Sometimes I'm unsure what exactly is the difference between

a) cast(t)x
b) (t)x
c) t(x)

I know, (c) is a constructor call, but for basic types that's the same as (a) isn't it?
If t provides a constructor for typeof(x) and x provides opCast to type t, which one is called?
Does it depend on the form (a), (b), or (c)?
Does all three forms work if only the constructor or only the opCast is provided?
And is (b) always equivalent to (a)? (there is an odd example in the grammar where this is not the case)

On Wednesday, 3 May 2023 at 09:03:38 UTC, Dom DiSc wrote:

No, a cast allows for overflow cast(ubyte) 256, while the constructor needs an integer that fits. ubyte(256) is an error.

When casting, opCast has precedence over a constructor.

A constructor call will not be lowered to opCast, but a cast can be lowered to a constructor call.

C style cast syntax is not allowed in D.