.shuffle() and .shuffled() are part of Swift
Original historic question:
How do I randomize or shuffle the elements within an array in Swift? For example, if my array consists of 52 playing cards, I want to shuffle the array in order to shuffle the deck.
This answer details how to shuffle with a fast and uniform algorithm (Fisher-Yates) in Swift 4.2+ and how to add the same feature in the various previous versions of Swift. The naming and behavior for each Swift version matches the mutating and nonmutating sorting methods for that version.
Swift 4.2+
shuffle and shuffled are native starting Swift 4.2. Example usage:
let x = [1, 2, 3].shuffled()
// x == [2, 3, 1]
let fiveStrings = stride(from: 0, through: 100, by: 5).map(String.init).shuffled()
// fiveStrings == ["20", "45", "70", "30", ...]
var numbers = [1, 2, 3, 4]
numbers.shuffle()
// numbers == [3, 2, 1, 4]
Swift 4.0 and 4.1
These extensions add a shuffle() method to any mutable collection (arrays and unsafe mutable buffers) and a shuffled() method to any sequence:
extension MutableCollection {
/// Shuffles the contents of this collection.
mutating func shuffle() {
let c = count
guard c > 1 else { return }
for (firstUnshuffled, unshuffledCount) in zip(indices, stride(from: c, to: 1, by: -1)) {
// Change `Int` in the next line to `IndexDistance` in < Swift 4.1
let d: Int = numericCast(arc4random_uniform(numericCast(unshuffledCount)))
let i = index(firstUnshuffled, offsetBy: d)
swapAt(firstUnshuffled, i)
}
}
}
extension Sequence {
/// Returns an array with the contents of this sequence, shuffled.
func shuffled() -> [Element] {
var result = Array(self)
result.shuffle()
return result
}
}
Same usage as in Swift 4.2 examples above.
Swift 3
These extensions add a shuffle() method to any mutable collection and a shuffled() method to any sequence:
extension MutableCollection where Indices.Iterator.Element == Index {
/// Shuffles the contents of this collection.
mutating func shuffle() {
let c = count
guard c > 1 else { return }
for (firstUnshuffled , unshuffledCount) in zip(indices, stride(from: c, to: 1, by: -1)) {
// Change `Int` in the next line to `IndexDistance` in < Swift 3.2
let d: Int = numericCast(arc4random_uniform(numericCast(unshuffledCount)))
guard d != 0 else { continue }
let i = index(firstUnshuffled, offsetBy: d)
self.swapAt(firstUnshuffled, i)
}
}
}
extension Sequence {
/// Returns an array with the contents of this sequence, shuffled.
func shuffled() -> [Iterator.Element] {
var result = Array(self)
result.shuffle()
return result
}
}
Same usage as in Swift 4.2 examples above.
Swift 2
(obsolete language: you can't use Swift 2.x to publish on iTunes Connect starting July 2018)
extension MutableCollectionType where Index == Int {
/// Shuffle the elements of `self` in-place.
mutating func shuffleInPlace() {
// empty and single-element collections don't shuffle
if count < 2 { return }
for i in startIndex ..< endIndex - 1 {
let j = Int(arc4random_uniform(UInt32(count - i))) + i
guard i != j else { continue }
swap(&self[i], &self[j])
}
}
}
extension CollectionType {
/// Return a copy of `self` with its elements shuffled.
func shuffle() -> [Generator.Element] {
var list = Array(self)
list.shuffleInPlace()
return list
}
}
Usage:
[1, 2, 3].shuffle()
// [2, 3, 1]
let fiveStrings = 0.stride(through: 100, by: 5).map(String.init).shuffle()
// ["20", "45", "70", "30", ...]
var numbers = [1, 2, 3, 4]
numbers.shuffleInPlace()
// [3, 2, 1, 4]
Swift 1.2
(obsolete language: you can't use Swift 1.x to publish on iTunes Connect starting July 2018)
shuffle as a mutating array method
This extension will let you shuffle a mutable Array instance in place:
extension Array {
mutating func shuffle() {
if count < 2 { return }
for i in 0..<(count - 1) {
let j = Int(arc4random_uniform(UInt32(count - i))) + i
swap(&self[i], &self[j])
}
}
}
var numbers = [1, 2, 3, 4, 5, 6, 7, 8]
numbers.shuffle() // e.g., numbers == [6, 1, 8, 3, 2, 4, 7, 5]
shuffled as a non-mutating array method
This extension will let you retrieve a shuffled copy of an Array instance:
extension Array {
func shuffled() -> [T] {
if count < 2 { return self }
var list = self
for i in 0..<(list.count - 1) {
let j = Int(arc4random_uniform(UInt32(list.count - i))) + i
swap(&list[i], &list[j])
}
return list
}
}
let numbers = [1, 2, 3, 4, 5, 6, 7, 8]
let mixedup = numbers.shuffled() // e.g., mixedup == [6, 1, 8, 3, 2, 4, 7, 5]
Edit: As noted in other answers, Swift 4.2 finally adds random number generation to the standard library, complete with array shuffling.
However, the GKRandom / GKRandomDistribution suite in GameplayKit can still be useful with the new RandomNumberGenerator protocol — if you add extensions to the GameplayKit RNGs to conform to the new standard library protocol, you can easily get:
sendable RNGs (that can reproduce a "random" sequence when needed for testing)
RNGs that sacrifice robustness for speed
RNGs that produce non-uniform distributions
...and still make use of the nice new "native" random APIs in Swift.
The rest of this answer concerns such RNGs and/or their use in older Swift compilers.
There are some good answers here already, as well as some good illustrations of why writing your own shuffle can be error-prone if you're not careful.
