Re: Bug modulo operator on negative numbers

[Flyer31 Test <flyer31@...>]

I do not like the idea, that modulo 3 would provide any other result
than 0...2, also not for negative numbers. I think there are quite a
bit applications where you want a "clearly periodic" cutting of the
number axis, thus "hiding" the "complexity of large numbers", and I
would prefer to hide also the "complexity of negative numbers" in this
context, but this of course might be too philosophical argumentation,
sorry :).

I would regard the "C" result of your example as wrong ... . What C
compiler do you use there - or does this really any C do?

But if like your "C result" so much, you can also quite easily get
this result by Lua, if you write:
c= a > 0 and a%b or -(-a%b)

On Mon, Aug 16, 2021 at 9:33 AM Domingo Alvarez Duarte
<mingodad@gmail.com> wrote:
>


On Mon, Aug 16, 2021 at 9:33 AM Domingo Alvarez Duarte
<mingodad@gmail.com> wrote:
>
> There was some discussion about Lua modulo operator with negative
> numbers but it doesn't seem to be addressed properly yet, see the
> examples bellow.
>
> Example:
>
> ====
>
> local a = 4;
> local b = 3;
> local c = a % b;
> local d = a - c;
>
>
> local a2 = -4;
> local b2 = 3;
> local c2 = a2 % b2;
> local d2 = a2 - c2;
>
> print( a, b, c, d, a2, b2, c2, d2);
>
> ====
>
> Output:
>
> ====
>
> 4    3    1    3    -4    3    2    -6
>
> ====
>
> Code on lua.vm (note the comments):
>
> ====
>
> /*
> ** Integer modulus; return 'm % n'. (Assume that C '%' with
> ** negative operands follows C99 behavior. See previous comment
> ** about luaV_idiv.)
> */
> lua_Integer luaV_mod (lua_State *L, lua_Integer m, lua_Integer n) {
>    if (l_unlikely(l_castS2U(n) + 1u <= 1u)) {  /* special cases: -1 or 0 */
>      if (n == 0)
>        luaG_runerror(L, "attempt to perform 'n%%0'");
>      return 0;   /* m % -1 == 0; avoid overflow with 0x80000...%-1 */
>    }
>    else {
>      lua_Integer r = m % n;
>      if (r != 0 && (r ^ n) < 0)  /* 'm/n' would be non-integer negative? */
>        r += n;  /* correct result for different rounding */
>      return r;
>    }
> }
>
> ====
>
> C example:
>
> ====
>
> #include <stdio.h>
>
> int main(int argc, char *argv[]) {
>
>      int a = 4;
>      int b = 3;
>      int c = a % b;
>      int d = a - c;
>
>
>      int a2 = -4;
>      int b2 = 3;
>      int c2 = a2 % b2;
>      int d2 = a2 - c2;
>
>      printf("%d\t%d\t%d\t%d\t%d\t%d\t%d\t%d\n", a, b, c, d, a2, b2, c2, d2);
>
>      return 0;
> }
>
> ====
>
> C output:
>
> ====
>
> 4    3    1    3    -4    3    -1    -3
>
> ====
>
> Cheers !