Hello,
I just discovered this strange behavior:

struct A {
  void[10] x; // OK, compiles fine
}
class B {
  void[10] x; // Error: void does not have a default initializer
              // ! note, the message come without any line indication !
}

Does this mean that void arrays are a thing or it is a bug ?

badlink:

> Does this mean that void arrays are a thing or it is a bug ?

You can compile that with:

class Foo {
    void[10] x = void;
}


But this code shows a dmd bug (lack of line number):

class Foo {
    void[10] x;
}
void main() {}


Error: void does not have a default initializer


Take a look in bugzilla, if it's not already present file it.

Bye,
bearophile

> Take a look in bugzilla, if it's not already present file it.

https://issues.dlang.org/show_bug.cgi?id=13505

Bye,
bearophile

On Sat, Sep 20, 2014 at 10:09:25AM +0000, badlink via Digitalmars-d-learn wrote:
> Hello,
> I just discovered this strange behavior:
> 
> struct A {
>   void[10] x; // OK, compiles fine
> }
> class B {
>   void[10] x; // Error: void does not have a default initializer
>               // ! note, the message come without any line indication !
> }
> 
> Does this mean that void arrays are a thing or it is a bug ?

void[] is a wildcard array type that represents a block of memory that contains an array of some sort. It behaves like ubyte[], except that it may be GC-scanned for pointers (ubyte[] AFAIK is not scanned because you're not expected to store pointers in it, but void[] represents *any* array and thus may contain pointers).


T

-- 
Not all rumours are as misleading as this one.