Abstract
Test smells are patterns in test code that may indicate poor code quality. Some recent studies have cast doubt on the accuracy and usefulness of the test smells proposed and studied by the research community. In this study, we aimed to determine whether practitioners view these test smells as sources of technical debt worth spending effort to remove. We selected 12 substantial open-source software systems and mapped how 19 test smells from the literature were introduced and removed from the code base over time. Out of these 19 smells, our results show that: 1) four test smells were rarely detected in our selected projects; 2) three test smells are removed rapidly from the projects while another three are removed from code bases slowly; 3) the remaining nine test smells did not show a consistent pattern of quick or delayed removal. Our results suggest that the test smells currently being studied by researchers do not capture the true concerns of developers regarding test quality, with current testing tool sets, with only three of the 19 smells studied showing clear evidence of developer concern.
Original language | English |
---|---|
Title of host publication | IFIP-ICTSS 35th International Conference on Testing Software and Systems |
Publication status | Accepted/In press - 30 Jun 2023 |
Keywords
- Test Smells
- Software Testing
- Empirical Software Engineering