Systematic exhortation of code smell detection using JSmell for java source code
M. Sangeetha, P. Sengottuvelan · 2017 International Conference on Inventive Systems and Control (ICISC) · 2017
“Code Smell” or “Bad Smell”, at the very least, is an indicator of badly for source code and is often analytical of deeper problems in software design. In layman terms, it signals flaws in the core foundation or architecture of the software that can cause any number of more serious problems - from usability and runtime performance to supportability and enhancement. These problems can mostly be prevented by the systematic refactoring of the code. Code smells are symptoms of deep-rooted problems in design, which, in most common cases, reduce the understandability of the system for present and future programmers, therefore depiction the program un-maintainable. Identification of these code smells has been thought of as an spontaneous art rather than an exact science, as there are very few empirical measures or methodologies for doing so. This paper implements JSmell, which will follow a scientific approach to detect five of these 22 code smells. JSmell will give suggestions to refactor the code for all five of these smells. Further, the tool will provide an interactive process to refactor two of these cases; while for the rest, it will suggest an ideal refactoring technique that would need to be applied manually.