Re: [Pharo-users] trait question

2013-07-01 Thread Igor Stasenko
On 1 July 2013 14:01, Camille Teruel wrote: > > On 1 juil. 2013, at 12:19, Igor Stasenko wrote: > >> On 30 June 2013 15:48, Damien Cassou wrote: >>> On Sun, Jun 30, 2013 at 2:54 PM, Camillo Bruni >>> wrote: But you know that you can do this right now? >>> >>> >>> I think these examples onl

Re: [Pharo-users] trait question

2013-07-01 Thread Camille Teruel
On 1 juil. 2013, at 12:19, Igor Stasenko wrote: > On 30 June 2013 15:48, Damien Cassou wrote: >> On Sun, Jun 30, 2013 at 2:54 PM, Camillo Bruni >> wrote: >>> But you know that you can do this right now? >> >> >> I think these examples only show that some verifications are missing. >> I don't

Re: [Pharo-users] trait question

2013-07-01 Thread Igor Stasenko
On 30 June 2013 15:48, Damien Cassou wrote: > On Sun, Jun 30, 2013 at 2:54 PM, Camillo Bruni wrote: >> But you know that you can do this right now? > > > I think these examples only show that some verifications are missing. > I don't know any source code using that. To me it's more a bug than a >

Re: [Pharo-users] trait question

2013-06-30 Thread Tudor Girba
Hi, Theoretically, I could do the same for inheritance as well. But, that would be too complex for most cases, so it is the implicit behavior. To me, the decision for trait handling is similar. Cheers, Doru On Sun, Jun 30, 2013 at 2:54 PM, Camillo Bruni wrote: > On 2013-06-30, at 11:16, Sebas

Re: [Pharo-users] trait question

2013-06-30 Thread Stéphane Ducasse
On Jun 30, 2013, at 2:54 PM, Camillo Bruni wrote: > > uses: TMyOtherTrait

Re: [Pharo-users] trait question

2013-06-30 Thread Stéphane Ducasse
> > > But you know that you can do this right now? > > Object subclass: #Foo > uses: TMyTrait classTrait > > and the same on the class-side :/ > > Foo class > uses: TMyOtherTrait > instanceVariableNames: 'a b c' > > so from that

Re: [Pharo-users] trait question

2013-06-30 Thread Damien Cassou
On Sun, Jun 30, 2013 at 2:54 PM, Camillo Bruni wrote: > But you know that you can do this right now? I think these examples only show that some verifications are missing. I don't know any source code using that. To me it's more a bug than a feature. I would keep things simple: applying a trait t

Re: [Pharo-users] trait question

2013-06-30 Thread Camillo Bruni
On 2013-06-30, at 11:16, Sebastian Tleye wrote: > I also like this uniform because it's easier to program, > Up to now i haven't had the need to separate them. > > 2013/6/30 Tudor Girba >> +1 >> >> Keeping things uniform is a plus in this case, even though we might lose >> some flexibility. >>

Re: [Pharo-users] trait question

2013-06-30 Thread Sebastian Tleye
I also like this uniform because it's easier to program, Up to now i haven't had the need to separate them. 2013/6/30 Tudor Girba > +1 > > Keeping things uniform is a plus in this case, even though we might lose > some flexibility. > > Doru > > > On Sun, Jun 30, 2013 at 9:01 AM, Stéphane Ducass

Re: [Pharo-users] trait question

2013-06-30 Thread Tudor Girba
+1 Keeping things uniform is a plus in this case, even though we might lose some flexibility. Doru On Sun, Jun 30, 2013 at 9:01 AM, Stéphane Ducasse wrote: > Ok I slept and now my brain is working. > I think that having classTrait and trait together is important because > else it would be th

Re: [Pharo-users] trait question

2013-06-30 Thread Stéphane Ducasse
Ok I slept and now my brain is working. I think that having classTrait and trait together is important because else it would be the mess when you have a class method that uses an instance methods (both define in a trait). So conceptually this is one trait like in smalltalk you have one class and

Re: [Pharo-users] trait question

2013-06-29 Thread Camillo Bruni
so let's discuss that on monday with the trait specialists Damien et Sebastian :) On 2013-06-29, at 23:34, Igor Stasenko wrote: > yes, i don't know if it good or bad, but i'd rather prefer that traits > applied to Behavior, > but not to Class (which has two entangled behaviors). But maybe for >

Re: [Pharo-users] trait question

2013-06-29 Thread Igor Stasenko
yes, i don't know if it good or bad, but i'd rather prefer that traits applied to Behavior, but not to Class (which has two entangled behaviors). But maybe for human comprehension it is better.. but not for implementing, because it certainly makes model more complex. On 29 June 2013 22:52, Camillo

Re: [Pharo-users] trait question

2013-06-29 Thread Camillo Bruni
On 2013-06-29, at 22:32, Igor Stasenko wrote: > On 29 June 2013 21:21, Stéphane Ducasse wrote: >> >> On Jun 29, 2013, at 8:19 PM, Igor Stasenko wrote: >> >>> On 29 June 2013 20:12, Camillo Bruni wrote: Working on the new class builder I have a hard time figuring out the full trai

Re: [Pharo-users] trait question

2013-06-29 Thread Igor Stasenko
On 29 June 2013 21:21, Stéphane Ducasse wrote: > > On Jun 29, 2013, at 8:19 PM, Igor Stasenko wrote: > >> On 29 June 2013 20:12, Camillo Bruni wrote: >>> Working on the new class builder I have a hard time figuring out the full >>> trait >>> semantics. >>> >>> If I use a trait on instance side,

Re: [Pharo-users] trait question

2013-06-29 Thread Stéphane Ducasse
On Jun 29, 2013, at 8:19 PM, Igor Stasenko wrote: > On 29 June 2013 20:12, Camillo Bruni wrote: >> Working on the new class builder I have a hard time figuring out the full >> trait >> semantics. >> >> If I use a trait on instance side, does that imply that its classTrait on the >> metaclass?

Re: [Pharo-users] trait question

2013-06-29 Thread Igor Stasenko
On 29 June 2013 20:12, Camillo Bruni wrote: > Working on the new class builder I have a hard time figuring out the full > trait > semantics. > > If I use a trait on instance side, does that imply that its classTrait on the > metaclass? yes. Trait resembles same dual property as class (it has ins

[Pharo-users] trait question

2013-06-29 Thread Camillo Bruni
Working on the new class builder I have a hard time figuring out the full trait semantics. If I use a trait on instance side, does that imply that its classTrait on the metaclass? smime.p7s Description: S/MIME cryptographic signature