Hi Mark, > Read this: > > http://clhs.lisp.se/Body/07_ec.htm
Perfect, thanks for the link and the confirmation. Note that in the current implementation, guile's problem is not the [proper] initialization of the inherited slot, as I did show and as you can check, slot-ref works as expected on Diego's example. One can also use describe for that matter (oop goops describe) .. The problem is that getters, setters and accessors although [merely] inherited are not properly recomputed. Your temporary solution is nice, but only works if there is no need for a change of the inherited slot allocation option, and would not solve the 'quantum mechanic' slot states :) that my 'consider the following situation' example shows. Therefore, in the mean time, I personally would recommend goops users to redefine getters, setters and accessors they are using/plan to use on their subclasse's instances, since it is currently the only way to 'force' guile to do his job properly. > Here's a transcript of an SBCL session I tried on Debian: Yep, I've downloaded it too now, it's been a while I did not play with lisp :) I much prefer scheme and guile now. But then I am not surprised they got it right, for lispers clos is an old beast and lisp implementations have, by far, much much more code and many many more testers as well, indeed clos is so much more used by lispers then goops by schemers. Cheers, David
