arXiv subjects

Abir Bouraffa

Publications and source records attributed to Abir Bouraffa.

5 recordsLinked to original sources

arXiv2025

Not One to Rule Them All: Mining Meaningful Code Review Orders From GitHub

Developers use tools such as GitHub pull requests to review code, discuss proposed changes, and request modifications. While changed files are commonly presented in alphabetical order, this does not necessarily coincide with the reviewer's preferred navigation sequence. This study investigates the different navigation orders developers follow while commenting on changes submitted in pull requests. We mined code review comments from 23,241 pull requests in 100 popular Java and Python repositories on GitHub to analyze the order in which the reviewers commented on the submitted changes. Our analysis shows that for 44.6% of pull requests, the reviewers comment in a non-alphabetical order. Among these pull requests, we identified traces of alternative meaningful orders: 20.6% (2,134) followed a largest-diff-first order, 17.6% (1,827) were commented in the order of the files' similarity to the pull request's title and description, and 29% (1,188) of pull requests containing changes to both production and test files adhered to a test-first order. We also observed that the proportion of reviewed files to total submitted files was significantly higher in non-alphabetically ordered reviews, which also received slightly fewer approvals from reviewers, on average. Our findings highlight the need for additional support during code reviews, particularly for larger pull requests, where reviewers are more likely to adopt complex strategies rather than following a single predefined order.

cs.SE↗

arXiv2025

How Do Developers Use Code Suggestions in Pull Request Reviews?

GitHub introduced the suggestion feature to enable reviewers to explicitly suggest code modifications in pull requests. These suggestions make the reviewers' feedback more actionable for the submitters and represent a valuable knowledge for newcomers. Still, little is known about how code review suggestions are used by developers, what impact they have on pull requests, and how they are influenced by social coding dynamics. To bridge this knowledge gap, we conducted an empirical study on pull requests from 46 engineered GitHub projects, in which developers used code review suggestions. We applied an open coding approach to uncover the types of suggestions and their usage frequency. We also mined pull request characteristics and assessed the impact of using suggestions on merge rate, resolution time, and code complexity. Furthermore, we conducted a survey with contributors of the studied projects to gain insights about the influence of social factors on the usage and acceptance of code review suggestions. We were able to uncover four suggestion types: code style suggestions, improvements, fixes, and documentation with improvements being the most frequent. We found that the use of suggestions positively affects the merge rate of pull requests but significantly increases resolution time without leading to a decrease in code complexity. Our survey results show that suggestions are more likely to be used by reviewers when the submitter is a newcomer. The results also show that developers mostly search suggestions when tracking rationale or looking for code examples. Our work provides insights on the usage of code suggestions and their potential as a knowledge sharing tool.

cs.SE↗

arXiv2024

Explaining Explanation: An Empirical Study on Explanation in Code Reviews

Code reviews are central for software quality assurance. Ideally, reviewers should explain their feedback to enable authors of code changes to understand the feedback and act accordingly. Different developers might need different explanations in different contexts. Therefore, assisting this process first requires understanding the types of explanations reviewers usually provide. The goal of this paper is to study the types of explanations used in code reviews and explore the potential of Large Language Models (LLMs), specifically ChatGPT, in generating these specific types. We extracted 793 code review comments from Gerrit and manually labeled them based on whether they contained a suggestion, an explanation, or both. Our analysis shows that 42% of comments only include suggestions without explanations. We categorized the explanations into seven distinct types including rule or principle, similar examples, and future implications. When measuring their prevalence, we observed that some explanations are used differently by novice and experienced reviewers. Our manual evaluation shows that, when the explanation type is specified, ChatGPT can correctly generate the explanation in 88 out of 90 cases. This foundational work highlights the potential for future automation in code reviews, which can assist developers in sharing and obtaining different types of explanations as needed, thereby reducing back-and-forth communication.

cs.SE↗

arXiv2023

Developers' Visuo-spatial Mental Model and Program Comprehension

Previous works from research and industry have proposed a spatial representation of code in a canvas, arguing that a navigational code space confers developers the freedom to organise elements according to their understanding. By allowing developers to translate logical relatedness into spatial proximity, this code representation could aid in code navigation and comprehension. However, the association between developers' code comprehension and their visuo-spatial mental model of the code is not yet well understood. This mental model is affected on the one hand by the spatial code representation and on the other by the visuo-spatial working memory of developers. We address this knowledge gap by conducting an online experiment with 20 developers following a between-subject design. The control group used a conventional tab-based code visualization, while the experimental group used a code canvas to complete three code comprehension tasks. Furthermore, we measure the participants' visuo-spatial working memory using a Corsi Block test at the end of the tasks. Our results suggest that, overall, neither the spatial representation of code nor the visuo-spatial working memory of developers has a significant impact on comprehension performance. However, we identified significant differences in the time dedicated to different comprehension activities such as navigation, annotation, and UI interactions.

cs.SE↗

arXiv2022

Beyond Duplicates: Towards Understanding and Predicting Link Types in Issue Tracking Systems

Software projects use Issue Tracking Systems (ITS) like JIRA to track issues and organize the workflows around them. Issues are often inter-connected via different links such as the default JIRA link types Duplicate, Relate, Block, or Subtask. While previous research has mostly focused on analyzing and predicting duplication links, this work aims at understanding the various other link types, their prevalence, and characteristics towards a more reliable link type prediction. For this, we studied 607,208 links connecting 698,790 issues in 15 public JIRA repositories. Besides the default types, the custom types Depend, Incorporate, Split, and Cause were also common. We manually grouped all 75 link types used in the repositories into five general categories: General Relation, Duplication, Composition, Temporal / Causal, and Workflow. Comparing the structures of the corresponding graphs, we observed several trends. For instance, Duplication links tend to represent simpler issue graphs often with two components and Composition links present the highest amount of hierarchical tree structures (97.7%). Surprisingly, General Relation links have a significantly higher transitivity score than Duplication and Temporal / Causal links. Motivated by the differences between the link types and by their popularity, we evaluated the robustness of two state-of-the-art duplicate detection approaches from the literature on the JIRA dataset. We found that current deep-learning approaches confuse between Duplication and other links in almost all repositories. On average, the classification accuracy dropped by 6% for one approach and 12% for the other. Extending the training sets with other link types seems to partly solve this issue. We discuss our findings and their implications for research and practice.

cs.SE↗