Fostering Mutation Testing with Variability-Aware Execution

Abstract

Mutation Testing is a fault-based technique commonly used to assess the effectiveness of software tests. It basically consists on introducing syntactical changes (called mutants) into source code and checking whether the test cases distinguish them. Since there are dozens of distinct mutant types, one of the most challenging problems faced by the community is the high computational effort required for testing the whole test suites for each generated mutant. Over last four decades, researchers proposed techniques aiming at effort reduction in mutation testing phases, e.g., either by reducing the number of mutants or by optimization. In this Master Thesis, we proposed a technique for mutation testing cost reduction in running time, inspired on variability-aware execution, a novel approach that achieves cost reduction in the large configuration space of configurable software systems. We assessed viability of the proposed technique and conclude it does deserve further investigation.