Is there a good reason why we can't have ctors/dtors on structs?

I think they'd be extremely useful, and am currently stymied on a technique I wanted to use because of this.

I believe the given reason is that it would make the compiler implementation more complicated, and would complicate exception handling.  I know one of the design goals of D is a simple implementation, but this seems like a cop-out to me.

Sean

"Matthew Wilson" <matthew@stlsoft.org> wrote in message news:bmafdi$2u7n$2@digitaldaemon.com...
> Is there a good reason why we can't have ctors/dtors on structs?
>
> I think they'd be extremely useful, and am currently stymied on a
technique
> I wanted to use because of this.

I can see the point of a simple compiler implementation up to a point, but when it comes to hobbling the language, it's gone beyond its purpose.

Is it better to have a powerful language, or a simple compiler? What's going to result in the greater success of D in the long run? Seems pretty obvious to me (and you :))

"Sean L. Palmer" <palmer.sean@verizon.net> wrote in message news:bmcfq6$2icd$1@digitaldaemon.com...
> I believe the given reason is that it would make the compiler
implementation
> more complicated, and would complicate exception handling.  I know one of the design goals of D is a simple implementation, but this seems like a cop-out to me.
>
> Sean
>
> "Matthew Wilson" <matthew@stlsoft.org> wrote in message news:bmafdi$2u7n$2@digitaldaemon.com...
> > Is there a good reason why we can't have ctors/dtors on structs?
> >
> > I think they'd be extremely useful, and am currently stymied on a
> technique
> > I wanted to use because of this.
>
>