从不同文档的数组中获取多个对象MongoDb

Question

My collection coll is

/* 1 */
{
    "_id" : ObjectId("566121aa4b88d840eb7d1c50"),
    "batchCourseId" : ObjectId("566122ab94b792fbdf81bcf3"),
    "array" : [ 
        {
            "id" : 1
        }, 
        {
            "id" : 2
        }, 
        {
            "id" : 3
        }, 
        {
            "id" : 4
        }
    ]
}

/* 2 */
{
    "_id" : ObjectId("5661224a4b88d840eb7d1c51"),
    "batchCourseId" : ObjectId("566122ab94b792fbdf81bcf3"),
    "array" : [ 
        {
            "id" : 1
        }, 
        {
            "id" : 7
        }, 
        {
            "id" : 3
        }, 
        {
            "id" : 5
        }
    ]
}

what i need is to pull objects in array 'array' where batchCourseId = ObjectId("566122ab94b792fbdf81bcf3")

and 2<array.id<=5

expected output is

/* 1 */
    {
        "_id" : ObjectId("566121aa4b88d840eb7d1c50"),
        "array" : [             
            {
                "id" : 3
            }, 
            {
                "id" : 4
            }
        ]
    }

/* 2 */
{
    "_id" : ObjectId("5661224a4b88d840eb7d1c51"),
    "array" : [            
        {
            "id" : 3
        }, 
        {
            "id" : 5
        }
    ]
}

already tried

db.coll.find({"batchCourseId" : ObjectId("566122ab94b792fbdf81bcf3")},
    { array: { $elemMatch: { id: { $gt: 2,$lte: 5} } } })

the output is like

/* 1 */
{
    "_id" : ObjectId("566121aa4b88d840eb7d1c50"),
    "array" : [ 
        {
            "id" : 3
        }
    ]
}

/* 2 */
{
    "_id" : ObjectId("5661224a4b88d840eb7d1c51"),
    "array" : [ 
        {
            "id" : 3
        }
    ]
}

close but only the first matching object in array is in result

FYI this only a sample set of data the original data is more complex and big in count

so pls let me know the best practice to do this, performance is also important

thanks in advance

Answer 1

You can use aggregation for achieving the same. A sample is shown below:

db.coll.aggregate(
  {$match: {"batchCourseId" : ObjectId("566122ab94b792fbdf81bcf3")}},
  {$unwind: '$array'},
  {$match: {'array.id': { $gt: 2,$lte: 5}}},
  {$group: {_id: '$_id', array: {$push : '$array'}}}
)

Result:

{ "_id" : ObjectId("5661224a4b88d840eb7d1c51"), "array" : [ { "id" : 3 }, { "id" : 5 } ] }
{ "_id" : ObjectId("566121aa4b88d840eb7d1c50"), "array" : [ { "id" : 3 }, { "id" : 4 } ] }

Answer 2

In MongoDB aggregation $unwind creates Cartesian_product problem so in large data set is good way to avoid $unwind .

Let's check with your example if you use $unwind in aggregation then result looks like this

db.collectionName.aggregate([
    { "$match": { "batchCourseId": ObjectId("566122ab94b792fbdf81bcf3") }}, 
    { "$unwind": "$array" }
])

so result of above query is :

{ "_id" : ObjectId("566121aa4b88d840eb7d1c50"), "batchCourseId" : ObjectId("566122ab94b792fbdf81bcf3"), "array" : { "id" : 1 } }
{ "_id" : ObjectId("566121aa4b88d840eb7d1c50"), "batchCourseId" : ObjectId("566122ab94b792fbdf81bcf3"), "array" : { "id" : 2 } }
{ "_id" : ObjectId("566121aa4b88d840eb7d1c50"), "batchCourseId" : ObjectId("566122ab94b792fbdf81bcf3"), "array" : { "id" : 3 } }
{ "_id" : ObjectId("566121aa4b88d840eb7d1c50"), "batchCourseId" : ObjectId("566122ab94b792fbdf81bcf3"), "array" : { "id" : 4 } }
{ "_id" : ObjectId("5661224a4b88d840eb7d1c51"), "batchCourseId" : ObjectId("566122ab94b792fbdf81bcf3"), "array" : { "id" : 1 } }
{ "_id" : ObjectId("5661224a4b88d840eb7d1c51"), "batchCourseId" : ObjectId("566122ab94b792fbdf81bcf3"), "array" : { "id" : 7 } }
{ "_id" : ObjectId("5661224a4b88d840eb7d1c51"), "batchCourseId" : ObjectId("566122ab94b792fbdf81bcf3"), "array" : { "id" : 3 } }
{ "_id" : ObjectId("5661224a4b88d840eb7d1c51"), "batchCourseId" : ObjectId("566122ab94b792fbdf81bcf3"), "array" : { "id" : 5 } }

this create multiple documents and in large documents in collections it slow the performance and increase processing time.

Instead of $unwind use $map in aggregation with aggregation-set operator and the query is as below :

db.collection.aggregate([{
  "$match": {
    "batchCourseId": ObjectId("566122ab94b792fbdf81bcf3")
  }
}, {
  "$project": {
    "array": {
      "$setDifference": [{
          "$map": {
            "input": "$array",
            "as": "el",
            "in": {
              "$cond": {
                "if": {
                  "$and": [{
                    "$gt": ["$$el.id", 2]
                  }, {
                    "$lte": ["$$el.id", 5]
                  }]
                },
                "then": "$$el",
                "else": false
              }
            }
          }
        },
        [false]
      ]
    }
  }
}])