Analyzing the Impact of CI Sub-practices on Continuous Code Quality in Open-Source Projects: An Empirical Study
Resumo
Continuous Integration (CI) is a widely adopted practice that automates and enhances the frequency of code integration. Previous research has explored the relationship between CI sub-practices (such as frequent commit activity and high test coverage) and software quality. However, limited knowledge exists regarding the impact of specific CI sub-practices on the Continuous Code Quality (CCQ) inspection outcomes of software projects, such as technical debts, bug density, duplicated lines, and code smells. This paper aims to analyze the extent to which the adoption of CI sub-practices improves CCQ outcome metrics in software projects. First, we investigate the association between Travis CI adoption, Travis CI usage maturity, and improved quality inspection outcomes in a set of 75 popular open-source projects from GitHub. Additionally, we use data from other 97 open-source projects to identify specific CI sub-practices, such as maintaining shorter build durations, that exhibit a strong correlation with enhanced quality inspection outcomes. Our findings reveal that the quality inspection outcomes are improved in projects with higher test coverage and shorter build duration. Merely adopting a CI service is not a guarantee for improved quality outcomes. Therefore, it is essential for developers to consistently adhere to recommended CI sub-practices. By doing so, they can effectively harness the benefits of CI in their projects and achieve improved quality outcomes.
Palavras-chave:
software quality, software maintenance, empirical software engineering, continuous integration
Publicado
25/09/2023
Como Citar
FREITAS, Guilherme; BERNARDO, João Helis; SIZÍLIO, Gustavo; COSTA, Daniel Alencar Da; KULESZA, Uirá.
Analyzing the Impact of CI Sub-practices on Continuous Code Quality in Open-Source Projects: An Empirical Study. In: SIMPÓSIO BRASILEIRO DE ENGENHARIA DE SOFTWARE (SBES), 37. , 2023, Campo Grande/MS.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2023
.
p. 1–10.