Re: clos object wierdness ..

• To: berni@iml.iml.fhg.de
• Subject: Re: clos object wierdness ..
• From: rudolf mittelmann <rm@cast.uni-linz.ac.at>
• Date: Wed, 16 Feb 1994 10:54:47 +0100
• Cc: clim@BBN.COM
• Reply-To: rudolf mittelmann <rm@cast.uni-linz.ac.at>

In message <9402150852.AA24012@iml.fhg.de>  writes:
>     Essentially, I'm doing the following:
>     
>     (defclass a () ())
>     (defclass b () ())
>     (defclass c () ())
>     
>     (defclass d (a b c) ())
>     (define-presentation-type d (foo))
> 
> As far as I understood this is not a correct p-type definition
> in CLIM. presentation type d MUST :inherit-from (a b c), as the
> presentation-type hierarchy must always "parallel" the clos class-
> hierarchy.
>     

If I understand it right, you are wrong here: If you do not
define p-types *explicitly*, the p-types automatically parallel
the CLOS hierarchie.
You should define p-types explicitly in order to define
a *different* hierarchy, so this must be peerfectly legal.

In the rare case, that you want to define a p-type explicitly,
but of the default (CLOS) hierarchy, you have to supply
the full supertype list for the :inherit-from keyword.

I find it much easier to *not* define p-types explicitly.

• Prev: Re: problems with 1.1 output records under incremental redisplay
• Next: CLIM Documentation
• Index: CLIM Mailing list (1-Jun-94 to 30-Jun-95)
• Thread: CLIM Mailing list Index (1-Jun-94 to 30-Jun-95)