Home » Archived » M2M (model-to-model transformation) » New ATL use case: A Metamodel Independent Approach to Difference Representation
| |
Re: New ATL use case: A Metamodel Independent Approach to Difference Representation [message #32285 is a reply to message #30316] |
Tue, 17 April 2007 08:18 |
Eclipse User |
|
|
|
Originally posted by: andreas.ortwein.gmx.de
Dear Davide,
I'm a newbie on the ATL/AMMA framework and working about two months on the
problem of describing the diff of two uml models with ATL. Last weekend I
found this message of the new use case here on the ATL newsgroup. I read the
regarding technical report to the use case and I'm convinced that this
approach fits to my problem. For now I have just one question concerning the
transformation model used to transform a base model to a final model: How is
the diff-model generated (in the use case, it's the simpleUMLDiff.ecore
file)? Manually or automated? Which syntax and semantics are used to
describe such a transformation model?
Maybe this is already explained in the paper and I didn't get it. Any hint
would be appreciated.
Best regards,
Andreas
"Davide Di Ruscio" <diruscio@di.univaq.it> schrieb im Newsbeitrag
news:pan.2007.04.07.13.06.09.195456@di.univaq.it...
> Dear all,
>
> A new ATL use case is available: A Metamodel Independent Approach to
> Difference Representation.
>
> This work presents an ATL implementation of a metamodel independent
> approach to the representation of model differences which is agnostic of
> the calculation method. Given two models being differenced which conform
> to a metamodel, their difference is conforming to another metamodel
> derived from the former by an automated transformation. Interestingly,
> difference models are first-class artifacts which in turn induce other
> transformations, such that they can be applied to one of the differenced
> models to automatically obtain the other one.
>
> The use case is available at:
> http://www.eclipse.org/m2m/atl/usecases/MMIndApproachtoDiffR ep/
>
> Regards,
> Davide
>
> --
> ------------------------------------------------------------ ---
> Davide Di Ruscio, PhD in Computer Science
> Department of Computer Science - University of L'Aquila, Italy
> Email: diruscio@di.univaq.it
> davide@diruscio.org
> WWW: http://www.di.univaq.it/diruscio
> http://www.diruscio.org
> ------------------------------------------------------------ ---
|
|
|
Re: New ATL use case: A Metamodel Independent Approach to Difference Representation [message #32499 is a reply to message #32285] |
Wed, 18 April 2007 13:33 |
Davide Di Ruscio Messages: 5 Registered: July 2009 |
Junior Member |
|
|
Dear Andrea,
the approach deals with the representation and the animation of model
differences and it is independent from the method adopted for the
calculation of the differences.
The simpleUMLDiff.ecore available in the bundle has been manually
created. It conforms to a metamodel automatically generated from a sample
UML metamodel extended with the metaclasses required to represent the
differences between UML models. Such differences can be obtained by means
of any calculation method that you prefer, but if you represent them
according to the proposed approach, then you are able to animate the
calculated differences. In fact, difference models induce other
transformations (that are automatically generated through a high-order
transformation), such that they can be applied to one of the differenced
models to automatically obtain the other one.
I hope this answers your question.
Regards,
Davide
On Tue, 17 Apr 2007 10:18:01 +0200, Andreas Ortwein wrote:
> Dear Davide,
>
>
>
> I'm a newbie on the ATL/AMMA framework and working about two months on
> the problem of describing the diff of two uml models with ATL. Last
> weekend I found this message of the new use case here on the ATL
> newsgroup. I read the regarding technical report to the use case and I'm
> convinced that this approach fits to my problem. For now I have just one
> question concerning the transformation model used to transform a base
> model to a final model: How is the diff-model generated (in the use
> case, it's the simpleUMLDiff.ecore file)? Manually or automated? Which
> syntax and semantics are used to describe such a transformation model?
>
>
>
> Maybe this is already explained in the paper and I didn't get it. Any
> hint would be appreciated.
>
>
>
> Best regards,
>
> Andreas
>
>
>
>
>
> "Davide Di Ruscio" <diruscio@di.univaq.it> schrieb im Newsbeitrag
> news:pan.2007.04.07.13.06.09.195456@di.univaq.it...
>> Dear all,
>>
>> A new ATL use case is available: A Metamodel Independent Approach to
>> Difference Representation.
>>
>> This work presents an ATL implementation of a metamodel independent
>> approach to the representation of model differences which is agnostic
>> of the calculation method. Given two models being differenced which
>> conform to a metamodel, their difference is conforming to another
>> metamodel derived from the former by an automated transformation.
>> Interestingly, difference models are first-class artifacts which in
>> turn induce other transformations, such that they can be applied to one
>> of the differenced models to automatically obtain the other one.
>>
>> The use case is available at:
>> http://www.eclipse.org/m2m/atl/usecases/MMIndApproachtoDiffR ep/
>>
>> Regards,
>> Davide
>>
>> --
>> ------------------------------------------------------------ --- Davide
>> Di Ruscio, PhD in Computer Science Department of Computer Science -
>> University of L'Aquila, Italy Email: diruscio@di.univaq.it
>> davide@diruscio.org
>> WWW: http://www.di.univaq.it/diruscio
>> http://www.diruscio.org
>> ------------------------------------------------------------ ---
|
|
|
Re: New ATL use case: A Metamodel Independent Approach to Difference Representation [message #32911 is a reply to message #32499] |
Fri, 20 April 2007 13:09 |
Eclipse User |
|
|
|
Originally posted by: andreas.ortwein.gmx.de
Hello Davide,
thanks a lot for your answer! I think, now I understand your approach right.
With a calculation algorithm one can get the difference between two models,
but with your approach one can get the way (by means of a transformation(s))
to generate a final model from an initial model, e.g. described with the
transformation language ATL. And this is the transformative property, which
neither the edit scripts nor the coloring methods support. Right?
Best regards,
Andreas
"Davide Di Ruscio" <diruscio@di.univaq.it> schrieb im Newsbeitrag
news:pan.2007.04.18.13.33.17.418430@di.univaq.it...
> Dear Andrea,
>
> the approach deals with the representation and the animation of model
> differences and it is independent from the method adopted for the
> calculation of the differences.
>
> The simpleUMLDiff.ecore available in the bundle has been manually
> created. It conforms to a metamodel automatically generated from a sample
> UML metamodel extended with the metaclasses required to represent the
> differences between UML models. Such differences can be obtained by means
> of any calculation method that you prefer, but if you represent them
> according to the proposed approach, then you are able to animate the
> calculated differences. In fact, difference models induce other
> transformations (that are automatically generated through a high-order
> transformation), such that they can be applied to one of the differenced
> models to automatically obtain the other one.
>
> I hope this answers your question.
>
> Regards,
> Davide
>
> On Tue, 17 Apr 2007 10:18:01 +0200, Andreas Ortwein wrote:
>
>> Dear Davide,
>>
>>
>>
>> I'm a newbie on the ATL/AMMA framework and working about two months on
>> the problem of describing the diff of two uml models with ATL. Last
>> weekend I found this message of the new use case here on the ATL
>> newsgroup. I read the regarding technical report to the use case and I'm
>> convinced that this approach fits to my problem. For now I have just one
>> question concerning the transformation model used to transform a base
>> model to a final model: How is the diff-model generated (in the use
>> case, it's the simpleUMLDiff.ecore file)? Manually or automated? Which
>> syntax and semantics are used to describe such a transformation model?
>>
>>
>>
>> Maybe this is already explained in the paper and I didn't get it. Any
>> hint would be appreciated.
>>
>>
>>
>> Best regards,
>>
>> Andreas
>>
>>
>>
>>
>>
>> "Davide Di Ruscio" <diruscio@di.univaq.it> schrieb im Newsbeitrag
>> news:pan.2007.04.07.13.06.09.195456@di.univaq.it...
>>> Dear all,
>>>
>>> A new ATL use case is available: A Metamodel Independent Approach to
>>> Difference Representation.
>>>
>>> This work presents an ATL implementation of a metamodel independent
>>> approach to the representation of model differences which is agnostic
>>> of the calculation method. Given two models being differenced which
>>> conform to a metamodel, their difference is conforming to another
>>> metamodel derived from the former by an automated transformation.
>>> Interestingly, difference models are first-class artifacts which in
>>> turn induce other transformations, such that they can be applied to one
>>> of the differenced models to automatically obtain the other one.
>>>
>>> The use case is available at:
>>> http://www.eclipse.org/m2m/atl/usecases/MMIndApproachtoDiffR ep/
>>>
>>> Regards,
>>> Davide
>>>
>>> --
>>> ------------------------------------------------------------ --- Davide
>>> Di Ruscio, PhD in Computer Science Department of Computer Science -
>>> University of L'Aquila, Italy Email: diruscio@di.univaq.it
>>> davide@diruscio.org
>>> WWW: http://www.di.univaq.it/diruscio
>>> http://www.diruscio.org
>>> ------------------------------------------------------------ ---
|
|
|
[ATL] Re: New ATL use case: A Metamodel Independent Approach to Difference Representation [message #33402 is a reply to message #32911] |
Mon, 23 April 2007 13:39 |
Davide Di Ruscio Messages: 5 Registered: July 2009 |
Junior Member |
|
|
Hello Andreas,
sorry for the late answer. Yes, you understood our attempt.
Regards,
Davide
On Fri, 20 Apr 2007 15:09:54 +0200, Andreas Ortwein wrote:
> Hello Davide,
>
> thanks a lot for your answer! I think, now I understand your approach right.
> With a calculation algorithm one can get the difference between two models,
> but with your approach one can get the way (by means of a transformation(s))
> to generate a final model from an initial model, e.g. described with the
> transformation language ATL. And this is the transformative property, which
> neither the edit scripts nor the coloring methods support. Right?
>
>
> Best regards,
> Andreas
>
>
> "Davide Di Ruscio" <diruscio@di.univaq.it> schrieb im Newsbeitrag
> news:pan.2007.04.18.13.33.17.418430@di.univaq.it...
>> Dear Andrea,
>>
>> the approach deals with the representation and the animation of model
>> differences and it is independent from the method adopted for the
>> calculation of the differences.
>>
>> The simpleUMLDiff.ecore available in the bundle has been manually
>> created. It conforms to a metamodel automatically generated from a sample
>> UML metamodel extended with the metaclasses required to represent the
>> differences between UML models. Such differences can be obtained by means
>> of any calculation method that you prefer, but if you represent them
>> according to the proposed approach, then you are able to animate the
>> calculated differences. In fact, difference models induce other
>> transformations (that are automatically generated through a high-order
>> transformation), such that they can be applied to one of the differenced
>> models to automatically obtain the other one.
>>
>> I hope this answers your question.
>>
>> Regards,
>> Davide
>>
>> On Tue, 17 Apr 2007 10:18:01 +0200, Andreas Ortwein wrote:
>>
>>> Dear Davide,
>>>
>>>
>>>
>>> I'm a newbie on the ATL/AMMA framework and working about two months on
>>> the problem of describing the diff of two uml models with ATL. Last
>>> weekend I found this message of the new use case here on the ATL
>>> newsgroup. I read the regarding technical report to the use case and I'm
>>> convinced that this approach fits to my problem. For now I have just one
>>> question concerning the transformation model used to transform a base
>>> model to a final model: How is the diff-model generated (in the use
>>> case, it's the simpleUMLDiff.ecore file)? Manually or automated? Which
>>> syntax and semantics are used to describe such a transformation model?
>>>
>>>
>>>
>>> Maybe this is already explained in the paper and I didn't get it. Any
>>> hint would be appreciated.
>>>
>>>
>>>
>>> Best regards,
>>>
>>> Andreas
>>>
>>>
>>>
>>>
>>>
>>> "Davide Di Ruscio" <diruscio@di.univaq.it> schrieb im Newsbeitrag
>>> news:pan.2007.04.07.13.06.09.195456@di.univaq.it...
>>>> Dear all,
>>>>
>>>> A new ATL use case is available: A Metamodel Independent Approach to
>>>> Difference Representation.
>>>>
>>>> This work presents an ATL implementation of a metamodel independent
>>>> approach to the representation of model differences which is agnostic
>>>> of the calculation method. Given two models being differenced which
>>>> conform to a metamodel, their difference is conforming to another
>>>> metamodel derived from the former by an automated transformation.
>>>> Interestingly, difference models are first-class artifacts which in
>>>> turn induce other transformations, such that they can be applied to one
>>>> of the differenced models to automatically obtain the other one.
>>>>
>>>> The use case is available at:
>>>> http://www.eclipse.org/m2m/atl/usecases/MMIndApproachtoDiffR ep/
>>>>
>>>> Regards,
>>>> Davide
>>>>
>>>> --
>>>> ------------------------------------------------------------ --- Davide
>>>> Di Ruscio, PhD in Computer Science Department of Computer Science -
>>>> University of L'Aquila, Italy Email: diruscio@di.univaq.it
>>>> davide@diruscio.org
>>>> WWW: http://www.di.univaq.it/diruscio
>>>> http://www.diruscio.org
>>>> ------------------------------------------------------------ ---
|
|
|
Goto Forum:
Current Time: Tue Mar 19 01:12:17 GMT 2024
Powered by FUDForum. Page generated in 0.01576 seconds
|