In iOS 9, macOS 10.11, and tvOS 9 (or later), you don't have to write your own. There's an efficient, correct implementation of Fisher-Yates in GameplayKit (which, despite the name, is not just for games).
If you just want a unique shuffle:
let shuffled = GKRandomSource.sharedRandom().arrayByShufflingObjects(in: array)
If you want to be able to replicate a shuffle or series of shuffles, choose and seed a specific random source; e.g.
let lcg = GKLinearCongruentialRandomSource(seed: mySeedValue)
let shuffled = lcg.arrayByShufflingObjects(in: array)
In iOS 10 / macOS 10.12 / tvOS 10, there's also a convenience syntax for shuffling via an extension on NSArray. Of course, that's a little cumbersome when you're using a Swift Array (and it loses its element type on coming back to Swift):
let shuffled1 = (array as NSArray).shuffled(using: random) // -> [Any]
let shuffled2 = (array as NSArray).shuffled() // use default random source
But it's pretty easy to make a type-preserving Swift wrapper for it:
extension Array {
func shuffled(using source: GKRandomSource) -> [Element] {
return (self as NSArray).shuffled(using: source) as! [Element]
}
func shuffled() -> [Element] {
return (self as NSArray).shuffled() as! [Element]
}
}
let shuffled3 = array.shuffled(using: random)
let shuffled4 = array.shuffled()
In Swift 2.0, GameplayKit may come to the rescue! (supported by iOS9 or later)
import GameplayKit
func shuffle() {
array = GKRandomSource.sharedRandom().arrayByShufflingObjectsInArray(array)
}
Here's something possibly a little shorter:
sorted(a) {_, _ in arc4random() % 2 == 0}
Taking Nate's algorithm I wanted to see how this would look with Swift 2 and protocol extensions.
This is what I came up with.
extension MutableCollectionType where Self.Index == Int {
mutating func shuffleInPlace() {
let c = self.count
for i in 0..<(c - 1) {
let j = Int(arc4random_uniform(UInt32(c - i))) + i
swap(&self[i], &self[j])
}
}
}
extension MutableCollectionType where Self.Index == Int {
func shuffle() -> Self {
var r = self
let c = self.count
for i in 0..<(c - 1) {
let j = Int(arc4random_uniform(UInt32(c - i))) + i
swap(&r[i], &r[j])
}
return r
}
}
Now, any MutableCollectionType can use these methods given it uses Int as an Index
As of swift 4.2 there are two handy functions:
// shuffles the array in place
myArray.shuffle()
and
// generates a new array with shuffled elements of the old array
let newArray = myArray.shuffled()
In my case, I had some problems of swapping objects in Array. Then I scratched my head and went for reinventing the wheel.
// swift 3.0 ready
extension Array {
func shuffled() -> [Element] {
var results = [Element]()
var indexes = (0 ..< count).map { $0 }
while indexes.count > 0 {
let indexOfIndexes = Int(arc4random_uniform(UInt32(indexes.count)))
let index = indexes[indexOfIndexes]
results.append(self[index])
indexes.remove(at: indexOfIndexes)
}
return results
}
}
This is a version of Nate's implementation of the Fisher-Yates shuffle for Swift 4
(Xcode 9).
extension MutableCollection {
/// Shuffle the elements of `self` in-place.
mutating func shuffle() {
for i in indices.dropLast() {
let diff = distance(from: i, to: endIndex)
let j = index(i, offsetBy: numericCast(arc4random_uniform(numericCast(diff))))
swapAt(i, j)
}
}
}
extension Collection {
/// Return a copy of `self` with its elements shuffled
func shuffled() -> [Element] {
var list = Array(self)
list.shuffle()
return list
}
}
The changes are:
The constraint Indices.Iterator.Element == Index is now part
of the Collection protocol, and need not be imposed on the
extension anymore.
Exchanging elements must done by calling swapAt() on the collection,
compare SE-0173 Add MutableCollection.swapAt(_:_:).
Element is an alias for Iterator.Element.
Swift 4
Shuffle the elements of an array in a for loop where i is the mixing ratio
var cards = [Int]() //Some Array
let i = 4 // is the mixing ratio
func shuffleCards() {
for _ in 0 ..< cards.count * i {
let card = cards.remove(at: Int(arc4random_uniform(UInt32(cards.count))))
cards.insert(card, at: Int(arc4random_uniform(UInt32(cards.count))))
}
}
Or with extension Int
func shuffleCards() {
for _ in 0 ..< cards.count * i {
let card = cards.remove(at: cards.count.arc4random)
cards.insert(card, at: cards.count.arc4random)
}
}
extension Int {
var arc4random: Int {
if self > 0 {
print("Arc for random positiv self \(Int(arc4random_uniform(UInt32(self))))")
return Int(arc4random_uniform(UInt32(self)))
} else if self < 0 {
print("Arc for random negotiv self \(-Int(arc4random_uniform(UInt32(abs(self)))))")
return -Int(arc4random_uniform(UInt32(abs(self))))
} else {
print("Arc for random equal 0")
return 0
}
}
}
This is what I use:
func newShuffledArray(array:NSArray) -> NSArray {
var mutableArray = array.mutableCopy() as! NSMutableArray
var count = mutableArray.count
if count>1 {
for var i=count-1;i>0;--i{
mutableArray.exchangeObjectAtIndex(i, withObjectAtIndex: Int(arc4random_uniform(UInt32(i+1))))
}
}
return mutableArray as NSArray
}
Swift 3 solution, following #Nate Cook answer: (work if the index starts with 0, see comments below)
extension Collection {
/// Return a copy of `self` with its elements shuffled
func shuffle() -> [Generator.Element] {
var list = Array(self)
list.shuffleInPlace()
return list
} }
extension MutableCollection where Index == Int {
/// Shuffle the elements of `self` in-place.
