Android Source Code Smells: A Systematic Literature Review

Muhammad Fawad

doi:10.1002/spe.3401

Return

Android Source Code Smells: A Systematic Literature Review

2024-12-27

PRE

DOI：10.1002/spe.3401

Original request for help

Save

Abstract

En 中文

IntroductionModern software developers strive to develop applications that are robust, easy to maintain, and possess high levels of quality. However, code smells can hinder this goal as they are visible signs of underlying issues. Numerous techniques and tools have been proposed for detecting code smells in various contexts and programming languages. Despite this, research on Android-specific code smells and their impact on external quality attributes is still in its early stages.ObjectiveThis study aims to provide a comprehensive summary of state-of-the-art techniques, tools, and approaches used for detecting and refactoring code smells in Android applications.MethodologyA systematic literature review was conducted between November 2007 and December 2023, adhering to standard guidelines. In total, 79 primary studies were identified, analyzed, and synthesized.ResultsA total of 237 code smells were identified using 51 techniques and tools, based on seven distinct approaches. Efficiency was found to be the most affected external quality attribute. The code smell Durable Wakelock was the most studied.ChallengesDespite the extensive research, the software engineering community faces numerous challenges. These include a lack of in-depth investigation into Android-specific code smells, a limited number of studied quality attributes, insufficient involvement of industry experts in the research process, scarcity of Android-oriented metrics, and dependence on industry-exclusive datasets.Future DirectionsThis review suggests potential research directions that are valuable for researchers and practitioners seeking to gain insights into current open research problems in this domain.

Keywords:

android

antipattern

code smell

detection

mobile application

refactoring