fame package annotation related question

List overview All Threads
Download

newer

older

defining a clear set of change for...

Browser is broken in the image...

Stéphane Ducasse

23 Sep 2011 23 Sep '11

2:44 p.m.

I imagine that I cannot ask an FAME attribute the methods that defined it. Am I correct? So this means that if I want to know that an attribute got tagged with a package pragma: I have to query that at the smalltalk level. But this means that I also have to check that this is the correct attributes I got <MSEProperty: #annotationInstances type: #FAMIXAnnotationInstance opposite: #annotatedEntity> <multivalued> <derived> <package: #Java> Stef

Show replies by date

Tudor Girba

23 Sep 23 Sep

3:16 p.m.

Hi Stef, On 23 Sep 2011, at 14:44, Stéphane Ducasse wrote:

...

I imagine that I cannot ask an FAME attribute the methods that defined it. Am I correct?

No. You have implementingSelector. FAMIXPackage asMooseDescription attributes first implementingSelector --> #numberOfMethods http://www.themoosebook.org/book/internals/fame/smalltalk-integration

...

So this means that if I want to know that an attribute got tagged with a package pragma: I have to query that at the smalltalk level.

For the package pragma, we have to enhance the FMPragmaProcessor to take it into account. Cheers, Doru

...

But this means that I also have to check that this is the correct attributes I got <MSEProperty: #annotationInstances type: #FAMIXAnnotationInstance opposite: #annotatedEntity> <multivalued> <derived> <package: #Java> Stef _______________________________________________ Moose-dev mailing list Moose-dev(a)iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev

-- www.tudorgirba.com "The coherence of a trip is given by the clearness of the goal."

Stéphane Ducasse

9:17 p.m.

On Sep 23, 2011, at 3:16 PM, Tudor Girba wrote:

...

Hi Stef, On 23 Sep 2011, at 14:44, Stéphane Ducasse wrote:

I imagine that I cannot ask an FAME attribute the methods that defined it. Am I correct?

No. You have implementingSelector.

Excellent I will save my day.

...

FAMIXPackage asMooseDescription attributes first implementingSelector --> #numberOfMethods http://www.themoosebook.org/book/internals/fame/smalltalk-integration

So this means that if I want to know that an attribute got tagged with a package pragma: I have to query that at the smalltalk level.

For the package pragma, we have to enhance the FMPragmaProcessor to take it into account.

Ok for now I will play and enhance the metareport builder. Probably we will migrate everything to a subclass of FMPragmaProcessor or user. I want to make progress on FAMIX30.

...

Cheers, Doru

-- www.tudorgirba.com "The coherence of a trip is given by the clearness of the goal." _______________________________________________ Moose-dev mailing list Moose-dev(a)iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev

Stéphane Ducasse

9:46 p.m.

Now do you have an idea how I can get the Fame object when I get a method? I will look in FAMEPragmaProcessor Stef On Sep 23, 2011, at 3:16 PM, Tudor Girba wrote:

...

Hi Stef, On 23 Sep 2011, at 14:44, Stéphane Ducasse wrote:

I imagine that I cannot ask an FAME attribute the methods that defined it. Am I correct?

No. You have implementingSelector. FAMIXPackage asMooseDescription attributes first implementingSelector --> #numberOfMethods http://www.themoosebook.org/book/internals/fame/smalltalk-integration

So this means that if I want to know that an attribute got tagged with a package pragma: I have to query that at the smalltalk level.

For the package pragma, we have to enhance the FMPragmaProcessor to take it into account. Cheers, Doru

Stéphane Ducasse

9:53 p.m.

doru do you have an idea if this is correct that FMPragmaProcessor new processClass: FAMIXContainerEntity returns a FM3MetaDescription whose attributes are empty? FAMIXContainerEntity asMooseDescription attributes size -> 4 Do you think that we should fix processClass: Stef On Sep 23, 2011, at 3:16 PM, Tudor Girba wrote:

...

Hi Stef, On 23 Sep 2011, at 14:44, Stéphane Ducasse wrote:

I imagine that I cannot ask an FAME attribute the methods that defined it. Am I correct?

No. You have implementingSelector. FAMIXPackage asMooseDescription attributes first implementingSelector --> #numberOfMethods http://www.themoosebook.org/book/internals/fame/smalltalk-integration

So this means that if I want to know that an attribute got tagged with a package pragma: I have to query that at the smalltalk level.

For the package pragma, we have to enhance the FMPragmaProcessor to take it into account. Cheers, Doru

Stéphane Ducasse

10:03 p.m.

