A Metamorphic Relation based Approach for Testing a Chess Game

Abstract

Game software is becoming complex, therefore testing game software can be a tedious task as it can result in multiple complex and possible outcomes at a given point. Also most of testing is focused on certain parts of the software because of resource constraints. The game of chess also poses this problem as there are a large number of ways in which it can be played at a given point so testing it completely is impractical. We have proposed a metamorphic testing approach for testing chess game by using metamorphic relations for different moves of chess pieces to perform testing. We validated our proposed approach on an open source chess game by seeding errors. The results show that our approach for testing chess game through metamorphic testing reveals all the seeded errors. In the future, we can extend this approach to other complex games.