mutating func shuffleInPlace() {
// empty and single-element collections don't shuffle
if count < 2 { return }
let countInt = count as! Int
for i in 0..<countInt - 1 {
let j = Int(arc4random_uniform(UInt32(countInt - i))) + i
guard i != j else { continue }
swap(&self[i], &self[j])
}
}
}
This is how its done in a Simplest way.import Gamplaykit to your VC and use the below code. Tested in Xcode 8.
import GameplayKit
let array: NSArray = ["Jock", "Ellie", "Sue Ellen", "Bobby", "JR", "Pamela"]
override func viewDidLoad() {
super.viewDidLoad()
print(array.shuffled())
}
If you want to get a shuffled String from an Array you can use below code..
func suffleString() {
let ShuffleArray = array.shuffled()
suffleString.text = ShuffleArray.first as? String
print(suffleString.text!)
}
With Swift 3, if you want to shuffle an array in place or get a new shuffled array from an array, AnyIterator can help you. The idea is to create an array of indices from your array, to shuffle those indices with an AnyIterator instance and swap(_:_:) function and to map each element of this AnyIterator instance with the array's corresponding element.
The following Playground code shows how it works:
import Darwin // required for arc4random_uniform
let array = ["Jock", "Ellie", "Sue Ellen", "Bobby", "JR", "Pamela"]
var indexArray = Array(array.indices)
var index = indexArray.endIndex
let indexIterator: AnyIterator<Int> = AnyIterator {
guard let nextIndex = indexArray.index(index, offsetBy: -1, limitedBy: indexArray.startIndex)
else { return nil }
index = nextIndex
let randomIndex = Int(arc4random_uniform(UInt32(index)))
if randomIndex != index {
swap(&indexArray[randomIndex], &indexArray[index])
}
return indexArray[index]
}
let newArray = indexIterator.map { array[$0] }
print(newArray) // may print: ["Jock", "Ellie", "Sue Ellen", "JR", "Pamela", "Bobby"]
You can refactor the previous code and create a shuffled() function inside an Array extension in order to get a new shuffled array from an array:
import Darwin // required for arc4random_uniform
extension Array {
func shuffled() -> Array<Element> {
var indexArray = Array<Int>(indices)
var index = indexArray.endIndex
let indexIterator = AnyIterator<Int> {
guard let nextIndex = indexArray.index(index, offsetBy: -1, limitedBy: indexArray.startIndex)
else { return nil }
index = nextIndex
let randomIndex = Int(arc4random_uniform(UInt32(index)))
if randomIndex != index {
swap(&indexArray[randomIndex], &indexArray[index])
}
return indexArray[index]
}
return indexIterator.map { self[$0] }
}
}
Usage:
let array = ["Jock", "Ellie", "Sue Ellen", "Bobby", "JR", "Pamela"]
let newArray = array.shuffled()
print(newArray) // may print: ["Bobby", "Pamela", "Jock", "Ellie", "JR", "Sue Ellen"]
let emptyArray = [String]()
let newEmptyArray = emptyArray.shuffled()
print(newEmptyArray) // prints: []
As an alternative to the previous code, you can create a shuffle() function inside an Array extension in order to shuffle an array in place:
import Darwin // required for arc4random_uniform
extension Array {
mutating func shuffle() {
var indexArray = Array<Int>(indices)
var index = indexArray.endIndex
let indexIterator = AnyIterator<Int> {
guard let nextIndex = indexArray.index(index, offsetBy: -1, limitedBy: indexArray.startIndex)
else { return nil }
index = nextIndex
let randomIndex = Int(arc4random_uniform(UInt32(index)))
if randomIndex != index {
swap(&indexArray[randomIndex], &indexArray[index])
}
return indexArray[index]
}
self = indexIterator.map { self[$0] }
}
}
Usage:
var mutatingArray = ["Jock", "Ellie", "Sue Ellen", "Bobby", "JR", "Pamela"]
mutatingArray.shuffle()
print(mutatingArray) // may print ["Sue Ellen", "Pamela", "Jock", "Ellie", "Bobby", "JR"]
In Swift 4.2, there is now a method for both a mutable shuffle and immutable shuffled. You can read more about the random generation and array stuff here.
You can use generic swap function as well and implement mentioned Fisher-Yates:
for idx in 0..<arr.count {
let rnd = Int(arc4random_uniform(UInt32(idx)))
if rnd != idx {
swap(&arr[idx], &arr[rnd])
}
}
or less verbose:
for idx in 0..<steps.count {
swap(&steps[idx], &steps[Int(arc4random_uniform(UInt32(idx)))])
}
works!!. organisms is the array to shuffle.
extension Array
{
/** Randomizes the order of an array's elements. */
mutating func shuffle()
{
for _ in 0..<10
{
sort { (_,_) in arc4random() < arc4random() }
}
}
}
var organisms = [
"ant", "bacteria", "cougar",
"dog", "elephant", "firefly",
"goat", "hedgehog", "iguana"]
print("Original: \(organisms)")
organisms.shuffle()
print("Shuffled: \(organisms)")
Working Array Extension (mutating & non-mutating)
Swift 4.1 / Xcode 9
The top answer is deprecated, so I took it upon myself to create my own extension to shuffle an array in the newest version of Swift, Swift 4.1 (Xcode 9):
extension Array {
// Non-mutating shuffle
var shuffled : Array {
let totalCount : Int = self.count
var shuffledArray : Array = []
var count : Int = totalCount
var tempArray : Array = self
for _ in 0..<totalCount {
let randomIndex : Int = Int(arc4random_uniform(UInt32(count)))
let randomElement : Element = tempArray.remove(at: randomIndex)
shuffledArray.append(randomElement)
count -= 1
}
return shuffledArray
}
// Mutating shuffle
mutating func shuffle() {
let totalCount : Int = self.count
var shuffledArray : Array = []
var count : Int = totalCount
var tempArray : Array = self
for _ in 0..<totalCount {
let randomIndex : Int = Int(arc4random_uniform(UInt32(count)))
let randomElement : Element = tempArray.remove(at: randomIndex)
shuffledArray.append(randomElement)
count -= 1
}
self = shuffledArray
}
}
Call Non-Mutating Shuffle [Array] -> [Array]:
let array = [1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20]
print(array.shuffled)
This prints array in a random order.
