Extensible language-aware merging

Proceedings of the ACM on Programming Languages, 2019

In modern software development, developers rely on version control systems like Git to collaborate in the branch-based development workflow. One downside of this workflow is the conflicts occurred when merging contributions from different developers: these conflicts are tedious and error-prone to be correctly resolved, reducing the efficiency of collaboration and introducing potential bugs. The situation becomes even worse, with the popularity of refactorings in software development and evolution, because current merging tools (usually based on the text or tree structures of source code) are unaware of refactorings. In this paper, we present IntelliMerge, a graph-based refactoring-aware merging algorithm for Java programs. We explicitly enhance this algorithm's ability in detecting and resolving refactoring-related conflicts. Through the evaluation on 1,070 merge scenarios from 10 popular open-source Java projects, we show that IntelliMerge reduces the number of merge conflicts ...

Context: To reduce the integration effort arising from conflicting changes resulting from collaborative software development tasks, unstructured merge tools try to automatically solve part of the conflicts via textual similarity, whereas structured and semistructured merge tools try to go further by exploiting the syntactic structure of the involved artifacts. Objective: In this study, aiming at increasing the existing body of evidence and assessing results for systems developed under an alternative version control paradigm, we replicate an experiment conducted by Apel et al. to compare the unstructured and semistructured approach with respect to the occurrence of conflicts reported by both approaches. Method: We used both semistructured and unstructured merge in a sample 2.5 times bigger than the original study regarding the number of projects and 18 times bigger regarding the number of merge scenarios, and we compared the occurrence of conflicts. Results: Similar to the original study, we observed that semistructured merge reduces the number of conflicts in 55% of the scenarios of the new sample. However, the observed average conflict reduction of 62% in these scenarios is far superior than what has been observed before. We also bring new evidence that the use of semistructured merge can reduce the occurrence of conflicting merge scenarios by half. Conclusions: Our findings reinforce the benefits of exploiting the syntactic structure of the artifacts involved in code integration. Besides, the reductions observed in the number and size of conflicts suggest that the use of semistructured merge,when compared to the unstructured approach, might decrease integration effort without compromising correctness.

… 2012 International Conference on Software …, 2012

Merge conflicts cause software defects which if detected late may require expensive resolution. This is especially true when developers work too long without integrating concurrent changes, which in practice is common as integration generally occurs at check-in. Awareness of others' activities was proposed to help developers detect conflicts earlier. However, it requires developers to detect conflicts by themselves and may overload them with notifications, thus making detection harder. This paper presents a novel solution that continuously merges uncommitted and committed changes to create a background system that is analyzed, compiled, and tested to precisely and accurately detect conflicts on behalf of developers, before check-in. An empirical study confirms that our solution avoids overloading developers and improves early detection of conflicts over existing approaches. Similarly to what happened with continuous compilation, this introduces the case for continuous merging inside the IDE.

2003

IEEE Transactions on Software Engineering

When multiple developers change a software system in parallel, these concurrent changes need to be merged to all appear in the software being developed. Numerous merge techniques have been proposed to support this task, but none of them can fully automate the merge process. Indeed, it has been reported that as much as 10% to 20% of all merge attempts result in a merge conflict, meaning that a developer has to manually complete the merge. To date, we have little insight into the nature of these merge conflicts. What do they look like, in detail? How do developers resolve them? Do any patterns exist that might suggest new merge techniques that could reduce the manual effort? This paper contributes an in-depth study of the merge conflicts found in the histories of 2,731 open source Java projects. Seeded by the manual analysis of the histories of five projects, our automated analysis of all 2,731 projects: (1) characterizes the merge conflicts in terms of number of chunks, size, and programming language constructs involved, (2) classifies the manual resolution strategies that developers use to address these merge conflicts, and (3) analyzes the relationships between various characteristics of the merge conflicts and the chosen resolution strategies. Our results give rise to three primary recommendations for future merge techniques, that-when implemented-could on one hand help in automatically resolving certain types of conflicts and on the other hand provide the developer with tool-based assistance to more easily resolve other types of conflicts that cannot be automatically resolved.

