 |
 |
 |
 |
 |
Direct links |
The vital lead of my research activities deals with the development of object-oriented applications. In particular, I am interested in the study of the various techniques which on one hand, could improve the robustness, the readability, the documentation, the capacities of adaptation and evolution of the applications and on the other hand, would minimize the time necessary for their implementation on software platforms. We promote the idea that it is necessary to reduce the gap between the phases of design and programming until to have a continuous process of development, for example by transformation of models. Our contribution can stand out in four phases which took place sometimes successively, sometimes in parallel according to the reflections and the opportunities of collaboration.
A summary of my research activities in french
A summary of other activities related to research in french
Warning: Since the middle of 2012 my research activities are in stand by because of my implication as vice president of the university in charge of the finances (2012-2016). Since the middle of 2016 I am vice president of the university in charge of the finances, the information system and the evaluation of the means.
Aspect-Oriented Modelling - SmartAdapters |
Programming Language Composition
First version of SmartAdapters is based on JAdaptor. It is an experimental Software Platform dedicated to software architects. We aim for example to guide them when they want to integrate a design pattern into a java programming language. This platform allows to identify, split and reuse concerns (set of java classes). These concerns are equiped with adapters making possible the composition via a generic engine.
access to the site of the project
See publications here
Model Composition
The Aspect-Oriented Software Development (AOSD) paradigm offers to tame the complexity of large software systems by providing advanced mechanisms for separating and composing concerns, formerly at the code level. It is now gaining attention in the earlier steps of the software life-cycle, inspired by the Model-Driven Engineering paradigm. This leads to the creation of numerous Aspect-Oriented Modeling approaches that aim to compose aspect models into various base models. These approaches generally suffer from a lack of flexibility, limiting the reuse of aspect models to very related contexts. A reusable aspect model should ideally offers :
A base model may not provide all the elements required by the aspect but it still may be augmented with the aspect. We argue that the aspect should be able to adapt to di fferent contexts, by reusing existing elements or introducing some missing elements, provided that a set of mandatory elements is present in the base model. Inspired by the Software Product Line paradigm, the SmartAdapters approach integrates variability mechanisms in the different parts that constitute an aspect so that, it allows designing aspect models as a family of composition protocols. As a result, designers are able to address a wider range of situations where an aspect model could be composed.
See publications here
access to the site of the project
Software Product-Lines - FAMILIAR |
Feature models are a popular formalism for managing variability in software product lines (SPLs). Realistic SPLs involve the modeling of a large number of features to comprehensively represent different viewpoints, sub-systems or concerns of the software system. To manage complexity, there is a need to separate, inter-relate and compose several feature models while automating the reasoning on their compositions -- from validity checks to configuration process. We propose a Domain-Specific Language (DSL) that is dedicated to the management of feature models and that complements existing tool support.
The DSL, called FAMILIAR (for FeAture Model scrIpt Language for manIpulation and Automatic Reasoning), is an executable language that supports manipulating and reasoning about feature models.
access to the site of the project
See publications here
Reverse Inheritance relationship for the Eiffel programming language |
Generalization is common in object-oriented modelling. It would be useful in many situations also as a language mechanism, reverse inheritance, but there have been only few detailed proposals for that. We define reverse inheritance as a true inverse of ordinary inheritance, without changing anything else in the language that is extended. Eiffel is perhaps the most suitable language for that purpose because of its flexible inheritance principles. Moreover, there exists good previous work on Eiffel, on which we have built. Some of the most interesting aspects taken into account by our extension are: polymorphism, assertions, various reverse inheritance clauses and in particular redefinition and renaming, signature adaptation.
access to the site of the project
See publications here
Steering and Program committees, etc. |
Supervision of research |
In french DEA stands for "Diplôme d'études Approfondies"
Collaborations |
Other responsabilities |
(*) GDR means in french "Groupement De Recherche". They are group around the country dedicated to a set of related topics.
| Author: Philippe Lahire (version 1.0) |
Last update: 11-Nov-2016 6:29 PM |