Call Mutating Shuffle [Array] = [Array]:
var array = [1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20]
array.shuffle()
// The array has now been mutated and contains all of its initial
// values, but in a randomized shuffled order
print(array)
This prints array in its current order, which has already been randomly shuffled.
Hopes this works for everybody, if you have any questions, suggestions, or comments, feel free to ask!
In SWIFT 4
func createShuffledSequenceOfNumbers(max:UInt)->[UInt] {
var array:[UInt]! = []
var myArray:[UInt]! = []
for i in 1...max {
myArray.append(i)
}
for i in 1...max {
array.append(i)
}
var tempArray:[Int]! = []
for index in 0...(myArray.count - 1) {
var isNotFinded:Bool = true
while(isNotFinded){
let randomNumber = arc4random_uniform(UInt32(myArray.count))
let randomIndex = Int(randomNumber)
if(!tempArray.contains(randomIndex)){
tempArray.append(randomIndex)
array[randomIndex] = myArray[index]
isNotFinded = false
}
}
}
return array
}
If you want to use simple Swift For loop function use this ->
var arrayItems = ["A1", "B2", "C3", "D4", "E5", "F6", "G7", "H8", "X9", "Y10", "Z11"]
var shuffledArray = [String]()
for i in 0..<arrayItems.count
{
let randomObject = Int(arc4random_uniform(UInt32(items.count)))
shuffledArray.append(items[randomObject])
items.remove(at: randomObject)
}
print(shuffledArray)
Swift Array suffle using extension ->
extension Array {
// Order Randomize
mutating func shuffle() {
for _ in 0..<count {
sort { (_,_) in arc4random() < arc4random() }
}
}
}
This is how to shuffle one array with a seed in Swift 3.0.
extension MutableCollection where Indices.Iterator.Element == Index {
mutating func shuffle() {
let c = count
guard c > 1 else { return }
for (firstUnshuffled , unshuffledCount) in zip(indices, stride(from: c, to: 1, by: -1)) {
srand48(seedNumber)
let number:Int = numericCast(unshuffledCount)
let r = floor(drand48() * Double(number))
let d: IndexDistance = numericCast(Int(r))
guard d != 0 else { continue }
let i = index(firstUnshuffled, offsetBy: d)
swap(&self[firstUnshuffled], &self[i])
}
}
}
let shuffl = GKRandomSource.sharedRandom().arrayByShufflingObjects(in: arrayObject)
This is what I use:
import GameplayKit
extension Collection {
func shuffled() -> [Iterator.Element] {
let shuffledArray = (self as? NSArray)?.shuffled()
let outputArray = shuffledArray as? [Iterator.Element]
return outputArray ?? []
}
mutating func shuffle() {
if let selfShuffled = self.shuffled() as? Self {
self = selfShuffled
}
}
}
// Usage example:
var numbers = [1,2,3,4,5]
numbers.shuffle()
print(numbers) // output example: [2, 3, 5, 4, 1]
print([10, "hi", 9.0].shuffled()) // output example: [hi, 10, 9]
Simple Example:
extension Array {
mutating func shuffled() {
for _ in self {
// generate random indexes that will be swapped
var (a, b) = (Int(arc4random_uniform(UInt32(self.count - 1))), Int(arc4random_uniform(UInt32(self.count - 1))))
if a == b { // if the same indexes are generated swap the first and last
a = 0
b = self.count - 1
}
swap(&self[a], &self[b])
}
}
}
var array = [1,2,3,4,5,6,7,8,9,10]
array.shuffled()
print(array) // [9, 8, 3, 5, 7, 6, 4, 2, 1, 10]
Here's some code that runs in playground. You won't need to import Darwin in an actual Xcode project.
import darwin
var a = [1,2,3,4,5,6,7]
func shuffle<ItemType>(item1: ItemType, item2: ItemType) -> Bool {
return drand48() > 0.5
}
sort(a, shuffle)
println(a)
It stop at "swap(&self[i], &self[j])" when I upgrade the xCode version to 7.4 beta.
fatal error: swapping a location with itself is not supported
I found the reason that i = j (function of swap will exploded)
So I add a condition as below
if (i != j){
swap(&list[i], &list[j])
}
YA! It's OK for me.
Related
If I have two int arrays such as
var array1 = [1,2,3]
var array2 = [1,2,3,5]
I'd like to be able to add the first element of the first array with the first element of the second array, and so on. However if an array has a different length than the other I'd like to keep the element that was not added in the return array. For this example my return array would be [2,4,6,5].
I tried using zip(array1,array2).map(+) but it would exclude the 5 from array2.
After adding the elements at the index positions which are common to both arrays (what you already did with zip and map) just append the remaining elements from both arrays (using append(contentsOf:) and dropFirst):
let array1 = [1, 2, 3]
let array2 = [1, 2, 3, 5]
var combined = zip(array1, array2).map(+)
let commonCount = combined.count
combined.append(contentsOf: array1.dropFirst(commonCount))
combined.append(contentsOf: array2.dropFirst(commonCount))
print(combined) // [2, 4, 6, 5]
AnySequence(zip: (1...3, [1, 2, 3, 5])).map {
Optional($0).map(+) ?? firstNonNil($0)!
