An Event-Driven Tool for Context-Aware Code Smell Detection Using SmellDSL Journal: Science of Computer Programming

Matheus dos Santos Viegas, Adrian Gabriel Keller dos Santos, Kleinner Farias, Robson Keemps da Silva · 2026

Read An Event-Driven Tool for Context-Aware Code Smell Detection Using SmellDSL Journal: Science of Computer Programming by Matheus dos Santos Viegas, Adrian Gabriel Keller dos Santos, Kleinner Farias, Robson Keemps da Silva (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-20675.txt.

Introduction

Matheus dos Santos Viegas and three colleagues present SmellHunter, a context-aware tool for detecting code smells, design flaws that degrade software quality. Most detectors use static metrics alone and ignore the setting in which flaws arise and get fixed. SmellHunter reads scripts in SmellDSL, a domain language for smell rules, and mixes metrics with context such as team traits and project stage. An event-driven service bus runs validation, interpretation, and storage as separate services.

Catalog record

CREATORMatheus dos Santos Viegas, Adrian Gabriel Keller dos Santos, Kleinner Farias, Robson Keemps da Silva
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE29356
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2605.20675

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

An Event-Driven Tool for Context-Aware Code Smell Detection Using SmellDSL Journal: Science of Computer Programming Matheus dos Santos Viegas, Adrian Gabriel Keller dos Santos, Kleinner Farias, Robson Keemps da Silva Abstract Code smells signal violations of design principles that degrade the internal quality of evolving software systems. Although many tools detect such anomalies using static metrics, they often ignore the development context in which smells arise and are resolved. This limitation can lead to misleading warnings and weak support for refactoring decisions. To address this problem, we present SmellHunter , a context-aware tool that interprets scripts written in the SmellDSL domain-specific language to detect and contextualize code smells. SmellHunter integrates static code metrics with contextual information (such as team characteristics, project stage, and geographic metadata) to produce richer, more actionable analyses. The tool adopts an event-driven architecture in which a service bus orchestrates validation, interpretation, and persistence services through asynchronous events. This architecture enables scalable analysis while minimizing disruption to developers’ workflows. SmellHunter is integrated into the Eclipse development environment via a dedicated plugin and provides aggregated insights via a mobile application, allowing developers to explore smell occurrences by type, severity, and location. By linking smell detection with contextual data and collaborative visualization, SmellHunter supports developers acting as smell hunters , helping teams identify recurring quality issues emerging from a particular location and assign refactoring tasks to developers with relevant expertise. We describe the architecture of SmellHunter , the interpretation process of SmellDSL scripts, and the integration of contextual data to support more informed refactoring decisions in modern software development environments. Metadata The metadata for the proposed tool are shown in Table 1 . Nr. Code metadata description Please fill in this column C1 Current code version V1.0 C2 Permanent link to code/repository used for this code version https://github.com/MathDEV-0/SmellHunterAPI https://github.com/MathDEV-0/SmellHunter-Eclipse-Plugin C3 Permanent link to Reproducible Capsule C4 Legal Code License MIT License C5 Code versioning system used Git C6 Software code languages, tools, and services used Python, Flask (interpreter); Java, Eclipse RCP (plugin) C7 Compilation requirements, operating environments and dependencies Python 3.10+; Java 21+, Eclipse RCP C8 If available, link to developer documentation/manual SmellHunter API: https://github.com/MathDEV-0/SmellHunterAPI/blob/main/README.md SmellHunter Eclipse Plugin: https://github.com/MathDEV-0/SmellHunter-Eclipse-Plugin/blob/main/README.md Demonstration: https://www.youtube.com/watch?v=WKONlb5o1TY C9 Support email for questions [email protected] Table 1: Code metadata. 