Variation Control Systems
- Concepts, Operations, and Feasibility of a Projection-Based Variation Control System[Abstract, PDF, Code/Data]IEEE Int. Conf. on Software Maintenance and Evolution (ICSME), 2016, 323–333
Highly configurable software often uses preprocessor annotations to handle variability. However, understanding, maintaining, and evolving code with such annotations is difficult, mainly because a developer has to work with all variants at a time. Dedicated methods and tools that allow working on a subset of all variants could ease the engineering of highly configurable software. We investigate the potential of one kind of such tools: projection-based variation control systems integrated with projectional editors. For such systems we aim to understand: (i) what end-user operations they need to support, and (ii) whether they can realize the actual evolution of real-world, highly configurable software. We conduct an experiment that investigates variability-related evolution patterns and that evaluates the feasibility of a projection-based variation control system by replaying parts of the history of a real-world 3D printer firmware project. Among others, we show that the prototype variation control system does indeed support the evolution of a highly configurable system and that in general, it does not degrade the code.
- Projectional Editing of Variational SoftwareACM SIGPLAN Int. Conf. on Generative Programming and Component Engineering (GPCE), 2014, 29–38Best paper[Abstract, PDF]
Editing the source code of variational software is complicated by the presence of variation annotations, such as #ifdef statements, and by code that is only included in some configurations. When editing some configurations and not others, it would be easier to edit a simplified version of the source code that includes only the configurations we currently care about. In this paper, we present a projectional editing model for variational software. Using our approach, a programmer can partially configure a variational program, edit this simplified view of the code, and then automatically update the original, fully variational source code. The model is based on an isolation principle where edits affect only the variants that are visible in the view. We show that this principle has several nice properties that are suggested by related work on bidirectional transformations.
- #ifdef Confirmed Harmful: Promoting Understandable Software Variation[Abstract, PDF]IEEE Int. Symp. on Visual Languages and Human-Centric Computing (VL/HCC), 2011, 143–150
Maintaining variation in software is a difficult problem that poses serious challenges for the understanding and editing of software artifacts. Although the C preprocessor (CPP) is often the default tool used to introduce variability to software, because of its simplicity and flexibility, it is infamous for its obtrusive syntax and has been blamed for reducing the comprehensibility and maintainability of software. In this paper, we address this problem by developing a prototype for managing software variation at the source code level. We evaluate the difference between our prototype and CPP with a user study, which indicates that the prototype helps users reason about variational code faster and more accurately than CPP. Our results also support the research of others, providing evidence for the effectiveness of related tools, such as CIDE and FeatureCommander.