Add field (boolean) to returned objects, when a specified value is in array, without including the array itself

Question

I have a mongoose Schema that looks likes this :

var AnswerSchema = new Schema({
  author: {type: Schema.Types.ObjectId, ref: 'User'},
  likes: [{type: Schema.Types.ObjectId, ref: 'User'}]
  text: String,
  ....
});

and I have an API endpoint that allow to get answers posted by a specific user (which exclude the likes array). What I want to do is add a field (with "true/false" value for example) to the answer(s) returned by the mongoose query, when a specific user_id is (or is not) in the likes array of an answer. This way, I can display to the user requesting the answers if he already liked an answer or not.

How could I achieve this in an optimised way ? I would like to avoid fetching the likes array, then look into it myself in my Javascript code to check if specified userId is present in it, then remove it before sending it back to the client... because it sounds wrong to fetch all this data from mongoDB to my node app for nothing. I'm sure there is a better way by using aggregation but I never used it and am a bit confused on how to do it right. The database might grow very large so it must be quick and optimised.

Answer 1

One approach you could take is via the aggregation framework which allows you to add/modify fields via the $project pipeline, applying a host of logical operators that work in cohort to achieve the desired end result. For instance, in your above case this would translate to:

Answer.aggregate()
    .project({
        "author": 1,            
        "matched": {
            "$eq": [ 
                { 
                    "$size": { 
                        "$ifNull": [
                            { "$setIntersection": [ "$likes", [userId] ] }, 
                            []
                        ] 
                    } 
                },
                1
            ]
        }
    })
    .exec(function (err, docs){
        console.log(docs);
    })

As an example to test in mongo shell, let's insert some few test documents to the test collection:

db.test.insert([
    {
        "likes": [1, 2, 3]
    },
    {
        "likes": [3, 2]
    },
    {
        "likes": null
    },
    {
        "another": "foo"
    }
])

Running the above aggregation pipeline on the test collection to get the boolean field for userId = 2:

var userId = 2;
db.test.aggregate([
    { 
        "$project": {               
            "matched": {
                "$eq": [ 
                    { 
                        "$size": { 
                            "$ifNull": [
                                { "$setIntersection": [ "$likes", [userId] ] }, 
                                []
                            ] 
                        } 
                    },
                    1
                ]
            }
        }
    }
])

gives the following output:

{
    "result" : [ 
        {
            "_id" : ObjectId("564f487c7d3c273d063cd21e"),
            "matched" : true
        }, 
        {
            "_id" : ObjectId("564f487c7d3c273d063cd21f"),
            "matched" : true
        }, 
        {
            "_id" : ObjectId("564f487c7d3c273d063cd220"),
            "matched" : false
        }, 
        {
            "_id" : ObjectId("564f487c7d3c273d063cd221"),
            "matched" : false
        }
    ],
    "ok" : 1
}