A clue'd be nice.

"Matthew Wilson" <dmd@synesis.com.au> wrote in message news:cd4tu5$1hc5$1@digitaldaemon.com...
> A clue'd be nice.

Scratch that. I *must* have the name of the unimplemented method. I'm pretty sure I've precipitated a compiler error, since this code was working well in March. Now I'm compiling with 0.95, it's giving me that rather clueless response. I'm perfectly happy to have to hack a workaround, but I need somewhere to start.

Matthew Wilson wrote:

> "Matthew Wilson" <dmd@synesis.com.au> wrote in message
> news:cd4tu5$1hc5$1@digitaldaemon.com...
> 
>> A clue'd be nice.
> 
> Scratch that.
<snip>

No need to scratch anything.  For the benefit of all of us, the compiler ought to indicate which method is unimplemented, causing the class to become abstract.  Borland C++ manages it without any trouble.

But to be honest, I don't see what was wrong with making D like Java in this respect - a class is abstract iff it is declared as abstract.

Just looking at the docs, abstract doesn't seem to be explained at all!

Stewart.

-- 
My e-mail is valid but not my primary mailbox, aside from its being the unfortunate victim of intensive mail-bombing at the moment.  Please keep replies on the 'group where everyone may benefit.

> Just looking at the docs, abstract doesn't seem to be explained at all!

it isn't, as far as i can tell.  and i can't find any reference to abstract classes, let alone an explanation of what the "abstract" keyword does.  i'm happy that Matthew found a way to make classes abstract!  i was beginning to wonder if it were possible..