Done. https://pharo.fogbugz.com/f/cases/12731/Traits-modifications-cause-a-DNU
On Tue, Jan 28, 2014 at 2:13 PM, Tudor Girba <[email protected]> wrote: > Thank you! I can review it tomorrow morning. > > Doru > > > On Tue, Jan 28, 2014 at 2:10 PM, Martin Dias <[email protected]> wrote: > >> hhahah, ok, I'm on it >> >> >> On Tue, Jan 28, 2014 at 2:06 PM, Tudor Girba <[email protected]>wrote: >> >>> Yes, please! >>> >>> Doru >>> >>> >>> On Tue, Jan 28, 2014 at 2:04 PM, Esteban Lorenzano >>> <[email protected]>wrote: >>> >>>> yes, please :) >>>> >>>> On 28 Jan 2014, at 13:47, Martin Dias <[email protected]> wrote: >>>> >>>> >>>>> So, what to do? >>>>> - Don't send ClassModifiedClassDefinition in >>>>> SystemAnnouncer>>traitDefinitionChangedFrom: oldTrait to: newTrait ? >>>>> - Implement Trait>>layout ? >>>>> - Test for oldClassDefinition isTrait in >>>>> ClassModifiedClassDefinition>>isPropagating ? >>>>> >>>>> >>>>> >>>> I discussed with Camille and we think it's better this other >>>> alternative: to fix class builder to only announce >>>> ClassModifiedClassDefinition for the class that really changed its >>>> definition. For the subclasses, it won't be announced. This way, it's not >>>> necessary to check if it is a propagation. We can remove the two >>>> implementors and the unique sender. >>>> >>>> I verified that this was the behavior of "old class builder" (in Pharo >>>> 2). Do you agree? >>>> >>>> I can submit a slice this afternoon. >>>> >>>> MartÃn >>>> >>>> >>>> >>> >>> >>> -- >>> www.tudorgirba.com >>> >>> "Every thing has its own flow" >>> >> >> > > > -- > www.tudorgirba.com > > "Every thing has its own flow" >
