Definition and detection of bad smells of aspect-oriented program

Abstract

“Bad smell” is a metaphor describing software patterns that are generally associated with bad designs and bad programmings. It can be removed by using the refactoring technique which improves the quality of the software. Aspect-Oriented (AO) software development, which involves new notions and different ways of thinking for developing software and solving the crosscutting problem, possibly introduces different kinds of design flaws. Defining bad-smell kinds hidden in AO software, in order to point out bad designs and bad programmings, is then necessary. This research proposes the definition of five new kinds of AO bad smells affecting coupling of software. Moreover, appropriate solution to eliminate each kind of bad smell is presented in terms of refactoring procedure. Also, metrics are designed and thresholds are determined to support for detecting such kinds of bad smells. Bad-smell detector is further developed as a tool to support for automatic bad-smell detection. The results of bad-smell validation show that after removing the bad-smell kinds, software coupling is decreased.