Hi Anne,
Thank you for all this work.
I just have a small question : what is the impact on the projects of Famix users ?
If I understand, there is no.
In our case, we defined the generator and our MM based on the FamixNG DSL. So, the fact
that you change the traits should not have any impact on our structure.
Am I right ?
Cheers,
--
~~Jannik Laval~~
Enseignant-chercheur
Responsable Pédagogique Licence Coordonnateur de Projet
IUT Lumière, Université Lumière Lyon 2
laboratoire DISP
+33 4 78 77 43 06
http://www.jannik-laval.eu
http://www.phratch.com
http://www.approchealpes.info
Le 1 juil. 2019 à 17:28, Anne Etien
<anne.etien@univ-lille.fr<mailto:anne.etien@univ-lille.fr>> a écrit :
Hi everyone,
As already explained, we are currently cleaning the traits in FamixNG. Pavel did a great
job by providing a back office to create new meta model with a tiny DSL and by building
the old meta model using the new infrastructure. Now, we want to:
- clean elementary traits (for example when there is a FMMany slot, sometime there is an
addXXX method and sometime not)
- compose traits (FamixTClass is not a composition of traits, it has to be done when we
build the meta model), to provide semantically rich traits.
- be sure that tests are passing. They were tests in FamixNG, only to ensure compatibility
with the old meta model.
Consequently, I developed a generator to create tests associated to elementary traits.
Normally you don’t need to use this generator if you use only build traits by composing
existing ones.
Anyway, if needed, you can get the generator from FamixNG-TestGenerators package in
TestGenerator3 branch:
fttg := FamixTestAndTraitGenerator new.
fttg runFor: FamixTClass
I developed a second generator to generate tests from the trait composition. We will use
it after cleaning FamixNG. Obviously, later the generator has no vocation to remain in the
image, tests will be generated once and you can add tests by hand.
fcg := FamixComposingTestTraitsGenerator new.
fcg addingTraitsToTestTraitFor: FamixTClass
For using this later, we need that a correction in Pharo concerning composition of traits
with slots is back tracked in Pharo 7. Or you can use the attached file.
We already get 135 tests for FamixNG and we will continue.
If you have any question don’t hesitate.
Anne
_______________________________________________
Moose-dev mailing list
Moose-dev@list.inf.unibe.ch<mailto:Moose-dev@list.inf.unibe.ch>
https://www.list.inf.unibe.ch/listinfo/moose-dev
<TaSequence-validateSlots.st<http://TaSequence-validateSlots.st>>