UTF-8 support for Nix
Strings in Nix are byte strings, and builtin functions like substring
(and by extension some lib functions in nixpkgs)
processes bytes instead of UTF-8 code points.
That means these functions can create invalid strings when given strings with UTF-8.
This library basically allows you to convert it to a list of UTF-8 code points.
Try it out with flakes
nix repl github:figsoda/utf8#lib --extra-experimental-features "flakes nix-command repl-flake"or locally
nix repl -f .Type: String -> [ String ]
Split a string into a list of code points
nix-repl> chars "你好，世界！"
[ "你" "好" "，" "世" "界" "！" ]
Type: String -> String
Return the first code point of the string
nix-repl> head "你好，世界！"
"你"
Type: String -> String
Return the string without the first code point
nix-repl> tail "你好，世界！"
"好，世界！"
Type: String -> Int
Return the number of code points in the string
nix-repl> length "你好，世界！"
6
nix run ./dev # regenerate table.nix
nix develop ./dev
namaka check # run tests
namaka review # review pending snapshots