A Pairwise Based Method for Automated Test Data Generation for $\mathrm{C}/\mathrm{C}++$ Projects
Hoang-Viet Tran, Lam Nguyen Tung, Phạm Ngọc Hùng · 2022 RIVF International Conference on Computing and Communication Technologies (RIVF) · 2022
Pairwise testing is an effective test data generation method which is based on an observation where most software errors are caused by interactions of two factors. This testing method maintains a high defect finding ability of the generated test data set whilst keeping a small number of generated test data. This paper presents a method named PMC (a$\boldsymbol{p}$airwise based test data generation method for$\boldsymbol{C}/\boldsymbol{C}++$projects) for automated test data generation of unit testing$\mathbf{C}/\mathbf{C}++$projects. The method employs IPO (in-parameter-order) pairwise test data generation strategy and Eclipse CDT$(\mathbf{C}/\mathbf{C}++$Development Tooling) library for parsing the given unit source code to get the corresponding abstract syntax tree (AST). By traversing this AST, we retrieve the list of input parameters and simple conditions to generate associated values. These values are used as inputs of IPO strategy to generate the required test data set. Experiments are performed on some common unit functions with a significantly reduced number of generated test data in comparison with the combination test data generation method (CM). We give some discussions about the experimental results in the paper.