如何在OCaml中快速将UTF-8字符串拆分为字符？

OCaml的字符串本质是字节序列，char类型仅表示单字节值，标准库确实没有直接提供UTF-8解码为Unicode字符的内置函数，你可以通过以下两种方式解决：

方案1：手动实现UTF-8解码逻辑 #

基于UTF-8的编码规则，自己编写函数解析字节序列，组合成Unicode码点（使用标准库Uchar类型表示）：

let rec utf8_decode s i acc =
  if i >= String.length s then List.rev acc
  else
    let c = Char.code s.[i] in
    (* 单字节ASCII字符 *)
    if c < 0x80 then
      utf8_decode s (i + 1) (Uchar.of_int c :: acc)
    (* 双字节UTF-8序列 *)
    else if c < 0xE0 then
      if i + 1 >= String.length s then failwith "无效的UTF-8序列"
      else
        let c1 = Char.code s.[i+1] in
        if (c1 land 0xC0) <> 0x80 then failwith "无效的UTF-8序列"
        let code = ((c land 0x1F) lsl 6) lor (c1 land 0x3F) in
        utf8_decode s (i + 2) (Uchar.of_int code :: acc)
    (* 三字节UTF-8序列 *)
    else if c < 0xF0 then
      if i + 2 >= String.length s then failwith "无效的UTF-8序列"
      else
        let c1 = Char.code s.[i+1] in
        let c2 = Char.code s.[i+2] in
        if (c1 land 0xC0) <> 0x80 || (c2 land 0xC0) <> 0x80 then failwith "无效的UTF-8序列"
        let code = ((c land 0x0F) lsl 12) lor ((c1 land 0x3F) lsl 6) lor (c2 land 0x3F) in
        utf8_decode s (i + 3) (Uchar.of_int code :: acc)
    (* 四字节UTF-8序列 *)
    else if c < 0xF8 then
      if i + 3 >= String.length s then failwith "无效的UTF-8序列"
      else
        let c1 = Char.code s.[i+1] in
        let c2 = Char.code s.[i+2] in
        let c3 = Char.code s.[i+3] in
        if (c1 land 0xC0) <> 0x80 || (c2 land 0xC0) <> 0x80 || (c3 land 0xC0) <> 0x80 then failwith "无效的UTF-8序列"
        let code = ((c land 0x07) lsl 18) lor ((c1 land 0x3F) lsl 12) lor ((c2 land 0x3F) lsl 6) lor (c3 land 0x3F) in
        if code > 0x10FFFF then failwith "超出Unicode码点范围"
        utf8_decode s (i + 4) (Uchar.of_int code :: acc)
    else
      failwith "无效的UTF-8起始字节"

let utf8_split s = utf8_decode s 0 []

调用utf8_split "你好🌍"会返回包含对应Unicode码点的Uchar.t list。

方案2：使用第三方库 #

如果不想手动处理边界情况，可以用成熟的UTF-8处理库uutf：

1. 先通过opam安装：

opam install uutf

2. 编写拆分函数：

open Uutf

let utf8_split s =
  let decoder = decoder (`String s) in
  let rec decode acc =
    match decode decoder with
    | `Uchar u -> decode (u :: acc)
    | `End -> List.rev acc
    | `Malformed _ -> failwith "无效的UTF-8字符串"
  in
  decode []

这个库会自动处理各种UTF-8编码细节，错误处理更完善，适合生产环境使用。

内容的提问来源于stack exchange，提问作者user8203231