A Classification of Refactoring Methods Based on Software Quality Attributes

Abstract

Refactoring is the process of improving the design of existing code by changing its internal structure without affecting its external behavior. Refactoring tends to improve software quality by improving design, improving readability and reducing 'bugs'. There are many different refactoring methods, each having a particular purpose and effect. Consequently, the effect of refactoring methods on software quality attributes may vary. Moreover, it is often unclear to software designers how to use refactoring methods to improve specific quality attributes. In this paper, we propose a classification of refactoring methods based on their measurable effect on software quality attributes. This, in turn, helps software designers choose appropriate refactoring methods that will improve the quality of their designs, based on the design objectives. It also enables them to predict the quality drift caused by using particular refactoring methods.

“Refactoring” Refactoring

“…Remediation is prescribed as primarily grammatical or structural often amenable to formulaic manipulations that may be automated in code editors or IDE's [13,25]. The four non-structural exceptions are instances where refactoring actions apply "naming" or "renaming" of code elements to clarify or signify the programmer's motivation; these refactorings incorporate knowledge of the purpose of the code that cannot be gleaned from the source code alone in either the activity of recognition or remediation.…”

Optimising Pair Programming in a Scholastic Framework: A Design Pattern Perspective

“…• Data Abstraction Coupling (DAC): This metric measures the number of instantiations of other classes within a single class (Elish and Alshayeb, 2011) DAC = number of ADTs defined in a class. ADT's refer to abstract data types.…”

The Downsides of Software Refactoring

“…Further research by Dandashi revealed that complexity and LOC were found to be directly proportional to indirect software quality attributes, while NOC, DIT, RFC, and CBO are inversely proportional to indirect software quality attributes. (Dandashi, 2002) C. More Empirical Results of Refactoring -the Elish and Alshayeb Study (Elish, 2011) Elish's study is very significant as it expanded the work by Dandashi by taking code components, measuring their internal directly measurable metrics (based upon Dandashi's metrics) and then performing nine different refactoring activities on that code. After refactoring activities, he then re-analyzed a number of software metrics to note any changes due to the refactoring.…”