Automated slicing scheme for test case prioritization in regression testing
Manju Kaushal, Satheesh Abimannan · International Journal of Engineering & Technology · 2019
Motivation: The testing approach which ensures that the software does not have any adverse affects due to the changes made in the existing features or addition of some new features is called regression testing. For testing the changes made in the previous versions of software, this type of testing is performed. To ensure that the numbers of test cases available in the software is not too large, it is important to select the regression tests and to do so, several techniques have been designed. To detect the individual functions from the software, the existing work applied the slicing technique. The parameters which are used in this approach are calculated manually to analyze importance of individual functions. The number of times the function is encountered and the number of functions relevant to the specific function are the two different parameters calculated here. A list of changes in the source code and the execution traces generated from the test cases which are run on previous versions are used to combine the modification, minimization and prioritization-based selection which thus generates a hybrid technique.Problem Statement: In the existing system, the manual slicing technique is applied to perform test case prioritization. In manual slicing, the total number of times a function is triggered and the total numbers of functions attached are calculated manually to generate final function importance. This approach is very time consuming and inaccurate.Method: In this paper, we studied that to prioritize the test cases based on the changes, a type of regression testing is used which is test case prioritization. The test cases of the functions which have higher priority are executed first and so on. Based on the changes made, the test cases are prioritized. For identifying maximum number of faults from the modified software, manual slicing and automated slicing are applied in this work. The proposed method will be the enhancement of manual slicing technique. The automated slicing technique will automatically calculate the functional importance based on number of attached functions and number of times function triggered. The proposed method has low execution time and detects more number of defects from the software. The dataset of ten different projects is used to test the performances of proposed and existing algorithms in MATLAB. Each project has seven functions and four numbers of changes are defined for the regression testing.Results: The simulation results achieved at the end show that in comparison to manual methods, the implementation of automated test case prioritization has provided improvement in the fault detection rate and reduction in the execution time.