CODEWITHSUNDEEP
haskellcurryingpartial-application
brander
brander

Reputation: 121

How To with Partial Function Application in Haskell

I am trying to use partial application to shorten the following valid function definition:

ltest50 n = take 50 (iterate ltestonce n)

I thought something like:

ltest50 = take 50 (iterate ltestonce)

or

ltest50 = take 50 (iterate ltestonce$)

would do the trick, but no joy. What is the best way to do this in Haskell?

Upvotes: 2

Views: 324

Answers (1)

willeM_ Van Onsem
willeM_ Van Onsem

Reputation: 477749

Haskell has the "dot operator" or "function composition" (.) :: (b -> c) -> (a -> b) -> a -> c for this:

ltest50 n = take 50 (iterate ltestonce n)

is equivalent to:

ltest50 = take 50 . iterate ltestonce

(.) is defined as:

(.) :: (b -> c) -> (a -> b) -> a -> c
(.) f g x = f (g x)

or a bit shorter:

(.) :: (b -> c) -> (a -> b) -> a -> c
(.) f g x = f $ g x

Since it is a symbol placed between brackets, that means it can be used as . as an infix operator, so:

ltest50 = take 50 . iterate ltestonce

is "syntactical sugar" for:

ltest50 = (.) (take 50) (iterate ltestonce)

At least in Flanders (the Dutch part of Belgium) the operator is sometimes informally called "after" because you apply f after you have applied g on a variable.

Upvotes: 4

Related Questions