Test Input Data Generation for Choiceless Program Nets

Abstract

Software testing is an important problem in designing a large software system and this problem is difficult to be solved due to its computational complexity. We try to use program nets to approach this problem and have proposed algorithms to divide a whole program net into subnets that can structurally cover the original one based on divide-and-conquer method. This paper aims to solve the remaining task of our approach, that is how to find test input data for each subnet to be called choiceless program net. Firstly, definitions of program nets are extended and the properties of choiceless program nets are analysed. Secondly, polynomial algorithms are proposed to get all constraint conditions of any given choiceless program nets. Finally, a method to generate test input data under the obtained constraint conditions is proposed by adopting an SMT (Satisfiability Modulo Theories) solver, $Z3$ prover, and also an example is given to show the usefulness of our method.