Le 24/09/2013 16:50, Igor Stasenko a écrit :

On 24 September 2013 15:31, Goubier Thierry <thierry.goubier@cea.fr <mailto:thierry.goubier@cea.fr>> wrote:

I'm managing to get MNUs in Pharo 3.0 in a quite fun way.

I've compiled methods with nil as protocol... The result is kind of interesting.

i think it is because nil is no longer valid protocol. sure thing it should not lead to dnu but to some more clear error message.

Well, nil wasn't a protocol either in 2.0 but it was ignored :)

anyways, thanks for finding the bug

I triggered it with aClass compile: 'test ^true' classified: nil notifying: nil :) There is also another behavior difference between 2.0 and 3.0: the protocol #'as yet unclassified' disappear when empty in 2.0 and it stays, even empty, in 2.0. Do you intend it that way? Regards, Thierry -- Thierry Goubier CEA list Laboratoire des Fondations des Systèmes Temps Réel Embarqués 91191 Gif sur Yvette Cedex France Phone/Fax: +33 (0) 1 69 08 32 92 / 83 95

On Sep 24, 2013, at 6:20 PM, Igor Stasenko <siguctua@gmail.com> wrote:

On 24 September 2013 17:06, Goubier Thierry <thierry.goubier@cea.fr> wrote:

Le 24/09/2013 16:50, Igor Stasenko a écrit :

On 24 September 2013 15:31, Goubier Thierry <thierry.goubier@cea.fr <mailto:thierry.goubier@cea.fr>> wrote:

I'm managing to get MNUs in Pharo 3.0 in a quite fun way.

I've compiled methods with nil as protocol... The result is kind of interesting.

i think it is because nil is no longer valid protocol. sure thing it should not lead to dnu but to some more clear error message.

Well, nil wasn't a protocol either in 2.0 but it was ignored :)

anyways, thanks for finding the bug

I triggered it with

aClass compile: 'test ^true' classified: nil notifying: nil

:)

There is also another behavior difference between 2.0 and 3.0: the protocol #'as yet unclassified' disappear when empty in 2.0 and it stays, even empty, in 2.0.

Do you intend it that way?

I don't know the details, there was a lot of work done by Esteban refactoring this.. so i letting him to answer your question.

no, is a bug (known), but still not time to fix it :)

Regards,

Thierry -- Thierry Goubier CEA list Laboratoire des Fondations des Systèmes Temps Réel Embarqués 91191 Gif sur Yvette Cedex France Phone/Fax: +33 (0) 1 69 08 32 92 / 83 95

-- Best regards, Igor Stasenko.

On Sep 24, 2013, at 8:38 PM, GOUBIER Thierry <thierry.goubier@cea.fr> wrote:

Ok.

At the same time, I can appreciate the way the change was done. I just merged my system browser from 2.0 to 3.0 with very few problems, just tracking a few deprecations. And given how close I am to the class organisation / RPackage and friends, I can appreciate the care in maintaining a very close API while refactoring.

Thanks because we hate this part of the system and this is not what we are dreaming of. if you would see esteban coming to steal fruits in my office these days :) Stef

Thierry De : Pharo-dev [pharo-dev-bounces@lists.pharo.org] de la part de Esteban Lorenzano [estebanlm@gmail.com] Date d'envoi : mardi 24 septembre 2013 18:34 À : Igor Stasenko Cc: Pharo Development List Objet : Re: [Pharo-dev] Pharo 3 is interesting

On Sep 24, 2013, at 6:20 PM, Igor Stasenko <siguctua@gmail.com> wrote:

...

On 24 September 2013 17:06, Goubier Thierry <thierry.goubier@cea.fr> wrote:

Le 24/09/2013 16:50, Igor Stasenko a écrit :

On 24 September 2013 15:31, Goubier Thierry <thierry.goubier@cea.fr <mailto:thierry.goubier@cea.fr>> wrote:

I'm managing to get MNUs in Pharo 3.0 in a quite fun way.

I've compiled methods with nil as protocol... The result is kind of interesting.

i think it is because nil is no longer valid protocol. sure thing it should not lead to dnu but to some more clear error message.

Well, nil wasn't a protocol either in 2.0 but it was ignored :)

anyways, thanks for finding the bug

I triggered it with

aClass compile: 'test ^true' classified: nil notifying: nil

:)

There is also another behavior difference between 2.0 and 3.0: the protocol #'as yet unclassified' disappear when empty in 2.0 and it stays, even empty, in 2.0.

Do you intend it that way?

I don't know the details, there was a lot of work done by Esteban refactoring this.. so i letting him to answer your question.

no, is a bug (known), but still not time to fix it :)

...

Regards,

Thierry -- Thierry Goubier CEA list Laboratoire des Fondations des Systèmes Temps Réel Embarqués 91191 Gif sur Yvette Cedex France Phone/Fax: +33 (0) 1 69 08 32 92 / 83 95

-- Best regards, Igor Stasenko.