}
public extension AnySequence {
/// Like `zip`, but with `nil` elements for the shorter sequence after it is exhausted.
init<Sequence0: Sequence, Sequence1: Sequence>(
zip zipped: (Sequence0, Sequence1)
) where Element == (Sequence0.Element?, Sequence1.Element?) {
self.init(
sequence(
state: (zipped.0.makeIterator(), zipped.1.makeIterator())
) { iterators in
((iterators.0.next(), iterators.1.next()) as Optional)
.filter { $0 != nil || $1 != nil }
}
)
}
}
public extension Optional {
/// Exchange two optionals for a single optional tuple.
/// - Returns: `nil` if either tuple element is `nil`.
init<Wrapped0, Wrapped1>(_ optionals: (Wrapped0?, Wrapped1?))
where Wrapped == (Wrapped0, Wrapped1) {
switch optionals {
case let (wrapped0?, wrapped1?):
self = (wrapped0, wrapped1)
default:
self = nil
}
}
/// Transform `.some` into `.none`, if a condition fails.
/// - Parameters:
/// - isSome: The condition that will result in `nil`, when evaluated to `false`.
func filter(_ isSome: (Wrapped) throws -> Bool) rethrows -> Self {
try flatMap { try isSome($0) ? $0 : nil }
}
}
public func firstNonNil<Element>(_ tuple: (Element?, Element?)) -> Element? {
switch tuple {
case (let _0?, _):
return _0
case (nil, let _1?):
return _1
case (nil, nil):
return nil
}
}
func combine2Arrays(array1:[Int], array2:[Int]) -> [Int] {
var finalArray:[Int] = []
let maxSize = max(array1.count, array2.count)
for i in 0..<maxSize {
let valToAdd1 = (array1.count > i ? array1[i] : 0)
let valToAdd2 = (array2.count > i ? array2[i] : 0)
let finalVal = valToAdd1 + valToAdd2
finalArray.append(finalVal)
}
return finalArray
}
print(combine2Arrays(array1: [1,2,3], array2: [1,2,3,5]))
OR
func combine2Arrays(array1:[Int], array2:[Int]) -> [Int] {
var finalArray:[Int] = zip(array1,array2).map(+)
let largerArray = array1.count > array2.count ? array1 : array2
let smallerArray = array1.count > array2.count ? array2 : array1
let min = smallerArray.count
let max = largerArray.count
for i in min..<max {
finalArray.append(largerArray[i])
}
return finalArray
}
print(combine2Arrays(array1: [1,2,3], array2: [1,2,3,5]))
You can fill your smaller array with zeroes, then use zip. inout means that arrays are mutable, or you can make the copy of function parameters inside the function to make them mutable.
private func combineArrays(array1: inout [Int], array2: inout [Int]) -> [Int] {
let maxSize = max(array1.count, array2.count)
if (array1.count > array2.count) {
array2.append(contentsOf: [Int](repeating: 0, count: maxSize - array2.count))
} else if (array2.count > array1.count) {
array1.append(contentsOf: [Int](repeating: 0, count: maxSize - array1.count))
}
return zip(array1, array2).map(+)
}
//javaScript
var array1 = [1,2,3,4,5];
var array2 = [9,7,8,6,5,6,7];
let a= array1.length;
let b = array2.length;
var array3 = [];
let c = a>b?a:b;
for(let i=0; i<c; i++){
if(i < a && i < b){
array3.push(array1[i] + array2[i]);
} else if(i >= a){
array3.push(array2[i])
} else{
array3.push(array1[i])
}
}
console.log(array3)
I'd like to write an extension on CollectionType in Swift that will find the x objects after an object in an array. Obviously it needs be protected to work even if there are no objects after the item.
In my head the signatures something like this:
func itemsAfterItem(item: T, limit: Int?) -> [T]
I can't figure out how to implement it though, could someone help?
A possible implementation for arbitrary collections of
Equatable elements (explanations inline). The main
challenge is to get the parameter types and constraints right.
extension CollectionType where Generator.Element: Equatable,
SubSequence.Generator.Element == Generator.Element {
func itemsAfterItem(item: Generator.Element, limit: Index.Distance?) -> [Generator.Element] {
if let idx = indexOf(item) where idx != endIndex {
// Start after the given item:
let from = idx.advancedBy(1)
// Up to min(from + limit, endIndex):
let to = limit.map { from.advancedBy($0, limit: endIndex) } ?? endIndex
// Return slice as an array:
return Array(self[from..<to])
} else {
// Item not found, or only at the last position.
return []
}
}
}
Understanding the
let to = limit.map { from.advancedBy($0, limit: endIndex) } ?? endIndex
part is left as an exercise to the reader :)
Examples:
[1, 2, 3, 4, 5, 6].itemsAfterItem(2, limit: 2) // [3, 4]
["x", "y", "z"].itemsAfterItem("y", limit: 4) // ["z"]
[1, 2, 3].itemsAfterItem(7, limit: 4) // []
[1.1, 2.2, 3.3].itemsAfterItem(1.1, limit: nil) // [2.2, 3.3]
Example for a non-array collection:
"abcdef".characters.itemsAfterItem("b", limit: 2) // ["c", "d"]
Just because I liked the challenge ;)
extension Array where Element : Equatable {
func itemsAfterItem(item: Element, limit: Int? = nil) -> [Element] {
if let from = self.indexOf(item) where from < self.count - 1 {
if let limit = limit where from + limit < self.count {
return Array(self[from+1...from + limit])
}
return Array(self[from+1...self.count-1])
} else {
return []
}
}
}
For the input
let arr = [1, 2, 4, 6, 9]
It results in
arr.itemsAfterItem(2) // [4, 6, 9]
arr.itemsAfterItem(2, limit: 2) // [4, 6]
arr.itemsAfterItem(2, limit: 100) // [4, 6, 9]
arr.itemsAfterItem(9, limit: 2) // []
arr.itemsAfterItem(3, limit: 100) // []
I think you can try this:
func itemsAfterItem(item: T, limit: Int?) -> [T] {
var counter: Int = 0
var isAfter: Bool = false
let array = [T]()
let newLimit = limit != nil ? limit : myArray.count
for tmpItem in myArray {
if tmpItem == T {
isAfter = true
}
if isAfter && counter < limit {
array.append(tmpItem)
counter += 1
}
}
}
This function will put your T item at the start of the array.
