VRP and division/remainder by 0

[deadalnix via Digitalmars-d]

On Monday, 4 July 2016 at 00:58:49 UTC, Jonathan M Davis wrote:
> I confess that my first thought was that it should just result 
> in an error, since it's obviously wrong, but then you have an 
> error that depends on how good VRP is, which may not be a great 
> idea. Still, I don't see any reason why you'd want it to 
> actually compile when it's obviously going to crash the program 
> if it's ever hit.
>
> - Jonathan M Davis

Yup, also, consider :

a == 0 ? b : 10 / a