Lua中实现字符串分割函数split

方法一

function split(input, delimiter)
    input = tostring(input)
    delimiter = tostring(delimiter)
    if (delimiter == "") then return false end
    local pos, arr = 0, {}
    for st, sp in function() return string.find(input, delimiter, pos, true) end do
        table.insert(arr, string.sub(input, pos, st - 1))
        pos = sp + 1
    end
    table.insert(arr, string.sub(input, pos))
    return arr
end

local testStr1 = "hello,world"
local testStr2 = "a,,b,c"
local testStr3 = "a+-b-+c"
local testStr4 = "www.baidu.com"

print(unpack(split(testStr1, ",")))
print(unpack(split(testStr2, ",")))
print(unpack(split(testStr3, "+-")))
print(unpack(split(testStr4, ".")))

-- hello	world
-- a		b	c
-- a	b-+c
-- www	baidu	com

方法二

function split(input, delimiter)
    if type(delimiter) ~= "string" or #delimiter <= 0 then
        return
    end
    local start = 1
    local arr = {}
    while true do
        local pos = string.find(input, delimiter, start, true)
        if not pos then
            break
        end
        table.insert (arr, string.sub (input, start, pos - 1))
        start = pos + string.len (delimiter)
    end
    table.insert (arr, string.sub (input, start))
    return arr
end

local testStr1 = "hello,world"
local testStr2 = "a,,b,c"
local testStr3 = "a+-b-+c"
local testStr4 = "www.baidu.com"

print(unpack(split(testStr1, ",")))
print(unpack(split(testStr2, ",")))
print(unpack(split(testStr3, "+-")))
print(unpack(split(testStr4, ".")))

-- hello	world
-- a		b	c
-- a	b-+c
-- www	baidu	com

ps

在网上看到了不少使用正则表达式来进行分割的，例如：

function split(input, delimiter)
    local arr = {}
    string.gsub(input, '[^' .. delimiter ..']+', function(w) table.insert(arr, w) end)
    return arr
end

local testStr1 = "hello,world"
local testStr2 = "a,,b,c"
local testStr3 = "a+-b-+c"
local testStr4 = "www.baidu.com"

print(unpack(split(testStr1, ",")))
print(unpack(split(testStr2, ",")))  -- 无法正确分割
print(unpack(split(testStr3, "+-")))  -- 无法正确分割
print(unpack(split(testStr4, ".")))

-- hello	world
-- a	b	c
-- a	b	c
-- www	baidu	com

可以看到对于某些字符串是无法正确分割的，所以分割字符串最好还是不要用正则表达式了吧，毕竟 Lua 对正则表达式的支持并不是特别完整！