String elements in Swift

StringElements.swift

// Demonstrates some rather elementary points about code points vs elements

import Foundation
import Swift

extension String {
    func canonize() -> [Character] {
        return Array(precomposedStringWithCanonicalMapping)
    }    
    static func decanonize(chars: [Character]) -> String {
        return "".join ( chars.map { String($0) })
    }
}

func test(s:String) -> String {
    let c = s.canonize()
    let ß = String.decanonize(c)
    let equal = ß == s ? "equals" : "does not equal"
    return "\(s) has \(s.utf16Count) UTF-16 code points, \(c.count) elements, \(ß.utf16Count) precomposed code points, \(equal)"
}

test("Åström") 
// "Åström has 6 UTF-16 code points, 6 elements, 6 precomposed code points, equals"

test("\u{0041}\u{030A}str\u{006F}\u{0308}m")
// "Åström has 8 UTF-16 code points, 6 elements, 6 precomposed code points, does not equal"

let ö1 = "Åström".canonize()[4]
let ö2 = "\u{0041}\u{030A}str\u{006F}\u{0308}m".canonize()[4]
ö1 == ö2 // true