ChatGPT解决这个技术问题 Extra ChatGPT

创建后如何在 mongoose 中填充子文档?

I am adding a comment to an item.comments list. I need to get the comment.created_by user data before I output it in the response. How should I do this?

    Item.findById(req.param('itemid'), function(err, item){
        var comment = item.comments.create({
            body: req.body.body
            , created_by: logged_in_user
        });

        item.comments.push(comment);

        item.save(function(err, item){
            res.json({
                status: 'success',
                message: "You have commented on this item",

//how do i populate comment.created_by here???

                comment: item.comments.id(comment._id)
            });
        }); //end item.save
    }); //end item.find

I need to populate the comment.created_by field here in my res.json output:

                comment: item.comments.id(comment._id)

comment.created_by is a user reference in my mongoose CommentSchema. It currently is only giving me a user id, I need it populated with all the user data, except for password and salt fields.

Here is the schema as people have asked:

var CommentSchema = new Schema({
    body          : { type: String, required: true }
  , created_by    : { type: Schema.ObjectId, ref: 'User', index: true }
  , created_at    : { type: Date }
  , updated_at    : { type: Date }
});

var ItemSchema = new Schema({
    name    : { type: String, required: true, trim: true }
  , created_by  : { type: Schema.ObjectId, ref: 'User', index: true }
  , comments  : [CommentSchema]
});
Could you elaborate the schema
i've pasted what I've tried, res.json()

a
andyvn22

In order to populate referenced subdocuments, you need to explicitly define the document collection to which the ID references to (like created_by: { type: Schema.Types.ObjectId, ref: 'User' }).

Given this reference is defined and your schema is otherwise well defined as well, you can now just call populate as usual (e.g. populate('comments.created_by'))

Proof of concept code:

// Schema
var mongoose = require('mongoose');
var Schema = mongoose.Schema;

var UserSchema = new Schema({
  name: String
});

var CommentSchema = new Schema({
  text: String,
  created_by: { type: Schema.Types.ObjectId, ref: 'User' }
});

var ItemSchema = new Schema({
   comments: [CommentSchema]
});

// Connect to DB and instantiate models    
var db = mongoose.connect('enter your database here');
var User = db.model('User', UserSchema);
var Comment = db.model('Comment', CommentSchema);
var Item = db.model('Item', ItemSchema);

// Find and populate
Item.find({}).populate('comments.created_by').exec(function(err, items) {
    console.log(items[0].comments[0].created_by.name);
});

Finally note that populate works only for queries so you need to first pass your item into a query and then call it:

item.save(function(err, item) {
    Item.findOne(item).populate('comments.created_by').exec(function (err, item) {
        res.json({
            status: 'success',
            message: "You have commented on this item",
            comment: item.comments.id(comment._id)
        });
    });
});

I re-read your item.save() function, this mostly works, except I get all comments in the collection back, I just want the one I added.
Ok, I had a small bug, populate('comments.created_by username') didn't work, but .populate('comments.created_by', 'username') did. Also I noticed I can use -salt -password_hash to not return those fields. Thanks a bunch! accepted answer.
Oh wow! I didn't know you can use - prefix. Thanks for teaching me as well! :)
@chovy can you please update the code with how you did the -salt -passowrd_hash. I searched the mongoose doc, but there was no reference to -salt.
2014 update: Check user1417684's answer below. stackoverflow.com/a/17505554/825980
u
user1417684

This might have changed since the original answer was written, but it looks like you can now use the Models populate function to do this without having to execute an extra findOne. See: http://mongoosejs.com/docs/api.html#model_Model.populate. You'd want to use this inside the save handler just like the findOne is.


This answer is much more correct that the one by @jsalonen. This way you don't have to fetch the document all over again.
References are helpful thanks. An example goes much further.
C
Christian Hartmann

@user1417684 and @chris-foster are right!

excerpt from working code (without error handling):

var SubItemModel = mongoose.model('subitems', SubItemSchema);
var ItemModel    = mongoose.model('items', ItemSchema);

var new_sub_item_model = new SubItemModel(new_sub_item_plain);
new_sub_item_model.save(function (error, new_sub_item) {

  var new_item = new ItemModel(new_item);
  new_item.subitem = new_sub_item._id;
  new_item.save(function (error, new_item) {
    // so this is a valid way to populate via the Model
    // as documented in comments above (here @stack overflow):
    ItemModel.populate(new_item, { path: 'subitem', model: 'subitems' }, function(error, new_item) {
      callback(new_item.toObject());
    });
    // or populate directly on the result object
    new_item.populate('subitem', function(error, new_item) {
      callback(new_item.toObject());
    });
  });

});

N
Naveen Kumar

I faced the same problem,but after hours of efforts i find the solution.It can be without using any external plugin:)

applicantListToExport: function (query, callback) {
  this
   .find(query).select({'advtId': 0})
   .populate({
      path: 'influId',
      model: 'influencer',
      select: { '_id': 1,'user':1},
      populate: {
        path: 'userid',
        model: 'User'
      }
   })
 .populate('campaignId',{'campaignTitle':1})
 .exec(callback);
}