I've not test this function
I am looking for iterator to infinitely iterate collection in a loop mode. So that when end index of collection is reached, then iterator should return element at start index.
The following solution seems working, but I hope it can be made in a better way.
public struct LoopIterator<T: Collection>: IteratorProtocol {
private let collection: T
private var startIndexOffset: T.IndexDistance
public init(collection: T) {
self.collection = collection
startIndexOffset = 0
}
public mutating func next() -> T.Iterator.Element? {
guard !collection.isEmpty else {
return nil
}
let index = collection.index(collection.startIndex, offsetBy: startIndexOffset)
startIndexOffset += T.IndexDistance(1)
if startIndexOffset >= collection.count {
startIndexOffset = 0
}
return collection[index]
}
}
extension Array {
func makeLoopIterator() -> LoopIterator<Array> {
return LoopIterator(collection: self)
}
}
// Testing...
// Will print: 1, 2, 3, 1, 2, 3
var it = [1, 2, 3].makeLoopIterator()
for _ in 0..<6 {
print(it.next())
}
Is it a right way to do custom iterator? What can be improved?
Thanks!
In Swift 3 (which you're using), indexes are intended to be advanced by the collection itself. With that, you can simplify this as follows:
public struct LoopIterator<Base: Collection>: IteratorProtocol {
private let collection: Base
private var index: Base.Index
public init(collection: Base) {
self.collection = collection
self.index = collection.startIndex
}
public mutating func next() -> Base.Iterator.Element? {
guard !collection.isEmpty else {
return nil
}
let result = collection[index]
collection.formIndex(after: &index) // (*) See discussion below
if index == collection.endIndex {
index = collection.startIndex
}
return result
}
}
Now we simply move the index forward, and if it now points to the end, reset it to the beginning. No need for count or IndexDistance.
Note that I've used formIndex here, which exists to improve performance in somewhat obscure cases (specifically around AnyIndex) since your Iterator works on any Collection (and therefore any Index). The simpler version would be index = collection.index(after: index), and that may be better in most cases.
For all the gory details on Swift 3 indices, see SE-0065.
With Swift 5, you can use one of the following examples in order to solve your problem.
#1. Using AnyIterator
As an alternative to creating a new type that conforms to IteratorProtocol, you can use AnyIterator. The following code, based on Rob Napier's answer, shows how to use it:
extension Array {
func makeInfiniteLoopIterator() -> AnyIterator<Element> {
var index = self.startIndex
return AnyIterator({
if self.isEmpty {
return nil
}
let result = self[index]
index = self.index(after: index)
if index == self.endIndex {
index = self.startIndex
}
return result
})
}
}
Usage:
let infiniteLoopIterator = [1, 2, 3].makeInfiniteLoopIterator()
for val in infiniteLoopIterator.prefix(5) {
print(val)
}
/*
prints:
1
2
3
1
2
*/
let infiniteLoopIterator = [1, 2, 3].makeInfiniteLoopIterator()
let array = Array(infiniteLoopIterator.prefix(7))
print(array) // prints: [1, 2, 3, 1, 2, 3, 1]
let infiniteLoopIterator = [1, 2, 3].makeInfiniteLoopIterator()
let val1 = infiniteLoopIterator.next()
let val2 = infiniteLoopIterator.next()
let val3 = infiniteLoopIterator.next()
let val4 = infiniteLoopIterator.next()
print(String(describing: val1)) // prints: Optional(1)
print(String(describing: val2)) // prints: Optional(2)
print(String(describing: val3)) // prints: Optional(3)
print(String(describing: val4)) // prints: Optional(1)
#2. Using AnySequence
A similar approach is to use AnySequence:
extension Array {
func makeInfiniteSequence() -> AnySequence<Element> {
return AnySequence({ () -> AnyIterator<Element> in
var index = self.startIndex
return AnyIterator({
if self.isEmpty {
return nil
}
let result = self[index]
self.formIndex(after: &index) // alternative to: index = self.index(after: index)
if index == self.endIndex {
index = self.startIndex
}
return result
})
})
}
}
Usage:
let infiniteSequence = [1, 2, 3].makeInfiniteSequence()
for val in infiniteSequence.prefix(5) {
print(val)
}
/*
prints:
1
2
3
1
2
*/
let infiniteSequence = [1, 2, 3].makeInfiniteSequence()
let array = Array(infiniteSequence.prefix(7))
print(array) // prints: [1, 2, 3, 1, 2, 3, 1]
Say I have an array [1, 2, 3, 4, 5]. How can I iterate two at a time?
