Combined mining of several kinds of concerns

Abstract

Over the last years, many concern mining techniques were proposed aiming to identify concerns in source-code of systems. The main motivation is to restructure a system as soon as the concerns are identified, leading to a target system adherent to new paradigms, more modular and with better maintainability levels. In general, these techniques are used in maintainability or reengineering process of legacy systems, where the first step is to understand the structure and internal characteristics of these systems. However, mining techniques can also be used to mine features when the aim is to start the building of a product line. Nowadays, most of the techniques and tools have focus on mining just one kind of concern, usually the classical crosscutting concerns, making the mining of other concern kinds, such as features, design patterns and architectural styles, hard or even impossible. In order to restructure a system in a suitable way, taking into account the best design decisions, it must be possible to visualize not only the classical existing crosscutting concerns as well as the impact that these concerns have on the others. Therefore, the aim in this project is to evolve an existing mining tool called ComSCID to be possible to mine three other concern types: features, design patterns and architectural styles. In this way, it will be possible to analyze the impact that a specific crosscutting concern has on the others. It is expected the tool provides useful information so that the one responsible for a restructuring process of by the construction of a product line can analyze different target scenarios previous to decide for one of them. (AU)