CODEWITHSUNDEEP
haskellswiftfunctional-programming
Sinisa Drpa
Sinisa Drpa

Reputation: 917

Functional thinking in Swift?

I started learning Haskell and Swift. I'm wondering if it's the correct way of thinking in "functional" way? The problem is to create card deck: I need to loop through suits and ranks - for every suit create card with given suit and rank. In "imperative" way it'd be:

let suits: Array<Character> = ...
let ranks: Array<Int> = ...

var cards: [Card]
for suit in suits {
    for rank in ranks {
        cards.addObject(Card(suit: suit, rank: rank))
    }
}

Then I tried with pure functions using recursion, it works but, can it be done with less code ? To me "functional" in Swift is less readable, or probably I'm doing it wrong ...

let cards = cardsWithSuits(suits, ranks, [Card]());

func cardsWithSuits(suits: [Character], ranks: [Int], cards: [Card]) -> [Card] {
   if suits.count == 0 { return cards }
   let suit: Character = head(suits)!
   let acc = cardsWithRanks(ranks, suit, cards)

   return cardsWithSuits(drop(1, suits), ranks, acc)
}

func cardsWithRanks(ranks: [Int], suit: Character, cards: [Card]) -> [Card] {
   if ranks.count == 0 { return cards }
   let acc = cards + [Card(suit: suit, rank: head(ranks)!)]

   return cardsWithRanks(drop(1, ranks), suit, acc)
}

Upvotes: 1

Views: 456

Answers (3)

webcpu
webcpu

Reputation: 3422

Swift isn't a functional language. However, you can write swift code in functional style.

func map<T: Collection, U>( _ transform: (T.Iterator.Element) -> U, _ xs: T) -> [U] {
    return xs.reduce([U](), {$0 + [transform($1)]})
}

func concatMap<A, B> (_ process: (A)->[B], _ xs: [A]) -> [B] {
    return xs.reduce([B](), {$0 + process($1)})
}

infix operator <*>
func <*><A, B>(_ xs: [A], _ ys: [B]) -> [(A, B)]{
    let transform: (A, B) -> (A, B) = {($0, $1)}
    return concatMap({x in map({transform(x, $0)}, ys)}, xs)
}

struct Card {
    let suit : Character
    let rank : Int

    static func build(_ sr: (s:Character, r:Int)) -> Card {
        return Card(suit: sr.0, rank: sr.1)
    }
}

func test() {
    let suits : [Character] = [ "C", "D", "H", "S"]
    let ranks = Array(1...13)
    let cards = map(Card.build, suits <*> ranks)
    print(cards)
}

If you want to learn Haskell and Swift, you could refer to https://github.com/unchartedworks/HaskellSwift

Upvotes: 0

Aaron Rasmussen
Aaron Rasmussen

Reputation: 13316

This may not be pretty, and I don't think it is functional programming, but it is less code and it uses Swift's terrific map and reduce functions:

struct Card {
    let suit: String
    let rank: Int
}

let cards = ["Heart", "Diamond", "Club", "Spade"].reduce([Card]()) { (cards, suit) in
    return cards + map(1...13) { rank in return Card(suit: suit, rank: rank) }
}

Upvotes: 1

David Berry
David Berry

Reputation: 41226

Building on the usage of Haskell's applicative concept and <$> and <*> you might find the following generally useful (I think I've translated correctly, although it's based on arrays not sequences):

// use <^> because <$> is already used
infix operator <^> { associativity left }
public func <^> <T, U>(left:(T)->U, right:[T]) -> [U] {
    return map(right) { return left($0) }
}

public func flatten<T>(input:[[T]]) -> [T] {
    return input.reduce([], +)
}

infix operator <*> { associativity left }
public func <*> <T, U>(left:[(T)->U], right:[T]) -> [U] {
    return flatten(map(left) { (function) -> [U] in
        return map(right) { return function($0) }
    })
}

Which then allows you to use the following:

let suits : [Character] = [ "C", "D", "H", "S"]
let ranks = Array(2...14)

struct Card {
    let suit : Character
    let rank : Int

    static func build(suit:Character)(rank:Int) -> Card {
        return Card(suit: suit, rank:rank)
    }
}

Card.build <^> suits <*> ranks

Upvotes: 5

Related Questions