@ Arrrrrrrrr

> why? Is this to avoid the criticism of undefined behavior?

Not really, but I've seen too many people compiling their production servers 
with -d:release.

> Does something like ... still work?

Yes.

> What's the reason behind this decision? This looks redundant to me, sure 
> there must be an explanation on why is needed. Can't the compiler figure out 
> when a variable/literal is of type int/float?

It was error prone, code like `x &= y.len` compiled and produced the wrong 
result. Type safety works better when not every type works with every other 
type. 

Reply via email to