[Bese-devel] allegro mop compatibility

Gary King gwking at metabang.com
Fri Mar 31 17:37:07 UTC 2006


Hi Nathan,

I'd suggest building on Pascal Costanza's excellent Closer to MOP  
project. It might also make sense to use my moptilities (which does  
rely on Pascal's package). His package is at http://common-lisp.net/ 
project/closer/closer-mop.html; mine is at http://common-lisp.net/ 
project/cl-containers/moptilities/ .

regards,


On Mar 31, 2006, at 11:22 AM, Nathan Bird wrote:

> I finally traced down an error I was having, same one documented  
> at: http://paste.lisp.org/display/15624
>
>
>
> It’s happening because it is trying to query the class-slots of a  
> class that isn’t finalized yet. The classes are finalizable, but  
> apparently allegro doesn’t do this until you actually make- 
> instance. This is apparently in line with http://www.lisp.org/mop/ 
> concepts.html.
>
>
>
> So… ?
>
>
>
> (defun mopp::safe-class-slots (class)
>
>                (unless (mopp:class-finalized-p class)
>
>                  (mopp:finalize-inheritance class))
>
>                (mopp:class-slots class))
>
>
>
> Alternative we could make mopp:class-slots do the safely wrapping  
> the implementations version.
>
>
>
> On a side note, I found the comment in arnesi:
>
> ;;;; The code pre-dates Pascal Costanza's closer-mop package. If
>
> ;;;; you're looking for a compatability layer you should probably look
>
> ;;;; there instead.
>
>
>
> I haven’t looked at the packages as a whole to much so I don’t know  
> if it would be an easy transition, but aside from the work is there  
> a reason to not start using closer? I use that package for one or  
> two other things. So it won’t really be a NEW requirement for me,  
> but it might be for others.
>
>
>
> Nathan Bird
>
>
>
> _______________________________________________
> bese-devel mailing list
> bese-devel at common-lisp.net
> http://common-lisp.net/cgi-bin/mailman/listinfo/bese-devel


-- 
Gary Warren King
metabang.com
http://www.metabang.com/





More information about the bese-devel mailing list