CTFE Status

[Stefan Koch via Digitalmars-d]

On Thursday, 1 December 2016 at 17:52:02 UTC, Stefan Koch wrote:
> I just fixed  a bug cause the non-debug build to fail on a 
> expression that it should have bailed out on.
>
> Currently I have troubles with this  code :
>
> int fn(int y)
> {
>   int x = void;
>   ++x;
>   return x+y;
> }
>
> Since x is uninitialized the ctfe interpreter should fail on it.
> I will need to reserve a certain value as the uninitialized 
> flag.
> since stackAddress zero can never be used it'd be perfect for 
> the job.
> This bug should be fixed by tomorrow.

I opted for bailing out as soon as void initialization is in the 
function.
This will need to handled properly at some point but now is not 
the time.