Lua UTF-8截断：首字节前缀定的字符宽度

flowchart LR
    A["字节游标 i = 1"] --> B{"string.byte(s, i)<br/>读首字节"}
    B -- "≤127 → 宽 1" --> C["i = i + 1"]
    B -- "192~223 → 宽 2" --> D["i = i + 2"]
    B -- "224~239 → 宽 3" --> E["i = i + 3"]
    C --> F{"i ≤ #str ?"}
    D --> F
    E --> F
    F -- 是 --> B
    F -- 否 --> G["遍历结束<br/>字符数 = 跳跃次数"]

local function chsize(byte)
    if not byte then
        return 0
    elseif byte > 240 then
        return 4
    elseif byte > 225 then
        return 3
    elseif byte > 192 then
        return 2
    else
        return 1
    end
end

-- 计算真实字数: utf8len("1你好") => 3
function UTF8Length(str)
    local len, i = 0, 1
    while i <= #str do
        i = i + chsize(string.byte(str, i))
        len = len + 1
    end
    return len
end

-- 按字符数截断: utf8sub("你好ABC", 2, 3) => "好AB"
function UTF8Sub(str, startChar, numChars)
    local i = 1
    while startChar > 1 do                       -- 先游标走到起始字符
        i = i + chsize(string.byte(str, i))
        startChar = startChar - 1
    end
    local j = i
    while numChars > 0 and j <= #str do          -- 再数 numChars 个字符
        j = j + chsize(string.byte(str, j))
        numChars = numChars - 1
    end
    return str:sub(i, j - 1)                     -- 落点必是字符边界
end