$lastN (accumulator operator)

Options

Null and Missing Values

• $lastN does not filter out null values.

• $lastN converts missing values to null.

Comparison of $lastN and $bottomN

You can use $lastN and $bottomN to return the last n elements from a group. However, the difference of when to use each depends on whether your input documents are already sorted.

• 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.

Find the Last Three Movies for a Single Genre

You can use the $lastN accumulator to find the most-recently released movies in a particular genre.

db.movies.aggregate( [
   {
      $match: { genres: "Horror", year: { $type: "int" } }
   },
   {
      $sort: { year: 1, title: 1 }
   },
   {
      $group:
         {
            _id: "Horror",
            lastThreeMovies:
               {
                  $lastN:
                  {
                     input: [ "$title", "$year" ],
                     n: 3
                  }
               }
         }
   }
] )

[
  {
    _id: 'Horror',
    lastThreeMovies: [
      [ 'The Final Girls', 2015 ],
      [ 'The Hoarder', 2015 ],
      [ 'The Witch', 2015 ]
    ]
  }
]

The preceding pipeline:

• Uses $match to filter the results to movies where Horror is included in the genres array, and where year is an integer.

• Sorts the results by year in ascending order, then by title alphabetically as a tiebreaker.

• Groups all resulting documents into a single group.

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

• Uses $lastN to return the last three documents (sorted by year ascending, with title as a tiebreaker) for the Horror genre with n: 3.

Find the Last Three Movies Across Multiple Genres

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

db.movies.aggregate( [
   {
      $match: { year: { $type: "int" } }
   },
   {
      $unwind: "$genres"
   },
   {
      $sort: { year: 1, title: 1 }
   },
   {
      $group:
         {
            _id: "$genres",
            movies:
               {
                  $lastN:
                     {
                        input: [ "$title", "$year" ],
                        n: 3
                     }
               }
         }
   },
   { $sort: { _id: 1 } },
   { $limit: 5 }
] )

[
  {
    _id: 'Action',
    movies: [
      [ 'War Pigs', 2015 ],
      [ 'Wolf Warrior', 2015 ],
      [ 'The Masked Saint', 2016 ]
    ]
  },
  {
    _id: 'Adventure',
    movies: [
      [ 'The SpongeBob Movie: Sponge Out of Water', 2015 ],
      [ 'Tomorrowland', 2015 ],
      [ 'Wolf Totem', 2015 ]
    ]
  },
  {
    _id: 'Animation',
    movies: [
      [ 'The Magic Mountain', 2015 ],
      [ 'The SpongeBob Movie: Sponge Out of Water', 2015 ],
      [ 'World of Tomorrow', 2015 ]
    ]
  },
  {
    _id: 'Biography',
    movies: [
      [ 'Twinsters', 2015 ],
      [ 'Whitney', 2015 ],
      [ 'The Masked Saint', 2016 ]
    ]
  },
  {
    _id: 'Comedy',
    movies: [
      [ 'Wondrous Boccaccio', 2015 ],
      [ 'World of Tomorrow', 2015 ],
      [ "You're Ugly Too", 2015 ]
    ]
  }
]

The preceding pipeline:

• Uses $match to filter to movies where year is an integer.

• Uses $unwind to create a separate document for each genre in the genres array.

• Sorts the results by year in ascending order, then by title alphabetically.

• Uses $group to group the results by genres.

• Uses $lastN to return the last three documents (by year ascending, with title as a tiebreaker) for each genre with n: 3.

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

• Limits the output to five documents to improve readability.

Set n Based on the Group Key

You can assign the value of n dynamically based on the group key. In this example, the $cond expression uses the rated field to change the value of n:

db.movies.aggregate( [
   {
      $match: {
         rated: { $in: [ "G", "PG", "PG-13", "R" ] },
         year: { $type: "int" },
         title: { $regex: "^[A-Za-z0-9]" }
      }
   },
   { $sort: { title: 1 } },
   {
      $group:
         {
            _id: { rated: "$rated" },
            movies:
               {
                  $lastN:
                     {
                        input: "$title",
                        n: { $cond: { if: { $eq: ["$rated", "PG"] }, then: 3, else: 1 } }
                     }
               }
         }
   },
   { $sort: { "_id.rated": 1 } }
] )

[
  { _id: { rated: 'G' }, movies: [ 'Zambezia' ] },
  {
    _id: { rated: 'PG' },
    movies: [ 'Zoom', 'Zorro: The Gay Blade', 'Zulu Dawn' ]
  },
  { _id: { rated: 'PG-13' }, movies: [ 'xXx: State of the Union' ] },
  { _id: { rated: 'R' }, movies: [ 'eXistenZ' ] }
]

The preceding pipeline:

• Filters for movies with content ratings of G, PG, PG-13, or R, where year is an integer and title starts with an alphanumeric character.

• Sorts the results alphabetically by title.

• Groups movies by rated (the content rating).

• Uses the $cond expression to return 3 movies for PG-rated movies and 1 movie for all other ratings.

In the output, the PG rating returns 3 movies, while G, PG-13, and R ratings each return 1 movie.