Investigating optimum refactoring sequence using hill-climbing algorithm

Abstract

The surface indication of a deeper problem in the source code is bad smells that required to be removed as soon as possible. Not all smells are harmful, but if ignored, they automatically become faults which will further deteriorate the quality of the software. Therefore, this damage can be prevented with the help of refactoring that changes the internal attributes without affecting its external behavior. Every class of the source code contains numerous types of bad smells; therefore, more than one refactoring technique needs to be applied to clean the code. In this paper, we have tried to figure out the most optimum refactoring sequence that should be applied in a critically affected class of an open-source software (OSS), jTDS, to maximize its maintainability value with minimal effort. The results indicate improvement in maintainability value by 14.06% after the application of refactoring techniques. This study would be beneficial for the software maintenance team as this will encourage them to complete their work with much ease and will make them more efficient.

Subject Classification:

• Primary 68N30
• Secondary 68T20

Keywords:

• Software refactoring
• Maintainability
• Quality
• Bad smell
• Hill-climbing algorithm