A Rust library providing generic string types parameterized by encoding.

stringly provides String<E> and Str<E> types that work like std's String and str, but are generic over their encoding. This allows you to:

• Write encoding-agnostic code that works with UTF-8, UTF-16, UTF-32, and legacy encodings
• Safely transcode between encodings
• Use the same API you already know from std

• Generic string types: String<E> (owned) and Str<E> (borrowed) parameterized by encoding
• Full std API compatibility: The same methods as std::string::String and str
• Pattern matching: Generic Pattern trait for searching, splitting, and matching
• Transcoding: Convert between any encodings via chars() iterators
• Feature-gated encodings: Enable only the encodings you need

CESU-8, GB18030, UTF-16BE, UTF-16LE, UTF-32BE, UTF-32LE, UTF-7, UTF-7-IMAP, UTF-8, UTF-EBCDIC, ATARIST, Big5, CP037, CP1006, CP1026, CP424, CP437, CP500, CP737, CP775, CP850, CP852, CP855, CP856, CP857, CP860, CP861, CP862, CP863, CP864, CP865, CP866, CP869, CP875, EUC-JP, EUC-KR, ISO-2022-JP, ISO-8859-1, ISO-8859-10, ISO-8859-11, ISO-8859-13, ISO-8859-14, ISO-8859-15, ISO-8859-16, ISO-8859-2, ISO-8859-3, ISO-8859-4, ISO-8859-5, ISO-8859-6, ISO-8859-7, ISO-8859-8, ISO-8859-9, KOI8-R, KOI8-U, KZ1048, MacArabic, MacCeltic, MacCenteuro, MacChinsimp, MacChintrad, MacCroatian, MacCyrillic, MacDevanaga, MacDingbats, MacFarsi, MacGaelic, MacGreek, MacGujarati, MacGurmukhi, MacHebrew, MacIceland, MacInuit, MacJapanese, MacKeyboard, MacKorean, MacRoman, MacRomanian, MacSymbol, MacThai, MacTurkish, Shift_JIS, Windows-1250, Windows-1251, Windows-1252, Windows-1253, Windows-1254, Windows-1255, Windows-1256, Windows-1257, Windows-1258, Windows-874, Windows-932, Windows-936, Windows-949, Windows-950

Add to your Cargo.toml:

[dependencies]
stringly = "0.0.0"

# Enable specific encodings
[features]
default = ["stringly/unicode"]  # UTF-16, UTF-32, GB18030, UTF-7, CESU-8, UTF-EBCDIC

use stringly::{String, Utf8};

// Create a UTF-8 string - same API as std
let s: String<Utf8> = String::from("Hello, world!");
assert!(s.starts_with("Hello"));

use stringly::{Str, Encoding};

// Write functions that work with any encoding
fn word_count<E: Encoding>(s: &Str<E>) -> usize {
    s.split_whitespace().count()
}

use stringly::{String, Utf8, Utf16Le, Iso8859_1};

// Universal encodings can convert infallibly via From
let utf8: String<Utf8> = String::from("Hello");
let utf16: String<Utf16Le> = String::from(utf8);

// Limited encodings may fail if characters aren't representable
let utf8: String<Utf8> = String::from("Hello");
let latin1: Option<String<Iso8859_1>> = utf8.try_transcode();

The library defines three levels of encoding capability:

• Encoding: Base trait for all encodings. Provides character iteration, validation, and basic operations.
• LimitedEncoding: Encodings that can represent a subset of Unicode (e.g., ISO-8859-1).
• UniversalEncoding: Encodings that can represent all of Unicode (UTF-8, UTF-16, UTF-32, GB18030, UTF-7, CESU-8, UTF-EBCDIC).

Licensed under either of Apache License, Version 2.0 or MIT license at your option.