On Feb 17, 2010, at 16:07 , Simon Denier wrote:
On 17 févr. 2010, at 15:09, Laval Jannik wrote:
For sure, I can :), But maybe you want a dsm to see cycles ?
To do simple, isAbstract is defined in Famix-Core, but all other isSomething are defined in FamixExtensions. It seems to be correct to move it with other test methods.
yeah, but why does isAbstract create a cycle?
because it uses modifierSet:for: wich is defined in Famix-Implementation. Famix-Implementation depends on FamixCore.
So there is a cycle.
Cheers, Jannik
For the ret, maybe we could do this:
extend method>> Smalltalk::FAMIXPackage.definedMethods() in Famix-Extensions.
it is an alias of self methods, it should be in Famix-Implementations
extend method>> Smalltalk::FAMIXPackage.localClasses() in Famix-Extensions.
it is an alias of self classes, it should be in Famix-Implementations
extend method>> Smalltalk::FAMIXPackage.extendedClasses() in Famix-Extensions. extend method>> Smalltalk::FAMIXPackage.extendedClassesGroup() in Famix-Extensions. extend method>> Smalltalk::FAMIXPackage.extensionClasses() in Famix-Extensions. extend method>> Smalltalk::FAMIXPackage.extensionClassesGroup() in Famix-Extensions. extend method>> Smalltalk::FAMIXPackage.extensionMethods() in Famix-Extensions. extend method>> Smalltalk::FAMIXPackage.localMethods() in Famix-Extensions. extend method>> Smalltalk::FAMIXPackage.localClassesGroup() in Famix-Extensions.
These methods are used by Cook, which is in Famix-Extension now. But I will try something else.
Jannik, if we rewrite the cook queries with the Chef API, those dependencies will disappear right? And we can keep smalltalkish stuff in Famix-Smalltalk. I still find it more disturbing that Cook depends on some Smalltalk stuff, and moving this stuff in Extensions feels a bit like sweeping it under the carpet.
-- Simon
Moose-dev mailing list Moose-dev@iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev