SB
S. Biennier
info
Please Note
<p>This page displays the records of the person named above and is not linked to a unique person identifier. This record may need to be merged to a profile.</p>
1 records found
1
Automated testing is essential for software reliability, yet test code frequently contains test smells that degrade maintainability. Prior research has largely examined these issues through software-quality perspectives, leaving their environmental impact underexplored. This study investigates how refactoring test smells affects energy consumption, execution time, and test quality in Java JUnit suites.
We curated a dataset of open-source systems, detected smells with tsDetect and manually validated refactorable instances. For each instance, we applied literature-backed, smell-specific refactorings and measured energy with EnergiBridge under controlled conditions. The results show that energy effects are smell-specific. Removing Ignored test instances yields clear energy savings, whereas refactoring the Lazy test (JUnit 5) smell via @ParameterizedTest incurs substantial energy increases. Most other smells exhibit small or inconsistent changes. Interestingly, we found that changes in energy were also strongly coupled with changes in execution time, within our evaluation context (a controlled, CPU-bound, sequential JUnit setting).
Overall, this study extends test smell research into software sustainability and highlights trade-offs between maintainability and energy efficiency. It provides a reproducible measurement pipeline and empirical guidance on when refactoring test smells is likely to be energy-beneficial. ...
We curated a dataset of open-source systems, detected smells with tsDetect and manually validated refactorable instances. For each instance, we applied literature-backed, smell-specific refactorings and measured energy with EnergiBridge under controlled conditions. The results show that energy effects are smell-specific. Removing Ignored test instances yields clear energy savings, whereas refactoring the Lazy test (JUnit 5) smell via @ParameterizedTest incurs substantial energy increases. Most other smells exhibit small or inconsistent changes. Interestingly, we found that changes in energy were also strongly coupled with changes in execution time, within our evaluation context (a controlled, CPU-bound, sequential JUnit setting).
Overall, this study extends test smell research into software sustainability and highlights trade-offs between maintainability and energy efficiency. It provides a reproducible measurement pipeline and empirical guidance on when refactoring test smells is likely to be energy-beneficial. ...
Automated testing is essential for software reliability, yet test code frequently contains test smells that degrade maintainability. Prior research has largely examined these issues through software-quality perspectives, leaving their environmental impact underexplored. This study investigates how refactoring test smells affects energy consumption, execution time, and test quality in Java JUnit suites.
We curated a dataset of open-source systems, detected smells with tsDetect and manually validated refactorable instances. For each instance, we applied literature-backed, smell-specific refactorings and measured energy with EnergiBridge under controlled conditions. The results show that energy effects are smell-specific. Removing Ignored test instances yields clear energy savings, whereas refactoring the Lazy test (JUnit 5) smell via @ParameterizedTest incurs substantial energy increases. Most other smells exhibit small or inconsistent changes. Interestingly, we found that changes in energy were also strongly coupled with changes in execution time, within our evaluation context (a controlled, CPU-bound, sequential JUnit setting).
Overall, this study extends test smell research into software sustainability and highlights trade-offs between maintainability and energy efficiency. It provides a reproducible measurement pipeline and empirical guidance on when refactoring test smells is likely to be energy-beneficial.
We curated a dataset of open-source systems, detected smells with tsDetect and manually validated refactorable instances. For each instance, we applied literature-backed, smell-specific refactorings and measured energy with EnergiBridge under controlled conditions. The results show that energy effects are smell-specific. Removing Ignored test instances yields clear energy savings, whereas refactoring the Lazy test (JUnit 5) smell via @ParameterizedTest incurs substantial energy increases. Most other smells exhibit small or inconsistent changes. Interestingly, we found that changes in energy were also strongly coupled with changes in execution time, within our evaluation context (a controlled, CPU-bound, sequential JUnit setting).
Overall, this study extends test smell research into software sustainability and highlights trade-offs between maintainability and energy efficiency. It provides a reproducible measurement pipeline and empirical guidance on when refactoring test smells is likely to be energy-beneficial.