How to reverse an unwind aggregation?

Question

consider the collection was

{ "_id" : 1, "item" : "ABC1", sizes: [ "S", "M", "L"] }

after unwind

{ "_id" : 1, "item" : "ABC1", "sizes" : "S" }
{ "_id" : 1, "item" : "ABC1", "sizes" : "M" }
{ "_id" : 1, "item" : "ABC1", "sizes" : "L" }

How to reverse this to original structure?

I want to get collection without changing the structure, But I tried group aggregation, it does not provide what I expected .

Actually my problem was, I didn't get original collection after performing unwind and group my input collection was :

{
    "journeys" : {
        "_id" : "8",
        "originDate" : ISODate("2017-06-27T03:55:00.000Z"),
        "destinationDate" : ISODate("2017-06-27T08:55:00.000Z"),


        "dir" : "OUTBOUND",
        "proposals" : [ 
            {
                "_id" : "35",
                "price" : {
                    "posCurrencyPrice" : "SEK 281.00",
                    "defaultCurrencyPrice" : "SEK 281.00"
                },
                "
                "trainClass" : "1st Class",
                "trainClassCode" : "FIRST",

            }, 
            {
                "_id" : "39",
                "price" : {
                    "posCurrencyPrice" : "SEK 377.00",
                    "defaultCurrencyPrice" : "SEK 377.00"
               },
                "trainClass" : "2nd Class",
                "trainClassCode" : "SECOND",


            }, 

}


{
    "journeys" : {
        "_id" : "10",
        "originDate" : ISODate("2017-06-27T04:20:00.000Z"),
        "destinationDate" : ISODate("2017-06-27T10:50:00.000Z"),

         "dir" : "OUTBOUND",
        "proposals" : [ 
            {
                "_id" : "49",
                "price" : {
                    "posCurrencyPrice" : "SEK 490.00",
                    "defaultCurrencyPrice" : "SEK 490.00"
                },

                "trainClass" : "1st Class",
                "trainClassCode" : "FIRST",



            }, 

    }
}

I want to filter out 2nd class from the proposals My expecting output is

{
    "journeys" : {
        "_id" : "8",
        "originDate" : ISODate("2017-06-27T03:55:00.000Z"),
        "destinationDate" : ISODate("2017-06-27T08:55:00.000Z"),


        "dir" : "OUTBOUND",
        "proposals" : [ 
            {
                "_id" : "35",
                "price" : {
                    "posCurrencyPrice" : "SEK 281.00",
                    "defaultCurrencyPrice" : "SEK 281.00"
                },
                "
                "trainClass" : "1st Class",
                "trainClassCode" : "FIRST",

            }, 


}


{
    "journeys" : {
        "_id" : "10",
        "originDate" : ISODate("2017-06-27T04:20:00.000Z"),
        "destinationDate" : ISODate("2017-06-27T10:50:00.000Z"),

         "dir" : "OUTBOUND",
        "proposals" : [ 
            {
                "_id" : "49",
                "price" : {
                    "posCurrencyPrice" : "SEK 490.00",
                    "defaultCurrencyPrice" : "SEK 490.00"
                },

                "trainClass" : "1st Class",
                "trainClassCode" : "FIRST",



            }, 

    }
}

performed following code and results another collection

db.searchResource.aggregate({ "$match" : { "_id" :  ObjectId("5951d7217f4a9810ccca7289")}},
{ "$project" : { "_id" : 0 , "journeys" : "$rssSearchResponse.journeys"}} , { "$unwind" : "$journeys"},
{ "$unwind" : "$journeys.proposals"},{ "$match" : { "journeys.proposals.trainClass" : "1st Class"}},
{$group:{"_id":"$journeys._id","originDate": { "$first": "$journeys.originDate" },
"destinationDate": { "$first": "$journeys.destinationDate" },


"dir": { "$first": "$journeys.dir" },
"proposals":{ "$push" : "$journeys.proposals" }}})

Output was :

 {
        "_id" : "8",
        "originDate" : ISODate("2017-06-27T03:55:00.000Z"),
        "destinationDate" : ISODate("2017-06-27T08:55:00.000Z"),


        "dir" : "OUTBOUND",
        "proposals" : [ 
            {
                "_id" : "35",
                "price" : {
                    "posCurrencyPrice" : "SEK 281.00",
                    "defaultCurrencyPrice" : "SEK 281.00"
                },
                "
                "trainClass" : "1st Class",
                "trainClassCode" : "SECOND",

            }, 






        "_id" : "10",
        "originDate" : ISODate("2017-06-27T04:20:00.000Z"),
        "destinationDate" : ISODate("2017-06-27T10:50:00.000Z"),

         "dir" : "OUTBOUND",
        "proposals" : [ 
            {
                "_id" : "49",
                "price" : {
                    "posCurrencyPrice" : "SEK 490.00",
                    "defaultCurrencyPrice" : "SEK 490.00"
                },

                "trainClass" : "1st Class",
                "trainClassCode" : "FIRST",



            }, 


}

Answer 1

Group is indeed the answer to your question

db.items.aggregate([
    {$match: {}},
    {$unwind: "$sizes"},
    {$group: {
        _id: "$_id",
        sizes: {$push: "$sizes"}
    }}
])