abf5fc9"),
"position" : [ 
    {
        "_id" : ObjectId("58d6b7e11e793c9a506ffe8f"),
        "title" : "Software Engineer",

        "applied" : [ 
            {
                "candidate_id" : "58d153e97e3317291gd80087",
                "name" : "Sample user"

            }, 
            {
                "candidate_id" : "58d153e97e3317291fd99001",
                "name" : "Sample User2"

            }
        ]
    }, 
    {
        "_id" : ObjectId("58c2871414cd3d209abf5fc0"),
        "title" : "Software Engineer",

    }
],

}

What is going on above is there 2 positions, one of which (first entry) has 2 applied candidates, What I need to do is return the nested object if it matches the mongoose query.

What is happening? Also do you have some example data that this is not working properly for? — Explosion Pills, Apr 02 '17 at 13:47
@ExplosionPills The array is coming back empty.. I will update the question to show some sample data! — user, Apr 02 '17 at 13:53

score 1 · Accepted Answer · answered Apr 03 '17 at 10:26

1

Your code seems fine to me I have implemented same and it works for me only possible issue can be that your position="58dc2bd4e7208a3ea143959e" it might be talking it as a string just convert it to objectId by using the following code and check it should work for you.

var mongoose = require('mongoose');

     var position = mongoose.Types.ObjectId("58dc2bd4e7208a3ea143959e");

          User.aggregate(
                {$unwind : "$position"},
                {$unwind : "$position.applied"},
                {$match:{'position.applied.candidate_id': position}}).exec(function (err, result) {
                  console.log(result);
                });
                res.render('applied', { title: 'applied',layout:'candidate'});
        });

answered Apr 03 '17 at 10:26

Shumi Gupta

1,505
2
19
28

I just realised that in my database the candidate_id field was just the number itself without the 'ObjectId' before it, so I am now able to pull the data but only the position array unwinds? the applied array is returned to me in the console as "applied: [Object] ".. and its just returning the entire user :( – user Apr 03 '17 at 11:05
yes, you will get all the data but now your applied array will be an object which will only have the result of the user whose condition matched with your given condition like if your position = "58d153e97e3317291gd80087" in your result you can access that by result[0].position[0].applied.name and it whould be equal to "Sample user" (considering your example) – Shumi Gupta Apr 03 '17 at 11:14
Thank you so much @Shumi Gupta, You have really helped me!! Its working now! :) – user Apr 03 '17 at 11:43

$unwind nested document and $match

1 Answers1