Test case generation of actor systems

Abstract

Testing is a vital part of the software development process. It is even more so in the context of concurrent languages, since due to undesired task interleavings and to unexpected behaviours of the underlying task scheduler, errors can go easily undetected. Test case generation (TCG) is the process of automatically generating test inputs for interesting coverage criteria, which are then applied to the system under test. This paper presents a TCG framework for actor systems, which consists of three main elements, which are the original contributions of this work: (1) a symbolic execution calculus, which allows symbolically executing the program (i.e., executing the program for unknown input data), (2) improved techniques to avoid performing redundant computations during symbolic execution, (3) new termination and coverage criteria, which ensure the termination of symbolic execution and guarantee that the test cases provide the desired degree of code coverage. Finally, our framework has been implemented and evaluated within the aPET system.