Software product lines can be implemented with different techniques. Those techniques can be grouped into annotation-based and composition-based approaches, with complementary strengths and weaknesses. A combination seems useful to utilize benefits of both groups but using two techniques in parallel may cause new problems. To our knowledge, there is no approach that integrates composition into an annotation-based approach or vice versa. We propose the use of an extended preprocessor to introduce physical separation of concerns similar to feature-oriented programming. In this paper, we i) present a preliminary implementation that integrates composition into annotation, ii) analyse its benefits and shortcomings, and iii) discuss implementation and tooling. Overall, we enable developers to keep on using familiar preprocessors but also to benefit from composition. Finally, we show the potential of our approach.
Sun 30 Oct
|15:40 - 16:10|
David WilleTU Braunschweig, Germany, Sandro SchulzeTU Hamburg, Germany, Ina SchaeferTU Braunschweig, GermanyDOI
|16:10 - 16:40|
Jacob KrügerHarz University of Applied Sciences, Germany, Ivonne SchröterUniversity of Magdeburg, Germany, Andy KennerMETOP Magdeburg, Germany, Christopher KruczekMETOP Magdeburg, Germany, Thomas LeichHarz University of Applied Sciences, GermanyDOI
|16:40 - 16:50|
Lightning Talk: Comparing Variability Mechanisms for Requirements and Test Artifacts in Behaviour-Driven Development
Neil MatherARC Technology Ltd.
|16:50 - 17:20|
Fine-Grained Test Case Prioritization for Integration Testing of Delta-Oriented Software Product Lines
Remo LachmannTechnische Universität Braunschweig, Sascha LityTechnische Universität Braunschweig, Mustafa Al-HajjajiMagdeburg University, Franz E. FürchtegottTechnische Universität Braunschweig, Ina SchaeferTU Braunschweig, GermanyDOI