Empirical Evaluation of a Live Environment for Code Refactoring
Complex software can be hard to read, adapt, and maintain. Refactoring it can create cleaner and self-explanatory code. Refactoring tools try to guide developers towards better code, with more quality. However, most of them take too long to provide feedback, support, and guidance on how developers should improve their software. To reduce this problem, we explored the concept of Live Refactoring, focusing on visually suggesting and applying refactorings, in real-time. With this in mind, we developed a Live Refactoring Environment that visually identifies, recommends, and applies Extract Method refactorings. To validate it, we conducted an empirical experiment. Early results showed that our approach improved several code quality metrics. Besides, we also concluded that our results were significantly different and better than the ones from refactoring the code manually without further help.
Rename Chains: An Exploratory Study on the Occurrence and Characteristics of Identifiers Undergoing Multiple Renamings
Identifier names play a significant role in program comprehension activities, with high-quality names improving developer productivity and system quality. To correct poor-quality names, developers rename identifiers to reflect their intended purpose better. However, renames do not always result in high-quality, long-lasting names; in many cases, developers perform multiple rename operations on the same identifier throughout the system’s lifetime. In this paper, we report on a large-scale empirical study that examines the occurrence of identifiers undergoing multiple renames (i.e., rename chains). Our findings show the presence of rename chains in almost every project, with methods typically having more rename chains than other identifier types. Furthermore, it is usually the same developer responsible for creating all renames within a chain, with most names maintaining the same grammatical structure. Understanding rename chains can help us provide stronger advice, and targeted research, on how to craft high-quality, long-lasting identifiers.