Lecture Notes in Computer Science, 2011

Versioning helps users to keep track of different sets of edits on a document. Version merging methods enable users to determine which parts of which version they wish to include in the next or final version. We explored several existing and two new methods (highlighting and overlay) in single and multiple window settings. We present the results of our quantitative user studies, which show that the new highlighting and overlay techniques are preferred for version merging tasks. The results suggest that the most useful methods are those which clearly and easily present information that is likely important to the user, while simultaneously hiding less important information. Also, multi window version merging is preferred over single window merging.

Malaysian Journal of Computer Science, 2016

Software Configuration Management (SCM) aims to provide a controlling mechanism for the evolution of software artifacts created during software development process. Controlling software artifacts evolution requires many activities to be carried out such as, construction and creation of versions, computation of mappings and differences between versions, merging (i.e. combining of two or more versions) and so on. Traditional SCM systems are file-based SCM systems, which are not adequate for performing software configuration management activities. File-based SCM systems consider software artifacts as a set of text files, while today's software development is model-driven and models are the main artifacts produced in the early phases of software development process. New challenges of model mappings, differencing, merging, and conflict detection arise when applying file-based solution to model-driven software. The goal of this paper is to develop a configuration management solution for model merging and conflict resolution that overcomes the challenges faced by traditional SCM systems for model-based development. We represent models at finegrained level as graph structures, which is an intermediate representation based o n graph t h e o r y. Our approach follows a 3-way model merge process, where a base and its derived versions are used for comparison. To differentiate between conflicted and non-conflicted cases, we have defined different merge cases, and established a merge policy based on merge cases. Merge cases are used along with the comparison result in order to perform conflict resolution and merge operation. We performed a controlled experiment using open source eclipse modeling framework and compare our approach with an open source tool Eclipse Modeling Framework (EMF) Compare. The results proved the accuracy and efficiency of our proposed approach.

Empirical studies show that merge conflicts frequently occur, impairing develop-ers' productivity, since merging conflicting contributions might be a demanding and tedious task. However, the structure of changes that lead to conflicts has not been studied yet. Understanding the underlying structure of conflicts, and the involved syntactic language elements might shed light on how to better avoid merge conflicts. To this end, in this paper we derive a catalog of conflict patterns expressed in terms of the structure of code changes that lead to merge conflicts. We focus on conflicts reported by a semi-structured merge tool that exploits knowledge about the underlying syntax of the artifacts. This way, we avoid analyzing a large number of spurious conflicts often reported by typical line based merge tools. To assess the occurrence of such patterns in different systems, we conduct an empirical study reproducing 70,047 merges from 123 GitHub Java projects. Our results show that most semi-structured merge conflicts in our sample happen because developers independently edit the same or consecutive lines of the same method. However, the probability of creating a merge conflict is approximately the same when editing methods, class fields, and modifier lists. Furthermore, we noticed that most part of conflicting merge scenarios, and merge conflicts, involve more than two developers. Also, that copying and pasting pieces of code, or even entire files, across different repositories is a common practice and cause of conflicts. Finally, we discuss how our results reveal the need for new research studies and suggest potential improvements to tools supporting collaborative software development.

International Journal of Computers and Applications, 2018

Distributed Software Development (DSD) is drawing a lot of attention from the research community due to its increasing significance. However, distance becomes a barrier in communication and coordination in such kind of development. Many tools have been used to help coordination in DSD. One such tool is Software Configuration Management (SCM) which has become an integral part of DSD. Although it coordinates the developers' effort, it has the limitation of not detecting conflicts at an early stage. Therefore, the Advance Version Control (AVC) tool has been designed and implemented to enhance the functionality of the version control system in a distributed environment. It performs versioning of files as well as detects the conflicts at an early stage and informs the concerned developers about conflicting changes. It provides code sharing to mutually resolve the inconsistencies. It also provides a communication medium because, sometimes, communication between conflicting developers can also solve the problem. This paper presents the results from an empirical evaluation of tool AVC. To measure the effectiveness of the tool, a metric suite is designed. Also, two experiments have been performed among the developers to check the efficiency of the tool. This work has been carried out as an extension of the work done in the paper [Kaur I, Singh H. Advanced Version Control (AVC): a paradigm shift from version control to conflict management. IJCA. June 2017;167(10):7-15].

