Customizable visualization of quality metrics for object-oriented variability implementations

Abstract

Many large-scale software systems intensively implement variability to reuse software and speed up development. Such mechanisms, however, bring additional complexity, which eventually leads to technical debt, threatening the software quality, and hampering maintenance and evolution. This is especially the case for variability-rich object-oriented (OO) systems that implement variability in a single codebase. They heavily rely on existing OO mechanisms to implement their variability, making them especially prone to variability debt at the code level. In this paper, we propose Vari-Metrics, an extension of a visualization relying on the city metaphor to reveal such zones of indebted OO variability implementations. VariMetrics extends the VariCity visualization and displays standard OO quality metrics, such as code duplication, code complexity, or test coverage, as additional visual properties on the buildings representing classes. Extended configuration options allow the user to choose and combine quality metrics, uncovering the critical zones of OO variability implementations. We evaluate VariMetrics both by reporting on the exposed quality-critical zones found on multiple large open-source projects, and by correcting the reported issues in such zones of one project, showing an improvement in quality. CCS CONCEPTS• Software and its engineering → Software product lines; Software reverse engineering; Object oriented architectures; • Human-centered computing → Visualization systems and tools.

IDE-assisted visualization of indebted OO variability implementations

“…All the toolchain components (symfinder, metrics fetching and VariMetrics) are packed in several Docker images to ease reuse of all or parts of the toolchain. The IDE integration is developed using the IntelliJ Platform SDK 7 and is installed as any other plugin for the IDE.…”

“…The VariMetrics visualization has been quantitatively evaluated on seven industrial open-source projects from several thousands to 2.4MLoC. Using these metrics, the exhibited classes were both critical and concentrating variability implementations [7]. A qualitative evaluation was also conducted on the JFreeChart project, with the refactoring of the indebted zones detected by VariMetrics.…”

“…2, where the two classes in the white dotted box, CategoryPlot and XYPlot, exhibit multiple method overloads (due to their height) and little test coverage (due to their cracks). Refactoring those classes led to an improvement of those classes and the project's overall quality [7]. More details on this evaluation are available on VariMetrics webpage 3 .…”

“…Moreover, with some better knowledge of the variability implementations, the need for determining their technical debt naturally arises. In a companion paper [7], the authors have proposed VariMetrics, an extension of VariCity to display zones of OO variability implementations with technical debt. It provides additional visual properties on the buildings representing classes to display some quality metrics, such as code duplication, code complexity, and test coverage.…”

“…External information about VariMetrics-IDE consists in a video available at https: //youtu.be/lRPK8nS5JMc and the experimental results from[7] available at https://deathstar3.github.io/varimetrics-demo/.…”