How to perform aggregations in Go using mongo-driver

问题

I have a MongoDB collection with 3 fields:

_id: ObjectId
field1: Number
field2: Number

I am doing this aggregation to get "distinct" field1/field2 rows and doing a total count of the results. This works OK in the client (Robo3t):

db.mycollection.aggregate([
    {
        $group: { 
            _id: { field1: "$field1", field2: "$field2" },
        }
    },
    { 
        $group: { 
            _id: null, count: { $sum: 1 } 
        } 
    }
])

Result:

{
    "_id" : null,
    "count" : 57.0
}

How could I do this aggregation in Go using mongo-driver?

There is this method to perform aggregations but the documentation is not clear to me. I understand that I should do some kind of bson query but I don't know where to start.

回答1:

group :=[]bson.M{bson.M{
    "$group": bson.M{
        "_id":bson.M{
            "field1": "$field1",
            "field2": "$field2"
        }
    }
},
bson.M {
    "$group": bson.M{
        "_id":nil,
        "count": bson.M{
            "$sum":1
        }
    }
}
} 

cursor, err := coll.Aggregate(context.Background(), mongo.Pipeline{group})
if err != nil {
    log.Fatal(err)
}

Try the above solution, It will works.

回答2:

To complement the accepted solution I am sharing the full code with the cursor iteration and decode. I changed the "_id": nil to "_id": "" to receive strings and be able to decode to a struct without issues.

pipeline := []bson.M{
    {
        "$group": bson.M{"_id": bson.M{"field1": "$field1", "field2": "$field2"}},
    },
    {
        "$group": bson.M{"_id": "", "count": bson.M{"$sum": 1}},
    },
}

cursor, err := coll.Aggregate(ctx, pipeline)
if err != nil {
    return err
}
type Result struct {
    ID    string `bson:"_id"`
    Count int    `bson:"count"`
}
var res Result
for cursor.Next(ctx) {
    err := cursor.Decode(&res)
    if err != nil {
        return err
    }
    fmt.Printf("Element %v", res)
}

来源：https://stackoverflow.com/questions/59513341/how-to-perform-aggregations-in-go-using-mongo-driver

标签

mongodb