std.variant benchmark

[Dmitry Olshansky]

On 30-Jul-12 00:11, Dmitry Olshansky wrote:
> I've found something far more evil:
>
> @property bool convertsTo(T)() const
>      {
>          TypeInfo info = typeid(T);
>          return fptr(OpID.testConversion, null, &info) == 0;
>      }
>
> Okay... now let me pull off another piece of rag:
>
> private VariantN opArithmetic(T, string op)(T other)
>      {
>          VariantN result;
>          static if (is(T == VariantN))
>          {
>             /*if (convertsTo!(uint) && other.convertsTo!(uint))
>                  result = mixin("get!(uint) " ~ op ~ " other.get!(uint)");
>              else*/ if (convertsTo!(int) && other.convertsTo!(int))
>                  result = mixin("get!(int) " ~ op ~ " other.get!(int)");
...
Apparently I'm spot on.
Just commenting one extra branch of this horror movie
gives interesting change:

2779us
2667us
3153762us

After:

2319us
2523us
288581us

Aye, 10x :o)

-- 
Dmitry Olshansky