Hi every one, in the past i tried to implement methods in FAMIX to handel the different types of dependencies between software entities. Now i have some thing like this:
Dependencies: Two subtypes of dependencies can occur: 1)References: based either on methods invocations or on classes accesses: a)Invocations: Where the method's signature is explicitly used. Two categories of invocations can occur: a1)Sure invocations: the receiving variable of the invocation is either -a class: {C1.m1 --> C3.new} is a sure outgoing invocation. -Or an implicit variable (self/super): {C2.m1 --> C1.m1} {C2.m2 --> C2.m1} represent also sure outgoing invocations. In consequence, I say that: C1 surely refers to C3, and C2 surely refers to C1, ... etc. a2)Potential invocations: the type of the receiving variable is not known: {C1.m1, C2.m1, and C4.m1} are all potentially invoked by C1.m1. In consequence, I say that {C2 and C4} are potentially referenced by C1.
b)Access or Static-References: where the name of the class is explicitly used: {C1.m1 --> C4} is a static reference.
2)Inheritance.
Now I work to clean up my code and i need your ideas about the vocabularies that i have used: invokedMethods/requestorMethods: -sureInvokedMethods/sureRequestorMethods -potentialInvokedMethods/potentialRequestorMethods referencedClasses/referencingClasses, ...etc. -sureReferencedClasses/sureReferencingClasses, ...etc. -potentialReferencedClasses/potentialReferencingClasses, ...etc. -staticReferencedClasses/staticReferencingClasses, ...etc. inheritedClasses/inheritedByClasses, ...etc. neededClasses/dependentClasses, ...etc.
do you have another proposition than using 'sure', 'potential', 'static' words? do you have any comment about the different categories of dependencies that i have used?
bests, hani
On Nov 27, 2007, at 15:29 PM, Hani ABDEEN wrote:
Hi all, in the attachement you find what it seems to be a solution to what you ask Stephane:
no?
bests hani
On Nov 26, 2007, at 14:58 PM, Stéphane Ducasse wrote:
On 26 nov. 07, at 09:09, Johan Brichau wrote:
Hi all,
In the importer I wrote, I have the following mappings:
- Java packages -> Famix namespaces (Simply because java packages
have control over name scoping).
- Java jar files -> Famix packages (it's the closest thing that
corresponds to any 'packaging' thing in Java).
I think it's strange to map Java packages onto Famix packages.. Just my 2 cents ;-)
Oops. I think that I would like to be much clever and know the answer. It seems that I will start to work on package and see how it goes. I have the impression that from a scenario point of view, I want to see how a class refer/is referenced by other classes. and this also at the level of the packages. so the scoping of the container is important now this would push me to put java package into namespace but to analyse I would like the inverse. So please help me.
Stef
Johan
ps: I will be looking at using the existing Eclipse importer directly from JavaConnect/Penumbra too.
On 25 Nov 2007, at 16:25, Toon Verwaest wrote:
I thought about it and I would like to have
java package -> extracted as package smalltalk package -> as package smalltalk namespace -> as namespace
Is it what you had in mind?
I don't necessarily think that package equals namespace in Java either. There are the same amount of namespaces as packages, but unlike packages, the namespaces are not nested, right?
Well on the other hand, namespaces are obviously deducable from java packages, so I guess it makes sense to just have packages and write code to get the namespace info (which is more or less important because package belongsTo != namespace belongsTo for java).
What is your opinion about the matter? @Sandro: get involved , this concerns you too, obviously :)
Johan Brichau johan.brichau@uclouvain.be
Moose-dev mailing list Moose-dev@iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev
Moose-dev mailing list Moose-dev@iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev
Moose-dev mailing list Moose-dev@iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev