Paper information
Title
Cardinality-based feature models and Model-Driven Engineering: fitting them together
Cardinality-based feature models and Model-Driven Engineering: fitting them together
Published in
Proceedings of the Fourth International Workshop on Variability Modelling of Software-Intensive Systems. ICB-research report No. 37. Pages 61--68. ISSN 1860 - 2770 (Print), ISSN 1866 - 5101 (Online) - 2010
Proceedings of the Fourth International Workshop on Variability Modelling of Software-Intensive Systems. ICB-research report No. 37. Pages 61--68. ISSN 1860 - 2770 (Print), ISSN 1866 - 5101 (Online) - 2010
Abstract
Feature Modeling is a technique which uses a specific visual notation to characterize the variability of product lines by means of diagrams. In this sense, the arrival of metamodeling frameworks in the Model-Driven Engineering field has provided the necessary background to exploit these diagrams (called feature models) in complex software development processes. However, these frameworks (such as the Eclipse Modeling Framework) have some limitations when they must deal with software artifacts at several abstraction layers. This paper presents a prototype that allows the developers to define cardinality-based feature models with constraints. These models are automatically translated to Domain Variability Models (DVM) by means of model-to-model transformations. Thus, such models can be instantiated, and each different instantiation is a configuration of the feature model. This appproach allows us to take advantage of existing generative programming tools, query languages and validation formalisms; and, what is more, DVMs can play a key role in MDE processes as they can be used as inputs in complex model transformations.
Feature Modeling is a technique which uses a specific visual notation to characterize the variability of product lines by means of diagrams. In this sense, the arrival of metamodeling frameworks in the Model-Driven Engineering field has provided the necessary background to exploit these diagrams (called feature models) in complex software development processes. However, these frameworks (such as the Eclipse Modeling Framework) have some limitations when they must deal with software artifacts at several abstraction layers. This paper presents a prototype that allows the developers to define cardinality-based feature models with constraints. These models are automatically translated to Domain Variability Models (DVM) by means of model-to-model transformations. Thus, such models can be instantiated, and each different instantiation is a configuration of the feature model. This appproach allows us to take advantage of existing generative programming tools, query languages and validation formalisms; and, what is more, DVMs can play a key role in MDE processes as they can be used as inputs in complex model transformations.
BibTeX
@misc{issi_web:id:335,
title = "Cardinality-based feature models and Model-Driven Engineering: fitting them together",
author = "Abel Gómez Llana and Isidro Ramos Salavert",
booktitle = "Proceedings of the Fourth International Workshop on Variability Modelling of Software-Intensive Systems. ICB-research report No. 37. Pages 61--68. ISSN 1860 - 2770 (Print), ISSN 1866 - 5101 (Online)",
year = "2010",
eprint = "http://issi.dsic.upv.es/publications/archives/f-1263570538890/document.pdf",
url = "http://www.vamos-workshop.net/proceedings/VaMoS_2010_Proceedings.pdf",
abstract = "Feature Modeling is a technique which uses a specific visual notation to characterize the variability of product lines by means of diagrams. In this sense, the arrival of metamodeling frameworks in the Model-Driven Engineering field has provided the necessary background to exploit these diagrams (called feature models) in complex software development processes. However, these frameworks (such as the Eclipse Modeling Framework) have some limitations when they must deal with software artifacts at several abstraction layers. This paper presents a prototype that allows the developers to define cardinality-based feature models with constraints. These models are automatically translated to Domain Variability Models (DVM) by means of model-to-model transformations. Thus, such models can be instantiated, and each different instantiation is a configuration of the feature model. This appproach allows us to take advantage of existing generative programming tools, query languages and validation formalisms; and, what is more, DVMs can play a key role in MDE processes as they can be used as inputs in complex model transformations."
}