Or moving those instance variables to their subclasses (Class and Metaclass)
2013/7/15 Camille Teruel <[email protected]> > > On 13 juil. 2013, at 14:44, Camillo Bruni wrote: > > > If you put all methods from ClassDescriptions in traits and move them on > > Class and Metaclass there is no need for ClassDescription. > > Except for "instanceVariables" and "organization" ivs of ClassDescription. > Were you thinking about stateful traits? > > > If previously you want something that inherited from ClassDescription you > > can simply include the traits now. > > > > In the case of a Behavior subclass nothing changes. > > > > > On 2013-07-13, at 14:38, Stéphane Ducasse <[email protected]> > wrote: > >> Why can't we keep ClassDescription. > >> Then in addition Behavior is a nice class to have to build strange > stuff. > >> > >> Stef > >> > >> On Jul 12, 2013, at 7:22 AM, Damien Cassou <[email protected]> > wrote: > >> > >>> On Thu, Jul 11, 2013 at 6:36 PM, Esteban Lorenzano < > [email protected]> wrote: > >>>> 11151 More changes needed to trait/classes polymorphism > >>>> https://pharo.fogbugz.com/f/cases/11151/ > >>> > >>> > >>> That's huge. Soon, we will be able to remove thr ClassDescription and > >>> Behavior classes. It makes sense because they are useless in the > >>> presence of traits. But do people want it? > >>> > >>> -- > >>> Damien Cassou > >>> http://damiencassou.seasidehosting.st > >>> > >>> "Success is the ability to go from one failure to another without > >>> losing enthusiasm." > >>> Winston Churchill > >>> > >> > >> > > > > > > >