Iteration 1: (1, 2)
Iteration 2: (3, 4)
Iteration 3: (5, nil)
You can use a progression loop called stride(to:, by:) to iterate over your elements every n elements:
let array = Array(1...5)
let pairs = stride(from: 0, to: array.endIndex, by: 2).map {
(array[$0], $0 < array.index(before: array.endIndex) ? array[$0.advanced(by: 1)] : nil)
} // [(.0 1, {some 2}), (.0 3, {some 4}), (.0 5, nil)]
print(pairs) // "[(1, Optional(2)), (3, Optional(4)), (5, nil)]\n"
To iterate your collection subsequences instead of tuples:
extension Collection {
func unfoldSubSequences(limitedTo maxLength: Int) -> UnfoldSequence<SubSequence,Index> {
sequence(state: startIndex) { start in
guard start < self.endIndex else { return nil }
let end = self.index(start, offsetBy: maxLength, limitedBy: self.endIndex) ?? self.endIndex
defer { start = end }
return self[start..<end]
}
}
}
let array = Array(1...5)
for subsequence in array.unfoldSubSequences(limitedTo: 2) {
print(subsequence) // [1, 2] [3, 4] [5]
}
This would work on any kind of collection:
let string = "12345"
for substring in string.unfoldSubSequences(limitedTo: 2) {
print(substring) // "12" "34" "5"
}
You can use sequence() and the iterator's next() method to iterate
over pairs of consecutive elements. This works for arbitrary sequences,
not only arrays:
let a = "ABCDE"
for pair in sequence(state: a.makeIterator(), next: { it in
it.next().map { ($0, it.next()) }
}) {
print(pair)
}
Output:
("A", Optional("B"))
("C", Optional("D"))
("E", nil)
The “outer” it.next() yields the elements at even positions, or nil
(in which case it.next().map { } evaluates to nil as well, and the
sequence terminates). The “inner” it.next() yields the elements
at odd positions or nil.
As an extension method for arbitrary sequences:
extension Sequence {
func pairs() -> AnyIterator<(Element, Element?)> {
return AnyIterator(sequence(state: makeIterator(), next: { it in
it.next().map { ($0, it.next()) }
}))
}
}
Example:
let seq = (1...).prefix(5)
for pair in seq.pairs() { print(pair) }
Note that the pairs are generated lazily, no intermediate array
is created. If you want an array with all pairs then
let pairs = Array([1, 2, 3, 4, 5].pairs())
print(pairs) // [(1, Optional(2)), (3, Optional(4)), (5, nil)]
does the job.
This is not identically what was asked, but I use an extension on Sequence that generates an array of arrays chunking the original sequence by any desired size:
extension Sequence {
func clump(by clumpsize:Int) -> [[Element]] {
let slices : [[Element]] = self.reduce(into:[]) {
memo, cur in
if memo.count == 0 {
return memo.append([cur])
}
if memo.last!.count < clumpsize {
memo.append(memo.removeLast() + [cur])
} else {
memo.append([cur])
}
}
return slices
}
}
So [1, 2, 3, 4, 5].clump(by:2) yields [[1, 2], [3, 4], [5]] and now you can iterate through that if you like.
Extension to split the array.
extension Array {
func chunked(into size: Int) -> [[Element]] {
return stride(from: 0, to: count, by: size).map {
Array(self[$0 ..< Swift.min($0 + size, count)]) }
}
}
let result = [1...10].chunked(into: 2)
I personally dislike looping through half the list (mainly because of dividing), so here is how I like to do it:
let array = [1,2,3,4,5];
var i = 0;
while i < array.count {
var a = array[i];
var b : Int? = nil;
if i + 1 < array.count {
b = array[i+1];
}
print("(\(a), \(b))");
i += 2;
}
You loop through the array by incrementing by 2.
If you want to have nil in the element, you need to use optionals.
If the array would have an even number of elements, you would be able to write something like this:
for i in 0..<arr.count/2 {
print(arr[2*i...2*i+1])
}
However that's not always the case. Moreover, nil is not always compatible with the type of elements in array, like the one in your example (nil is not compatible with Int, only with Int?).
Another solution would be to extend Array and add a pair() method, which returns a tuple (tuples can be heterogenous). You can use pair to walk within all pairs in the array, or, you can extend even more the Array struct and add pairs() that return an array of tuples. Note that since the second element in the tuple is an Optional you'll need to unwrap it before use.
extension Array {
func pair(at i: Index) -> (Element, Element?) {
return (self[i], i < self.count - 1 ? self[i+1] : nil)
}
func pairs() -> [(Element, Element?)] {
guard !isEmpty else { return [] }
var result = [(Element, Element?)]()
for i in 0...arr.count/2 {
result.append(self.pair(at: 2*i))
}
return result
}
}
let arr = [1, 2, 3, 4, 5]
for i in 0...arr.count/2 {
print(arr.pair(at: 2*i))
}
for pair in arr.pairs() {
print(pair)
}
Update Both above solutions can be simplified by using map instead of manually looping:
let pairs = (0..<arr.count/2).map { (arr[$0*2], arr[$0*2+1]) }
print(pairs) // prints [(1, 2), (3, 4)]
or, for the Array extension:
extension Array {
func pair(at i: Index) -> (Element, Element?) {
return (self[i], i < self.count - 1 ? self[i+1] : nil)
}
func pairs() -> [(Element, Element?)] {
guard !isEmpty else { return [] }
return (0..<(arr.count/2 + arr.count%2)).map { pair(at: $0*2) }
}
}
let arr = [1, 2, 3, 4, 5]
print(arr.pairs()) // [(1, Optional(2)), (3, Optional(4)), (5, nil)]
You can extend Collection instead, to have this pair functionality available for all collections:
extension Collection {
func pairs() -> [(Element, Element?)] {
guard !isEmpty else { return [] }
return (0..<count/2+count%2).map {
let i1 = index(startIndex, offsetBy: $0*2)
let i2 = index(after: i1)
return (self[i1], i2 < endIndex ? self[i2] : nil)
}
}
}
Here is my solution with one reduce and a few guards
extension Array {
var touplesOfTwo: [(Element,Element?)] {
self.reduce(into: [(Element,Element?)]()) {
guard let last = $0.last else { $0.append( ($1,nil) ); return }
let lastIndex = $0.count - 1
guard let _ = last.1 else { $0[lastIndex].1 = $1; return }
$0.append( ($1,nil) )
}
}
}
let list = [1,4,3,7,2,9,6,5]
let queues = list.map { $0 }
let touplesList = queues.touplesOfTwo
print("\(touplesList)")
// [(1, Optional(4)), (3, Optional(7)), (2, Optional(9)), (6, Optional(5))]
One approach would be to encapsulate the array in a class. The return values for getting pairs of items would be optionals to protect against out-of-range calls.
