Exploring The Effect of Code Coverage And Maintainability for Identifying Software Testability

Abstract

The ability of code to reveal its flaws, especially during automated testing, is known as software testability. The program being tested must be able to with stand testing. The coverage of the test data provided by a specific test data generation algorithm, on the other hand, is what determines whether a test will be successful. To clarify whether and how software testability affects test coverage. However little empirical evidence has been presented. In this article, we suggest a technique to clarify this issue. The testability of programs is determined using a variety of source code metrics, and our suggested framework builds machine learn ing models using the coverage of Software Under Test (SUT) provided by various automatically generated test suites.The cost of additional testing is decreased be cause the resulting models can anticipate the code coverage offered by a particular test data generation algorithm before the algorithm is even run.To measure the testability of source code, a concrete proxy called predicted coverage is used. The correlation between code coverage and maintainability is crucial in assessing the testability of software, as high code coverage combined with well-maintained code facilitates the creation of comprehensive test cases and ensures thorough testing of critical paths and edge cases.