Projets par an
Context: Software product lines (SPL) engineering defines a set of systems that share common features and artifacts to achieve high productivity, quality, market agility, low time to market, and cost. An SPL product is derived from a configuration of features which need to be compounded together without violating their particular specifications. While it is easy to identify the behavior of a feature in isolation, specifying and resolving interactions among features may not be a straightforward task. The feature interaction problem has been a challenging subject for decades. Objective: This study aims at surveying existing research on feature interaction in SPL engineering in order to identify common practices and research trends. Method: A systematic mapping study was conducted with a set of seven research questions, in which the 35 studies found are mainly classified regarding the feature interaction solution presented: detection, resolution and general analysis. Results: 43% of the papers deal with feature interaction at early phases of a software lifecycle. The remaining is shared among the other categories: source code detection, resolution and analysis. For each category, it was also identified the main strategies used to deal with interactions. Conclusions: The findings can help to understand the needs in feature interaction for SPL engineering, and highlight aspects that still demand an additional investigation. For example, often strategies are partial and only address specific points of a feature interaction investigation.