Example:
class Pairs {
let source = [1, 2, 3, 4, 5] // Or set with init()
var offset = 0
func nextpair() -> (Int?, Int?) {
var first: Int? = nil
var second: Int? = nil
if offset < source.count {
first = source[offset]
offset++
}
if offset < source.count {
second = source[offset]
offset++
}
return (first, second)
}
}
Given the following:
struct Weekdays: OptionSetType {
let rawValue: Int
init(rawValue: Int) { self.rawValue = rawValue }
static let Monday = Weekdays(rawValue: 1)
static let Tuesday = Weekdays(rawValue: 2)
static let Wednesday = Weekdays(rawValue: 4)
static let Thursday = Weekdays(rawValue: 8)
static let allOptions: [Weekdays] = [.Monday, .Tuesday, .Wednesday, .Thursday]
}
I can convert an array of Ints into a Weekdays object by doing this:
let arr = [1, 4]
let weekdays = arr.reduce(Weekdays()) { $0.union(Weekdays(rawValue: $1)) }
My question is, how do I take a Weekdays object and convert it into an array of Ints?
(Not necessarily better, but a different way to look at it and slightly
more general).
OptionSetType inherits from RawRepresentable and therefore can be
converted from and to the associated raw type, which in your case is
Int.
So the "missing link" is a conversion between the raw value (e.g. 5)
and an integer array of the bitwise components (e.g. [1, 4]).
This can be done with an Int extension method:
extension Int {
init(bitComponents : [Int]) {
self = bitComponents.reduce(0, combine: (+))
}
func bitComponents() -> [Int] {
return (0 ..< 8*sizeof(Int)).map( { 1 << $0 }).filter( { self & $0 != 0 } )
}
}
Then your conversion from an array to a Weekdays object becomes
let arr : [Int] = [1, 4]
let weekdays = Weekdays(rawValue: Int(bitComponents: arr))
print(weekdays)
// app.Weekdays(rawValue: 5)
and the reverse conversion
let array = weekdays.rawValue.bitComponents()
print(array)
// [1, 4]
Advantages:
The explicit definition of allOptions: is not needed.
It can be applied to other option set types (which have Int
as a raw value).
One could also try to define the conversions as a protocol extension,
e.g. of IntegerType, so that the same works with other integer raw types as well. However, this seems to be a bit complicated/ugly
because the left shift operator << is not part of the
IntegerType (or any) protocol.
Update for Swift 3:
extension Int {
init(bitComponents : [Int]) {
self = bitComponents.reduce(0, +)
}
func bitComponents() -> [Int] {
return (0 ..< 8*MemoryLayout<Int>.size).map( { 1 << $0 }).filter( { self & $0 != 0 } )
}
}
Not exactly answering the question, but might be useful to others. Based on Martin's answer I extract back the component objects:
extension FixedWidthInteger {
init(bitComponents : [Self]) {
self = bitComponents.reduce(0, +)
}
var bitComponents : [Self] {
(0 ..< Self.bitWidth).map { 1 << $0 } .filter { self & $0 != 0 }
}
}
extension OptionSet where RawValue: FixedWidthInteger, Self == Self.Element {
var components : [Self] { rawValue.bitComponents.map(Self.init) }
}
As I was writing the question, I figured it out:
let array = Weekdays.allOptions.filter { weekdays.contains($0) }.map { $0.rawValue }
Is there a better way?
You can improve the context of your extension by defining it conditionally on OptionSet.
extension OptionSet where RawValue: UnsignedInteger {
var individualCases: [Self] {
return (0..<(8 * MemoryLayout<RawValue>.size))
.map { bitsToShift in RawValue(1 << bitsToShift) } // Get every possible single-bit flag
.filter { (powerOfTwo: RawValue) -> Bool in rawValue & powerOfTwo != 0 } // filter out only the cases the receiver contains
.map { Self(rawValue: $0) } // create the `OptionSet` (single bit) type
}
}
let weekdays = Weekdays(rawValue: 0b11111)
weekdays.individualCases.map { $0.rawValue } // [1, 2, 4, 8, 16]
A warning: On my 13" 2019 MacBookPro, I had to provide all of the explicit types above to keep the methods type checking under 1500ms in Swift 5.0.
Thanks to MartinR for the inspiration to loop of the memory layout size.
To complete the example, I've updated the weekday type to Swift 5 below, and explicitly used the UInt8 type to make the individualCases more efficient. With UInt, it would loop over the first map and filter 64 times each, with UInt8 it only loops 8 times.
struct Weekdays: OptionSet {
let rawValue: UInt8
static let Monday = Weekdays(rawValue: 1)
static let Tuesday = Weekdays(rawValue: 2)
static let Wednesday = Weekdays(rawValue: 4)
static let Thursday = Weekdays(rawValue: 8)
static let Friday = Weekdays(rawValue: 16)
}