Um estudo sobre a natureza dos conflitos de merge de software na linguagem Python

Loading...
Thumbnail Image

Journal Title

Journal ISSN

Volume Title

Publisher

Universidade Federal de Juiz de Fora (UFJF)

DOI

Abstract

Introduction: The increasing complexity of software projects demands intense collaboration among developers, which often results in conflicts during code integration. Previous studies have investigated these conflicts but generally focus on languages like Java, leaving a gap in understanding conflicts in languages with distinct syntactic and structural characteristics, such as Python. Methods: This study analyzed merge conflicts in 50 Python repositories hosted on GitHub. An automated approach was developed to extract and classify merges from the project history, distinguishing between those with and without conflicts. Subsequently, conflicting chunks were identified from the conflict delimiters. The analysis included the extraction of metrics, such as the number and size of chunks (in lines of code), the identification of language constructs present in python files, and the categorization of developer decisions. Results: The data indicate that most failed merges involve a small number of conflicting chunks, with a median of 1 chunk per merge and 75% of chunks containing up to 13 lines of code, although extreme cases with thousands of lines occur. The analysis of language constructs revealed that elements such as method calls and variable declarations are the most prevalent, while the most common resolution strategies are the complete preservation of one of the versions (Version 1 and Version 2). Discussion: The findings suggest that while most results are similar between two different languages, it was possible to observe the influence that unique language characteristics can have on the nature of the Merge. Conclusion: This study expands the understanding of merge conflicts in Python projects, providing an empirical basis that complements studies conducted in other languages. The results offer a foundation for the development of future automated conflict resolution tools and for further investigations in the area of collaborative code integration.

Description

Citation

Endorsement

Review

Supplemented By

Referenced By

Creative Commons license

Except where otherwise noted, this item's license is described as Acesso Aberto