[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Incorrect CPL calculation in 4-29 Release?

I suspect that the class precedence list calculation for class objects
in the 4-29 release of PCL/PCLOS may be incorrect. Consider the following:

(defclass foo () () )

(class-precedence-list (class-named 'foo))


(#<Class FOO > #<Class OBJECT > #<Class T> )

implying that the class of the FOO class object is FOO...which is
incorrect. If class FOO has metaclass STANDARD-CLASS, then the
FOO class object should have STANDARD-CLASS as the first element
of the class precedence list (in fact, since STANDARD-CLASS is
not yet in the system, this will probably be CLASS). 

This should only be a problem for people using the metaobject
protocol, since only in that case are methods dispatching on
class objects relevent.

			Jim Kempf	kempf@hplabs.hp.com