Mongoose populate sub-sub document

Question

I have this setup in my MongoDB

Items:

title: String
comments: [] // of objectId's

Comments:

user: ObjectId()
item: ObjectId()
comment: String

Here's my Mongoose schema:

itemSchema = mongoose.Schema({
    title: String,
    comments: [{ type: Schema.Types.ObjectId, ref: 'comments' }],
});

Item = mongoose.model('items', itemSchema);

commentSchema = mongoose.Schema({
    comment: String,
    user: { type: Schema.Types.ObjectId, ref: 'users' },
});

Comment = mongoose.model('comments', commentSchema);

This is where I get my items along with the comments:

Item.find({}).populate('comments').exec(function(err, data){
    if (err) return handleError(err);
    res.json(data);
});

How do I populate the comments array with it's respective user? Since each comment has a user ObjectId()?

You can do it in two stages as shown in the accepted answer to this question: http://stackoverflow.com/questions/19222520/populate-nested-array-in-mongoose — JohnnyHK, Jun 25 '14 at 17:56

Antenka · Answer 1 · 2016-03-02T09:51:11.820

114

One more way (easier) to do this:

Item
  .find({})
  .populate({
 path:     'comments',   
 populate: { path:  'user',
      model: 'users' }
  })
  .exec(function(err, data){
    if (err) return handleError(err);
    res.json(data);
});

edited Mar 02 '16 at 09:51

answered Nov 12 '15 at 10:21

Antenka

1,519
2
19
29

4

Thanks for your answer. Your answer is great. – Bik Dec 15 '15 at 14:15
2

By the way there is no need for the first populate – Bazinga Mar 02 '16 at 03:58
1

This one was the unique answer that works for me in 4 hours of research, congratulations ! > – Yago Azedias May 17 '18 at 00:24
Way easier and helped to solve my problem! Thank you so much! – Ricardo Silva Aug 28 '18 at 16:26

score 31 · Accepted Answer · answered Jun 26 '14 at 08:46

As a complete example calling populate on the result objects:

Item.find({}).populate("comments").exec(function(err,data) {
    if (err) return handleError(err);

    async.forEach(data,function(item,callback) {
        User.populate(item.comments,{ "path": "user" },function(err,output) {
            if (err) throw err; // or do something

            callback();
        });
    }, function(err) {
        res.json(data);
    });

});

The call to .populate() in the form invoked from the model takes either a document or an array as it's first argument. So you loop through the returned results for each item and call populate this way on each "comments" array. The "path" tells the function what it is matching.

This is done using the "async" version of forEach so it is non-blocking, but generally after all the manipulation all of the items in the response are not only populated with comments but the comments themselves have the related "user" details.

@neil Lunn I am getting below error. Can you please help me with that. async.forEach(data,function(item,callback) { ^ ReferenceError: async is not defined — FarheenP, Jul 08 '15 at 11:52
@FarheenP - install it using `npm install async` and then you can include it by `var async = require('async');`. See https://github.com/caolan/async — petrkotek, Aug 04 '15 at 13:01

score 13 · Answer 3 · answered Apr 17 '16 at 16:58

13

Simpler

Item
  .find({})
  .populate({
    path: 'comments.user',
    model: 'users' }
  })
  .exec(function(err, data){
    if (err) return handleError(err);
    res.json(data);
});

answered Apr 17 '16 at 16:58

steampowered

11,809
12
78
98

score 10 · Answer 4 · answered Feb 10 '18 at 07:58

To add one final method that people may want to use to select only particular fields from sub-documents, you can use the following 'select' syntax:

  Model.findOne({ _id: 'example' })
    .populate({ 
      path: "comments", // 1st level subdoc (get comments)
      populate: { // 2nd level subdoc (get users in comments)
        path: "user",
        select: 'avatar name _id'// space separated (selected fields only)
      }
    })
    .exec((err, res) => { 
        // etc
     });

score 10 · Answer 5 · answered Jul 14 '20 at 12:15

I use this:

.populate({
            path: 'pathName',
            populate: [
                {
                    path: 'FirstSubPathName',
                    model: 'CorrespondingModel',
                },
                {
                    path: 'OtherSubPathName',
                    model: 'CorrespondingModel',
                },
                {
                    path: 'AnotherSubPathName',
                    model: 'CorrespondingModel',
                },
            ]
        });

it's the more easier way that i find to do this.I expect to help. :)

Yash Ojha · Answer 6 · 2018-10-22T10:16:46.917

To populate sub-sub document and populate from multiple schemas

ProjectMetadata.findOne({id:req.params.prjId})
.populate({
    path:'tasks',
    model:'task_metadata',
    populate:{
        path:'assigned_to',
        model:'users',
        select:'name employee_id -_id' // to select fields and remove _id field

    }
})
.populate({
    path:'client',
    model:'client'
})
.populate({
    path:'prjct_mgr',
    model:'users'
})
.populate({
    path:'acc_exec',
    model:'users'
})
.populate({
    path:'prj_type',
    model:'project_type'
}).then ( // .. your thing

or you can do it in following manner ..

   ProjectMetadata.findOne({id:req.params.prjId})
    .populate(
        [{
        path:'tasks',
        model:TaskMetadata,
        populate:[{
            path:'assigned_to',
            model:User,
            select:'name employee_id'
        },
        {
            path:'priority',
            model:Priority,
            select:'id title'
        }],
        select:"task_name id code assign_to stage priority_id"
    },
    {
        path:'client',
        model:Client,
        select:"client_name"
    },
    {
        path:'prjct_mgr',
        model:User,
        select:"name"
    },
    {
        path:'acc_exec',
        model:User,
        select:'name employee_id'
    },
    {
        path:'poc',
        model:User,
        select:'name employee_id'
    },
    {
        path:'prj_type',
        model:ProjectType,
        select:"type -_id"
    }

])

I found the second method (of using array) more useful when I had to get multiple sub-sub documents of same parent.

Vishal patel · Answer 7 · 2019-06-28T13:10:52.397

3

do it try it`s working find Project and get project related populate Task and Perticular Task User find

db.Project.find()
.populate({
    path: 'task',
    populate: { path: 'user_id'}
})
.exec(async(error,results)=>{

})

edited Jun 28 '19 at 13:10

answered Jun 28 '19 at 13:09

Vishal patel

41
6

1

Welcome to SO! Try to format your answer for better readability, this will help other stumbling on the same issue. – Bert H Jun 28 '19 at 13:17

score 1 · Answer 8 · answered Apr 12 '22 at 06:02

1

You can also populate subdocument by this in mongoose -

Item.find({}).populate("comments.user")

answered Apr 12 '22 at 06:02

Tanvir Islam Streame

442
1
5
15

score 0 · Answer 9 · answered May 14 '20 at 20:58

0

Check the screenshot below. This thing works like charm!!!

answered May 14 '20 at 20:58

Al Mobin

19
2

1

[Please don't post screenshots of text](https://meta.stackoverflow.com/a/285557/354577). They can't be searched or copied and offer poor usability. Instead, paste the code as text directly into your question. If you select it and click the `{}` button or Ctrl+K the code block will be indented by four spaces, which will cause it to be rendered as code. – ChrisGPT was on strike May 14 '20 at 22:06

score 0 · Answer 10 · answered May 14 '20 at 22:46

0

This worked for me:

i.e. no need for model

    .populate({
      path: 'filters',
      populate: {
        path: 'tags',
        populate: {
          path: 'votes.user'
        }
      }
    })
    .populate({
      path: 'members'
    })

answered May 14 '20 at 22:46

user2167687

1
1

Mongoose populate sub-sub document

10 Answers10

Linked