Software Merge: A Two-Decade Systematic Mapping Study

Software merging plays a key role in many software engineering tasks, e.g., reconciling source code developed in parallel. For this reason, many studies have been proposed in the last two decades. However, the current literature still lacks a classification of currently available approaches and research carried out considering software merge. Hence, a broad understanding of these already published works remains limited and inconclusive. Even worse, research gaps are not identified or prioritized. This study, therefore, provides a classification and thematic analysis of studies about software merge in the last two decades. We performed a systematic mapping study of the literature following PRISMA guidelines to explore nine research questions. After a careful selection process, 70 studies were selected, analyzed, and categorized (out of 308). The main results are that: (1) Most studies (71.42%) evaluated syntactic conflicts; (2) Most studies (57.14%) performed statistical analysis to evaluation software merge; (3) Most of the evaluated studies were published in the last five years (2017-2023), evidencing the importance and interest of the academic community and the software industry in this promising area of software engineering. This article benefits professionals and researchers by providing a body of knowledge about the current literature, which can be the starting point for future research. Finally, we present some worth-investigating challenges by the scientific community.