Strings & Unicode in Swift

StringsAndUnicode.swift

/*:
# Strings and Unicode
*/
import Foundation

/*: Some things we can't do */

//let ❤ = "❤" // error: Expected a pattern
//let ☆ = "☆" // error: Expected a pattern

/*: Some things we probbaly shouldn't be able to do */

let e⃝⃞⃟⃠⃣⃤⃥⃦⃪⃧꙰꙲꙱ = "e⃝⃞⃟⃠⃣⃤⃥⃦⃪⃧꙰꙲꙱" // Yeah, this valid

/*:
## Some Extensions
*/
extension String {
    var unicodeName: String {
        let cfstr = NSMutableString(string: self) as CFMutableString
        var range = CFRangeMake(0, CFStringGetLength(cfstr))
        CFStringTransform(cfstr, &range, kCFStringTransformToUnicodeName, false)
        return String(cfstr)
    }
}

"😄".unicodeName // \\N{SMILING FACE WITH OPEN MOUTH AND SMILING EYES}

/*
// This crashes the playground
extension UnicodeScalar : IntegerLiteralConvertible {
    init(integerLiteral value: Self.IntegerLiteralType) {
        self = UnicodeScalar(value)
    }
}
*/

extension UnicodeScalar {
    var codePointString: String {
        return "U+\(String(self.value, radix: 16))"
    }
}

/*: ## Simple string, Latin character */

let a₁ = "a"
String("\u{61}") // "a"
a₁.characters.count // 1
a₁.unicodeScalars.count // 1
a₁.utf16.count // 1
a₁.utf8.count // 1
let a₂: UnicodeScalar = "a" // 97
a₂.codePointString // U+61
String(a₂.value, radix: 16) // 61
String(UnicodeScalar(0x61)) // "A"

/*: 
## Some Chineese character
This is the Chineese character for "Toy" or "Toys" (accoring to Google translate).
*/

let 玩₁ = "玩"
String("\u{73a9}") // 玩
玩₁.characters.count // 1
玩₁.unicodeScalars.count // 1
玩₁.utf16.count // 1
玩₁.utf8.count // 3
let 玩₂: UnicodeScalar = "玩" // 29609
String(玩₂.value, radix: 16) // 73a9
玩₁.unicodeName // \\N{CJK UNIFIED IDEOGRAPH-73A9}
String(UnicodeScalar(0x73a9)) // "玩"

/*:
## Grapheme Cluster: Combining Characters
*/

let কী₁ = "কী" // U+0995 BENGALI LETTER KA
String("\u{0995}\u{09C0}") // "কী"
কী₁.characters.count // 1
কী₁.unicodeScalars.count // 2
কী₁.utf16.count // 2
কী₁.utf8.count // 6
কী₁.unicodeScalars.dropFirst() // ী // U+09C0 BENGALI VOWEL SIGN II
কী₁.unicodeScalars.dropLast() // ক
//let কী₂: UnicodeScalar = "কী" // error: cannot convert value of type 'String' to specified type 'UnicodeScalar'
কী₁.unicodeName // \N{BENGALI LETTER KA}\N{BENGALI VOWEL SIGN II}
String(UnicodeScalar(0x0995)) // "ক"
String(UnicodeScalar(0x09C0)) // "ী"

/*: 
## Grapheme Cluster: Emoji
*/

let 🇺🇸 = "🇺🇸"
🇺🇸.characters.count // 1
🇺🇸.unicodeScalars.count // 2
🇺🇸.utf16.count // 4
🇺🇸.utf8.count // 8
🇺🇸.unicodeScalars.dropFirst() // 🇸
🇺🇸.unicodeScalars.dropLast() // 🇺
🇺🇸.unicodeName // \N{REGIONAL INDICATOR SYMBOL LETTER U}\N{REGIONAL INDICATOR SYMBOL LETTER S}

/*:
## Zalgo Text
*/

let zalgo = "e⃝⃞⃟⃠⃣⃤⃥⃦⃪⃧꙰꙲꙱"
for s in zalgo.unicodeScalars {
    print("\(s)\t\(s.codePointString)\t\(String(s).unicodeName)")
}

/*
e	U+65	e
⃝	U+20dd	\N{COMBINING ENCLOSING CIRCLE}
⃞	U+20de	\N{COMBINING ENCLOSING SQUARE}
⃟	U+20df	\N{COMBINING ENCLOSING DIAMOND}
⃠	U+20e0	\N{COMBINING ENCLOSING CIRCLE BACKSLASH}
⃣	U+20e3	\N{COMBINING ENCLOSING KEYCAP}
⃤	U+20e4	\N{COMBINING ENCLOSING UPWARD POINTING TRIANGLE}
⃥	U+20e5	\N{COMBINING REVERSE SOLIDUS OVERLAY}
⃦	U+20e6	\N{COMBINING DOUBLE VERTICAL STROKE OVERLAY}
⃧	U+20e7	\N{COMBINING ANNUITY SYMBOL}
⃪	U+20ea	\N{COMBINING LEFTWARDS ARROW OVERLAY}
꙰	U+a670	\N{COMBINING CYRILLIC TEN MILLIONS SIGN}
꙲	U+a672	\N{COMBINING CYRILLIC THOUSAND MILLIONS SIGN}
꙱	U+a671	\N{COMBINING CYRILLIC HUNDRED MILLIONS SIGN}
*/

for s in zalgo.unicodeScalars { // e ⃝ ⃞ ⃟ ⃠ ⃣ ⃤ ⃥ ⃦ ⃧ ⃪ ꙰ ꙲ ꙱
    print("\(s) ", terminator: "")
}

let combined = "e\u{20dd}\u{20e7}" // "e⃝⃧"

/*:
## See Also

* [Why is Swift's String API So Hard? - mikeash.com](https://www.mikeash.com/pyblog/friday-qa-2015-11-06-why-is-swifts-string-api-so-hard.html)
* [What is the difference between ‘combining characters’ and ‘grapheme extenders’ in Unicode? - Stack Overflow](http://stackoverflow.com/questions/21722729)
* [Strings in Swift - Ole Begemann](http://oleb.net/blog/2014/07/swift-strings)
* [CFString​Transform - NSHipster](http://nshipster.com/cfstringtransform/)
* [Re: Origin of the U+nnnn notation](http://unicode.org/mail-arch/unicode-ml/y2005-m11/0060.html)
*/