Data.Map.Unboxed.Lifted

Documentation

data Map k v

A map from keys k to values v. The key type must have a Prim instance and the value type is unconstrained.

Instances

Prim k => Functor (Map k)

This fails the functor laws since fmap is strict.

Methods

fmap :: (a -> b) -> Map k a -> Map k b

(<$) :: a -> Map k b -> Map k a

(Prim k, Ord k) => IsList (Map k v)

Associated Types

type Item (Map k v) :: *

Methods

fromList :: [Item (Map k v)] -> Map k v

fromListN :: Int -> [Item (Map k v)] -> Map k v

toList :: Map k v -> [Item (Map k v)]

(Prim k, Eq k, Eq v) => Eq (Map k v)

Methods

(==) :: Map k v -> Map k v -> Bool

(/=) :: Map k v -> Map k v -> Bool

(Prim k, Ord k, Ord v) => Ord (Map k v)

Methods

compare :: Map k v -> Map k v -> Ordering

(<) :: Map k v -> Map k v -> Bool

(<=) :: Map k v -> Map k v -> Bool

(>) :: Map k v -> Map k v -> Bool

(>=) :: Map k v -> Map k v -> Bool

max :: Map k v -> Map k v -> Map k v

min :: Map k v -> Map k v -> Map k v

(Prim k, Show k, Show v) => Show (Map k v)

Methods

showsPrec :: Int -> Map k v -> ShowS

show :: Map k v -> String

showList :: [Map k v] -> ShowS

(Prim k, Ord k, Semigroup v) => Semigroup (Map k v)

Methods

(<>) :: Map k v -> Map k v -> Map k v

sconcat :: NonEmpty (Map k v) -> Map k v

stimes :: Integral b => b -> Map k v -> Map k v

(Prim k, Ord k, Semigroup v) => Monoid (Map k v)

Methods

mempty :: Map k v

mappend :: Map k v -> Map k v -> Map k v

mconcat :: [Map k v] -> Map k v

type Item (Map k v)

type Item (Map k v) = (k, v)

singleton :: Prim k => k -> v -> Map k v

O(1) Create a map with a single element.

lookup :: (Prim k, Ord k) => k -> Map k v -> Maybe v

O(log n) Lookup the value at a key in the map.

size :: Map k v -> Int

O(1) The number of elements in the map.

map :: Prim k => (v -> w) -> Map k v -> Map k w

O(n) Map over the values in the map.

mapMaybe :: Prim k => (v -> Maybe w) -> Map k v -> Map k w

O(n) Drop elements for which the predicate returns Nothing.

Folds

foldMapWithKey' :: (Monoid b, Prim k) => (k -> v -> b) -> Map k v -> b

O(n) Fold over the keys and values of the map with a strict monoidal accumulator. This function does not have left and right variants since the associativity required by a monoid instance means that both variants would always produce the same result.

Monadic Folds

foldlWithKeyM' :: (Monad m, Prim k) => (b -> k -> v -> m b) -> b -> Map k v -> m b

O(n) Left monadic fold over the keys and values of the map. This fold is strict in the accumulator.

foldrWithKeyM' :: (Monad m, Prim k) => (k -> v -> b -> m b) -> b -> Map k v -> m b

O(n) Right monadic fold over the keys and values of the map. This fold is strict in the accumulator.

foldlMapWithKeyM' :: (Monad m, Monoid b, Prim k) => (k -> v -> m b) -> Map k v -> m b

O(n) Monadic left fold over the keys and values of the map with a strict monoidal accumulator. The monoidal accumulator is appended to the left after each reduction.

foldrMapWithKeyM'

Arguments

:: (Monad m, Monoid b, Prim k)

=> (k -> v -> m b)

reduction

-> Map k v

map

-> m b

O(n) Monadic right fold over the keys and values of the map with a strict monoidal accumulator. The monoidal accumulator is appended to the right after each reduction.

List Conversion

fromList :: (Prim k, Ord k) => [(k, v)] -> Map k v

O(n*log n) Create a map from a list of key-value pairs. If the list contains more than one value for the same key, the last value is retained. If the keys in the argument are in nondescending order, this algorithm runs in O(n) time instead.

fromListAppend :: (Prim k, Ord k, Semigroup v) => [(k, v)] -> Map k v

O(n*log n) This function has the same behavior as fromList, but it combines values with the Semigroup instances instead of choosing the last occurrence.

fromListN

Arguments

:: (Prim k, Ord k)

=> Int

expected size of resulting Map

-> [(k, v)]

key-value pairs

-> Map k v

O(n*log n) This function has the same behavior as fromList regardless of whether or not the expected size is accurate. Additionally, negative sizes are handled correctly. The expected size is used as the size of the initially allocated buffer when building the Map. If the keys in the argument are in nondescending order, this algorithm runs in O(n) time.

fromListAppendN

Arguments

:: (Prim k, Ord k, Semigroup v)

=> Int

expected size of resulting Map

-> [(k, v)]

key-value pairs

-> Map k v

O(n*log n) This function has the same behavior as fromListN, but it combines values with the Semigroup instances instead of choosing the last occurrence.

Array Conversion

unsafeFreezeZip :: (Ord k, Prim k) => MutablePrimArray s k -> MutableArray s v -> ST s (Map k v)

O(n*log n) Zip an array of keys with an array of values. If they are not the same length, the longer one will be truncated to match the shorter one. This function sorts and deduplicates the array of keys, preserving the last value associated with each key. The argument arrays may not be reused after being passed to this function.

This is by far the fastest way to create a map, since the functions backing it are aggressively specialized. It internally uses a hybrid of mergesort and insertion sort provided by the primitive-sort package. It generates much less garbage than any of the fromList variants.