Example of encoding Unicode graphemes as string and taking its length

unicode_example.ml

(* Re: https://twitter.com/llaisdy/status/1558536851560054786?s=20&t=us8J3LvoJTlwVwod5bOqeQ *)

(* Use this if using the REPL, otherwise use dune to build with the library dependency *)
#require "uutf";;

(* Converts an array of ints (Unicode graphemes) into a UTF-8 encoded string. *)
let utf8_to_string uchars =
  let buf = Buffer.create (2 * Array.length uchars) in
  Array.iter (fun uchar -> Uutf.Buffer.add_utf_8 buf (Uchar.of_int uchar)) uchars;
  Buffer.contents buf

(* Calculates number of graphemes in a UTF-8 encoded string. *)
let utf8_len =
  let width_folder len _ = function
  | `Uchar _ -> succ len
  | `Malformed s -> invalid_arg s
  in
  Uutf.String.fold_utf_8 width_folder 0

(* Now we can replicate the examples in the other languages: *)

let s = utf8_to_string [|0x25105; 0x26159; 0x33521; 0x22269; 0x20154|] (* "𥄅𦅙𳔡𢉩𠅔" *)
let len = utf8_len s (* 5 *)