$lastN (aggregation accumulator)

Null and Missing Values

• $lastN does not filter out null values.

• $lastN converts missing values to null.

Consider the following aggregation that returns the last five documents from a group:

db.aggregate( [
   {
      $documents: [
         { playerId: "PlayerA", gameId: "G1", score: 1 },
         { playerId: "PlayerB", gameId: "G1", score: 2 },
         { playerId: "PlayerC", gameId: "G1", score: 3 },
         { playerId: "PlayerD", gameId: "G1"},
         { playerId: "PlayerE", gameId: "G1", score: null }
      ]
   },
   {
      $group:
      {
         _id: "$gameId",
         lastFiveScores:
            {
               $lastN:
                  {
                     input: "$score",
                     n: 5
                  }
            }
      }
   }
] )

In this example:

• $documents creates the literal documents that contain player scores.

• $group groups the documents by gameId. This example has only one gameId, G1.

• PlayerD has a missing score and PlayerE has a null score. These values are both considered as null.

• The lastFiveScores field is specified using input : "$score" and returned as an array.

• Since there is no sort criteria the last 5 score fields are returned.

[
   {
      _id: "G1",
      lastFiveScores: [ 1, 2, 3, null, null ]
   }
]

Comparison of $lastN and $bottomN

Both $lastN and $bottomN accumulators can accomplish similar results.

In general:

• If the documents coming into $group are already ordered, you should use $lastN.

• If you're sorting and selecting the bottom n elements then you can use $bottomN to accomplish both tasks with one accumulator.

• $lastN can be used as an aggregation expression, $bottomN cannot.

Window Function and Aggregation Expression Support

$lastN is supported as an aggregation expression.

For details on aggregation expression usage see Using $lastN as an Aggregation Expression.

$lastN is supported as a window operator.

Memory Limit Considerations

Aggregation pipelines which call $lastN are subject to the 100 MB limit. If this limit is exceeded for an individual group, the aggregation fails with an error.

Find the Last Three Player Scores for a Single Game

You can use the $lastN accumulator to find the last three scores in a single game.

db.gamescores.aggregate( [
   {
      $match : { gameId : "G1" }
   },
   {
      $group:
         {
            _id: "$gameId",
            lastThreeScores:
               {
                  $lastN:
                  {
                     input: ["$playerId", "$score"],
                     n:3
                  }
               }
         }
   }
] )

The example pipeline:

• Uses $match to filter the results on a single gameId. In this case, G1.

• Uses $group to group the results by gameId. In this case, G1.

• Specifies the fields that are output from $lastN with output : ["$playerId"," $score"].

• Uses $lastN to return the last three documents for the G1 game with n : 3.

The operation returns the following results:

[
   {
      _id: "G1",
      lastThreeScores: [ [ "PlayerB", 33 ], [ "PlayerC", 99 ], [ "PlayerD", 1 ] ]
   }
]

Finding the Last Three Player Scores Across Multiple Games

You can use the $lastN accumulator to find the last n input fields in each game.

db.gamescores.aggregate( [
      {
         $group:
         {
         _id: "$gameId", playerId:
            {
               $lastN:
                  {
                     input: [ "$playerId","$score" ],
                     n: 3
                  }
            }
         }
      }
] )

The example pipeline:

• Uses $group to group the results by gameId.

• Uses $lastN to return the last three documents for each game with n: 3.

• Specifies the fields that are input for $lastN with input : ["$playerId", "$score"].

The operation returns the following results:

[
   {
      _id: 'G2',
      playerId: [ [ 'PlayerB', 14 ], [ 'PlayerC', 66 ], [ 'PlayerD', 80 ] ]
   },
   {
      _id: 'G1',
      playerId: [ [ 'PlayerB', 33 ], [ 'PlayerC', 99 ], [ 'PlayerD', 1 ] ]
   }
]

Using $sort With $lastN

Using a $sort stage earlier in the pipeline can influence the results of the $lastN accumulator.

In this example:

• {$sort : { score : -1 } } sorts the highest scores to the back of each group.

• lastN returns the three lowest scores from the back of each group.

db.gamescores.aggregate( [
      { $sort : { score : -1 } },
      {
         $group:
         { _id: "$gameId", playerId:
            {
               $lastN:
                  {
                     input: [ "$playerId","$score" ],
                     n: 3
                  }
            }
         }
      }
   ] )

The operation returns the following results:

[
   {
      _id: 'G2',
      playerId: [ [ 'PlayerC', 66 ], [ 'PlayerB', 14 ], [ 'PlayerA', 10 ] ]
   },
   {
      _id: 'G1',
      playerId: [ [ 'PlayerB', 33 ], [ 'PlayerA', 31 ], [ 'PlayerD', 1 ] ]
   }
]

Computing n Based on the Group Key for $group

You can also assign the value of n dynamically. In this example, the $cond expression is used on the gameId field.

db.gamescores.aggregate([
   {
      $group:
      {
         _id: {"gameId": "$gameId"},
         gamescores:
            {
               $lastN:
                  {
                     input: "$score",
                     n: { $cond: { if: {$eq: ["$gameId","G2"] }, then: 1, else: 3 } }
                  }
            }
      }
   }
] )

The example pipeline:

• Uses $group to group the results by gameId.

• Specifies the fields that input for $lastN with input : "$score".

• If the gameId is G2 then n is 1, otherwise n is 3.

The operation returns the following results:

[
   { _id: { gameId: "G1" }, gamescores: [ 33, 99, 1 ] },
   { _id: { gameId: "G2" }, gamescores: [ 80 ] }
]

Using $lastN as an Aggregation Expression

You can also use $lastN as an aggregation expression.

In this example:

• $documents creates the literal document that contains an array of values.

• $project is used to return the output of $lastN.

• _id is omited from the output with _id : 0.

• $lastN uses the input array of [10, 20, 30, 40].

• The last three elements of the array are returned for the input document.

db.aggregate( [
   {
      $documents: [
         { array: [10, 20, 30, 40] } ]
   },
   { $project: {
      lastThreeElements:{
                           $lastN:
                           {
                              input: "$array",
                              n: 3
                           }
                        }
               }
   }
] )

The operation returns the following results:

[ { lastThreeElements: [ 20, 30, 40 ] } ]