Lua使用luastring.translate时子串匹配冲突导致翻译错误如何解决

问题背景 #

开发魔方转动公式转译工具，目标是将标准公式序列R U R' U'转译为自定义格式r^ u< rv u>。实际调用luastring库的translate接口执行替换时出现匹配逻辑错误：带撇号的转动标记（如R'）包含普通短标记子串R，接口默认短串优先匹配，输入R U R' U'时会错误输出r^ u< r^' u<'，撇号残留，结果不符合预期。
此前已尝试的实现思路：

• 编写自定义split字符串分割函数
• 定义基础转动标记数组moves与对应目标标记数组neomoves
• 构建translation_table翻译映射表
  但始终无法实现长标记（带'的转动符号）优先匹配，luastring库未提供相关匹配优先级配置，公开技术社区也未找到对应解法，原有测试代码如下：

io.write("put in your algorithm: ")
local alg = io.read()

function split(str, pat)
    local t = {}
    local fpat = "(.-)" .. pat
    local last_end = 1
    local s, e, cap = str:find(fpat, 1)
    while s do
        if s ~= 1 or cap ~= "" then table.insert(t, cap) end
        last_end = e + 1
        s, e, cap = str:find(fpat, last_end)
    end
    if last_end <= #str then
        cap = str:sub(last_end)
         table.insert(t, cap)
    end
    return t
end

-- table stuff so we can see
local moves = {"R", "U", "L", "F", "D", "B", "M", "E", "S", "R'", "U'", "L'", "F'", "D'", "B'", "M'", "E'", "S'"}
local neomoves = {"r^", "u<", "lv", "f>", "d>", "b<", "mv", "e>", "s>", "rv", "u>", "l^", "f<", "d<", "b>", "m^", "e<", "s<"}
local string = require("luastring")

local translation_table = {
    ["R'"] = "rv", ["U'"] = "u<", ["L'"] = "lv", ["F'"] = "f>", ["D'"] = "d>", ["B'"] = "b<", ["M'"] = "mv", ["E'"] = "e>", ["S'"] = "s>", ["R"] = "r^", ["U"] = "u<", ["L"] = "lv", ["F"] = "f>", ["D"] = "d>", ["B"] = "b<", ["M"] = "mv", ["E"] = "e>", ["S"] = "s>"
}

-- translate the moves to neomoves
local translated = string.translate(alg, translation_table)
print(translated)

注：原有代码的映射表本身存在笔误，U'对应的目标值按照neomoves数组定义应为u>，原代码错写为u<。

故障原因 #

luastring库的translate接口采用逐字符顺序替换逻辑，没有长串优先匹配机制，遍历到R'的首字符R时会直接命中短标记R的替换规则，不会向后检查是否存在更长的合法标记，最终导致撇号残留、转译错误。

修复方案 #

放弃依赖luastring的translate接口，手动实现逐位扫描+长标记优先匹配逻辑，不需要复杂的字符串分割：

1. 将所有合法转动标记按长度从大到小排序，保证2字符的带撇标记排在1字符普通标记前面
2. 逐位遍历输入字符串，跳过空格分隔符，每次从当前位置优先尝试匹配最长的合法标记
3. 匹配成功后追加对应转译结果，指针向后移动对应标记的长度，直到遍历完成整个输入

修复后的可运行代码如下：

io.write("put in your algorithm: ")
local alg = io.read()

-- 对齐moves与neomoves的正确映射关系
local translation_table = {
    ["R'"] = "rv", ["U'"] = "u>", ["L'"] = "l^", ["F'"] = "f<", ["D'"] = "d<", ["B'"] = "b>", ["M'"] = "m^", ["E'"] = "e<", ["S'"] = "s<",
    ["R"] = "r^", ["U"] = "u<", ["L"] = "lv", ["F"] = "f>", ["D"] = "d>", ["B"] = "b<", ["M"] = "mv", ["E"] = "e>", ["S"] = "s>"
}

-- 所有标记按长度降序排列，确保长标记优先匹配
local sorted_moves = {}
for move, _ in pairs(translation_table) do
    table.insert(sorted_moves, move)
end
table.sort(sorted_moves, function(a, b) return #a > #b end)

local result = {}
local pos = 1
local alg_len = #alg
while pos <= alg_len do
    -- 跳过空格分隔符
    if alg:sub(pos, pos) == " " then
        pos = pos + 1
        goto continue
    end
    -- 按优先级匹配标记
    local is_matched = false
    for _, move in ipairs(sorted_moves) do
        local move_len = #move
        if pos + move_len - 1 <= alg_len and alg:sub(pos, pos + move_len - 1) == move then
            table.insert(result, translation_table[move])
            pos = pos + move_len
            is_matched = true
            break
        end
    end
    -- 非法字符直接跳过，可按需改为抛出错误
    if not is_matched then
        pos = pos + 1
    end
    ::continue::
end

-- 拼接结果输出
local translated = table.concat(result, " ")
print(translated)

代码测试效果：输入R U R' U'时，将正确输出r^ u< rv u>，完全符合预期。后续如果需要新增更长的标记（如R2、r双层转、转体标记等），只需要在translation_table中添加对应键值对，排序逻辑会自动处理匹配优先级，不会再出现子串冲突问题。

内容的提问来源于stack exchange，提问作者Drogobo