How to include external array inside MongoDB aggregation?

Question

In aggregation project array element from index x to y where x is defined inside collection and y is an array element corresponding to the index of the array element which is matched. Please see below example then it will become quite easier to understand what i am trying to say.

coupons collection

{"coupon_id": "coupon01", "codes": ["FLAT30", "FLAT50", "FLAT70", "FLAT90"], "curr_index": 0}

For example see below example code here i am trying to get coupon codes starting from curr_index to (curr_index + n) where n is the number in coupon_ctrs corresponding to the index of coupons_ids example- for id "584559bd1f65d363bd5d25fd" n is 1, for id "58455a5c1f65d363bd5d2600" n is 2 and for the id "584878eb0005202c64b0cc5d" n is 3.

coupons_ctrs = [1, 2, 3];
coupons_ids = ["584559bd1f65d363bd5d25fd", "58455a5c1f65d363bd5d2600", "584878eb0005202c64b0cc5d"];

int n = 2;
couponmodel.aggregate(
        { $match : { '_id': { $in : coupons_ids }} },
        { $project: {_id:0, codes : /* How to use slice here so that codes array will be returned between cur_index and (curr_index + coupons_ctr corresponding to the index coupon id is found, example- for _id "584559bd1f65d363bd5d25fd" it should be 1 and so on) */} },
        function(err, docs) {
            if (err) {

            } else {

            }
        });

UPDATE

As suggested by Styvane i could use $zip and it would work perfect but as i am using mongoDB 3.2.11 so i can't use it, so what can be the solution for using functionality of $zip in mongodb 3.2.11 ??

Can anyone please tell me how can i include this coupon_ctrs array and use it inside aggregation pipeline.

Sede · Accepted Answer · 2016-12-10 12:11:48Z

1

You need to apply a $slice expression to each element in the "coupons_ctrs" array using the $map operator which means that we use the literal "coupons_ctrs" array as "input" to $map.

let coupons_ids = [
    "584559bd1f65d363bd5d25fd", 
    "58455a5c1f65d363bd5d2600", 
    "584878eb0005202c64b0cc5d"
];

let coupons_ctrs = [1, 2, 3];

db.couponmodel.aggregate(
    [
        { "$match" : { "_id": { "$in" : coupons_ids } } },
        { "$project": { 
            "codes":  { 
                "$map": { 
                    "input": coupons_ctrs, 
                    "as": "n", 
                    "in": { 
                        "$slice": [ 
                            "$codes", 
                            "$curr_index", 
                            { "$add": [ "$curr_index", "$$n" ] } 
                        ]
                    }
                }
            }
        }}
    ]
)

Which yields:

{
    "codes" : [
        [ "FLAT30" ],
        [ "FLAT30", "FLAT50" ],
        [ "FLAT30", "FLAT50",  "FLAT70" ]
    ]
}

In MongoDB 3.4 we can use the $zip operator to do this:

db.couponmodel.aggregate(     
    [ 
        { "$project": {
            "codes":  {
                "$map": {
                    "input": { 
                        "$zip": { 
                            "inputs": [ coupons_ids, coupons_ctrs ] 
                        } 
                    }, 
                    "as": "item",
                    "in": {  
                        "coupon_id": { "$arrayElemAt": [ "$$item", 0 ] },
                        "value": {          
                            "$slice": [  
                                "$codes",       
                                "$curr_index",    
                                { "$add": [ 
                                    "$curr_index", 
                                    { "$arrayElemAt": [ "$$item", 1 ] } 
                                ] }                          
                            ]
                         }
                    }     
                } 
            } 
        }}     
    ] 
)

which return something like this:

{
    "codes" : [
        {
             "coupon_id" : "584559bd1f65d363bd5d25fd",
             "value" : [ "FLAT30" ]
        },
        {
             "coupon_id" : "58455a5c1f65d363bd5d2600",
             "value" : [ "FLAT30", "FLAT50" ],
        },
        {
             "coupon_id" : "584878eb0005202c64b0cc5d",
             "value" : [ "FLAT30", "FLAT50",  "FLAT70" ]
        }
    ]
}

edited Dec 10, 2016 at 12:11

answered Dec 10, 2016 at 7:35

Sede

61.5k20 gold badges158 silver badges162 bronze badges

Sign up to request clarification or add additional context in comments.

10 Comments

Prakash Kumar Over a year ago

In your code where it is defined that for id: "584559bd1f65d363bd5d25fd" n will be 1, i mean is there any surety that for each id that is matching "n" will be the same index element in "coupon_ctr". ??

Sede Over a year ago

@PrakashKumar yes it will but you need to provide the "coupons_ids" elements index need to be in the same order as the "coupons_ctrs" elements as you've shown.

Prakash Kumar Over a year ago

yes it is in smae order but how can this code intenrally do that can you please explain ??

Sede Over a year ago

@PrakashKumar did you run the code? Is it working the way you expect?

Prakash Kumar Over a year ago

sorry but not for first id i am getting three codes [["GETFREE"],["GETFREE"],["GETFREE"]], for second id i am getting [["FET50"],["FET50"],["FET50"]] and for third i am getting [["versace"],["versace","versace"],["versace","versace"]].

|

Collectives™ on Stack Overflow

How to include external array inside MongoDB aggregation?

1 Answer 1

10 Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

10 Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related