I thought that I made a mistake but using MoosePragmaProcessor I get the same behavior. So is MoosePragmaProcessor when we reset the properties? I will check MooseModel resetMeta Stef On Sep 23, 2011, at 9:53 PM, Stéphane Ducasse wrote:

...

Hi Stef, On 23 Sep 2011, at 14:44, Stéphane Ducasse wrote:

I imagine that I cannot ask an FAME attribute the methods that defined it. Am I correct?

No. You have implementingSelector. FAMIXPackage asMooseDescription attributes first implementingSelector --> #numberOfMethods http://www.themoosebook.org/book/internals/fame/smalltalk-integration

So this means that if I want to know that an attribute got tagged with a package pragma: I have to query that at the smalltalk level.

For the package pragma, we have to enhance the FMPragmaProcessor to take it into account. Cheers, Doru

_______________________________________________ Moose-dev mailing list Moose-dev(a)iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev

Tudor Girba

24 Sep 24 Sep

10:10 a.m.

Hi Stef, ensureClass: is a private method (see the protocol). An example of how to use the class, can be found in the tests, or in the buildFM3. processor := FMPragmaProcessor new. processor queue: FMPragmaProcessorTestDummy. processor run. The MoosePragmaProcessor was just my attempt to not have the processor crash when there is no MSEClass: pragma in the class. Cheers, Doru On 23 Sep 2011, at 22:03, Stéphane Ducasse wrote:

...

Hi Stef, On 23 Sep 2011, at 14:44, Stéphane Ducasse wrote:

I imagine that I cannot ask an FAME attribute the methods that defined it. Am I correct?

No. You have implementingSelector. FAMIXPackage asMooseDescription attributes first implementingSelector --> #numberOfMethods http://www.themoosebook.org/book/internals/fame/smalltalk-integration

So this means that if I want to know that an attribute got tagged with a package pragma: I have to query that at the smalltalk level.

For the package pragma, we have to enhance the FMPragmaProcessor to take it into account. Cheers, Doru

_______________________________________________ Moose-dev mailing list Moose-dev(a)iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev

-- www.tudorgirba.com "In a world where everything is moving ever faster, one might have better chances to win by moving slower."

Stéphane Ducasse

10:23 a.m.

...

Ok I'm diving into that. What I will do is to hack the metamodelreport so that I get progress from this side. And I will look at all the meta stuff in parallel Now I'm still confused why

...

>> FMPragmaProcessor new processClass: FAMIXContainerEntity >

...

>> returns a FM3MetaDescription whose attributes are empty?

And this resetMeta is working.

...

Tudor Girba

10:39 a.m.

Hi Stef, On 24 Sep 2011, at 10:23, Stéphane Ducasse wrote:

...

Ok I'm diving into that. What I will do is to hack the metamodelreport so that I get progress from this side. And I will look at all the meta stuff in parallel Now I'm still confused why

>> FMPragmaProcessor new processClass: FAMIXContainerEntity >> >> returns a FM3MetaDescription whose attributes are empty?

And this resetMeta is working.

I told you that processClass: is an internal method (i.e., it is used in combination with others :)). So, if you look at the run method, you will see that it also calls resolveObjectReferences. Try this: processor := FMPragmaProcessor new. description := processor processClass: FMPragmaProcessorTestDummy. processor resolveObjectReferences. description inspect Cheers, Doru

...

_______________________________________________ Moose-dev mailing list Moose-dev(a)iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev

-- www.tudorgirba.com "It's not what we do that matters most, it's how we do it."

Stéphane Ducasse

10:44 p.m.

Ok I will look at it and write a comment as usual. Stef

...

Hi Stef, On 24 Sep 2011, at 10:23, Stéphane Ducasse wrote:

Ok I'm diving into that. What I will do is to hack the metamodelreport so that I get progress from this side. And I will look at all the meta stuff in parallel Now I'm still confused why

>> FMPragmaProcessor new processClass: FAMIXContainerEntity >> >> returns a FM3MetaDescription whose attributes are empty?

And this resetMeta is working.

_______________________________________________ Moose-dev mailing list Moose-dev(a)iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev

-- www.tudorgirba.com "It's not what we do that matters most, it's how we do it." _______________________________________________ Moose-dev mailing list Moose-dev(a)iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev

4838

days inactive

4839

days old

moose-dev@list.inf.unibe.ch

Manage subscription

9 comments

2 participants

tags (0)

participants (2)

Stéphane Ducasse
Tudor Girba