Global software developments intensify parallel changes. Although parallel changes can improve performance, their interferences contribute to faults. Current Software Configuration Management (SCM) systems can detect the interference between changes at textual level. However, our empirical study shows that, compared with textual interference, semantic approach is more effective and efficient in detecting interference in high- degree parallel changes. We propose to integrate semantic interference checking into SCM system. Semantic interferences detected during check in can alert developers to potential faults.

Proceedings of the 3rd international workshop on Software configuration management -, 1991

Software maintenance is the process of designing and integrating consistent changes to an existing software system. It is di cult for the maintainer to ascertain the complete e ect of a code change; the maintainer may make a change to a program that is syntactically and semantically legal, but has ripples into the parts of the program that were to remain unchanged.

2015

Developers typically change codebases in parallel from each other, which results in diverging codebases. Such diverging codebases must be integrated when finished. Integrating diverging codebases involves difficult activities. For example, two changes that are correct independently can introduce subtle bugs when integrated together. Integration can be difficult with existing tools, which, instead of dealing with the evolution of the actual program entities being changed, handle code changes as lines of text in files. Tools are important: software development tools have greatly improved from generic text editors to IDEs by providing high-level code manipulation such as automatic refactorings and code completion. This improvement was possible by the reification of program entities. Nevertheless, integration tools did not benefit from a similar reification of change entities to improve productivity in integration. In this work we first conducted a study to learn which integration activ...

2011

The use of version control has become ubiquitous in software development projects. Version control systems facilitate parallel development and maintenance through branching, the creation of isolated codelines. Merging is a consequence of branching and is the process of integrating codelines. However, there are unanswered questions about the use of version control to support parallel development; in particular, how are branching and merging used in practice? What defines a successful branching and merging strategy? As a first step towards answering these questions, we recruited a diverse sample of 140 version control users to participate in an online survey. In this paper, we present the survey results and 4 key observations about branching and merging practices in software development projects.

2014

The Envision project aims to develop an integrated development environment (IDE) for object-oriented languages that features a visual structured code editor and is used for large-scale software development. To achieve this goal Envision works directly on an abstract syntax tree instead of a text-based source code representation. This thesis features the design of a version control system based on an abstract syntax tree designed for Envision. The resulting system is more fine-grained than traditional text-based systems since changes are tracked on the basis of nodes in the abstract syntax tree and not lines as in a text file. Nevertheless traditional text-based systems contain functionality which can be used as foundation for a version control system for abstract syntax trees. Therefore Envision's version control system is build on top of a Git back-end. The system features a comparison algorithm which is able to detect move operations, a history functionality which is able to track substructures and a merge algorithm which automatically resolves conflicts on list types. The major benefit of a version control system based on an abstract syntax tree is its syntax awareness. As a consequence the comparison algorithm is not prone to formatting changes and the merge algorithm produces always a syntactically correct program. In addition a more fine-grained change categorization in terms of granularity and change types positively affects the quality of the comparison, history and merge algorithms. Further visual improvements in the presentation provide the user with more essential information.

While unstructured merge tools rely only on textual analysis to detect and resolve conflicts, semistructured merge tools go further by partially exploiting the syntactic structure and semantics of the involved artifacts. Previous studies compare these merge approaches with respect to the number of reported conflicts, showing, for most projects and merge situations, reduction in favor of semistructured merge. However, these studies do not investigate whether this reduction actually leads to integration effort reduction (productivity) without negative impact on the correctness of the merging process (quality). To analyze that, and better understand how merge tools could be improved, in this paper we reproduce more than 30,000 merges from 50 open source projects, identifying conflicts incorrectly reported by one approach but not by the other (false positives), and conflicts correctly reported by one approach but missed by the other (false negatives). Our results and complementary analysis indicate that, in the studied sample, the number of false positives is significantly reduced when using semistructured merge. We also find evidence that its false positives are easier to analyze and resolve than those reported by unstructured merge. However, we find no evidence that semistructured merge leads to fewer false negatives, and we argue that they are harder to detect and resolve than unstructured merge false negatives. Driven by these findings, we implement an improved semistructured merge tool that further combines both approaches to reduce the false positives and false negatives of semistructured merge. We find evidence that the improved tool, when compared to unstructured merge in our sample, reduces the number of reported conflicts by half, has no additional false positives, has at least 8% fewer false negatives, and is not prohibitively slower.