Search-Based Test Suite Generation for Rust

Rust is a robust programming language which promises safety and performance at the same time. Despite its young age, it has already convinced many and has been one of the most popular programming languages among developers since its first release. However, like any other software, Rust programs need to be tested extensively. In this work, we propose the first search-based tool, called RUSTYUNIT, for automatic generation of unit tests for Rust programs. RusTYUNIT incorporates a compiler wrapper, which statically analyzes and instruments a given program to generate and evaluate tests targeting high code coverage using a many-objective genetic algorithm. An initial empirical study using 6 real-world open-source Rust libraries demonstrates the feasibility of our approach but also highlights important differences and open challenges for test generation for Rust programs.