1 Motivation and significance Code smells are symptoms of poor design decisions in source code and can impair software maintainability [ 15 , 16 ] . Over time, these smells degrade design quality, increase maintenance costs, and may compromise architectural stability [ 7 , 4 , 14 ] . Moreover, code smells can significantly increase change-proneness [ 15 ] . Code smells are therefore widely regarded as early indicators of design problems that require refactoring [ 22 ] to prevent further deterioration of the software structure [ 7 ] . Numerous tools have been proposed to support smell detection, including SonarQube 1 1 1 SonarQube: https://www.sonarsource.com/products/sonarqube/ , Qodana 2 2 2 Qodana: https://www.jetbrains.com/qodana/ , and RefactoringMiner [ 9 , 22 ] . While these tools provide robust static analysis capabilities, refactoring heuristics, and rely on predefined metrics and rules, they often treat smells as universal violations of coding conventions. In practice, however, not all smells carry the same weight for every team [ 12 , 10 ] . Development teams often perceive certain smells as more critical [ 10 ] because they threaten architectural stability or violate local coding practices shaped by project constraints and organizational culture [ 12 ] . More critically, prior studies [ 17 , 10 ] reveal a marked lack of consensus among developers on how to detect even well-established code smells [ 21 ] , underscoring the inherently subjective nature of this process. SmellHunter addresses this concern by supporting the detection of team-sensitive smells . SmellHunter allows teams to specify priorities through SmellDSL scripts [ 12 ] and associate smell occurrences with contextual attributes, including project stage and geolocation. This functionality enables developers to identify where smells emerge and are resolved, allowing teams to map quality issues to specific regions, organizations, or development environments and to reveal locations where targeted improvements may be required. Prior empirical studies [ 6 , 5 ] have demonstrated the importance of incorporating contextual information in software development practices. This work addresses a scientific and technical problem: the lack of tools to interpret executable specifications of smell-detection rules, expressed in SmellDSL, in a context-aware development environment. SmellDSL is a domain-specific language to specific bad smells [ 12 ] . Existing approaches allow developers to define smell detection rules, but they rarely provide mechanisms to dynamically interpret these rules or integrate them with contextual signals that influence development decisions [ 1 , 12 ] . As a result, many existing tools remain limited to static rule evaluation and often fail to capture the broader context in which smells emerge. SmellHunter closes this gap by interpreting scripts written in the SmellDSL language and generating automated analyses of code anomalies enriched with contextual metadata. Developers receive feedback directly within the Eclipse development environment and through a complementary mobile application. This integration transforms smell detection from a static rule-based activity into an interactive, operational process. The scientific significance lies in enabling executable smell specifications that can be validated, interpreted, and acted through automated analysis pipelines that align technical metrics with the real constraints and priorities of development teams. Our work also opens new opportunities for scientific investigation in context-aware code quality analysis. Traditional smell detection mechanisms typically operate in batch, analyzing code snapshots with static metrics [ 9 ] . Such approaches rarely capture the dynamic and contextual nature of software development activities. SmellHunter reframes smell detection as an asynchronous, event-driven process that integrates contextual information into the analysis pipeline. Each smell-detection request becomes an event processed by a service bus that coordinates validation, interpretation, and persistence services. Our proposed architecture (Section 2.1 ) enables large-scale analysis while preserving the responsiveness and traceability of smell detection activities. Furthermore, by associating smell occurrences with contextual and geographical metadata, SmellHunter introduces the concept of location-aware quality analysis . Developers may specialize in detecting and resolving specific smell categories and act as “smell hunters” responsible for particular contexts or regions. Researchers can therefore investigate spatial and organizational patterns of code degradation and study how team practices influence the emergence and resolution of smells. From an operational perspective, our work supports development teams engaged in continuous software evolution. Teams define smell specifications using SmellDSL script and configure thresholds for relevant metrics based on their development practices and quality priorities . Through the Eclipse plugin, developers submit analysis requests that trigger asynchronous processing in the SmellHunter backend. A validation service verifies the script syntax and threshold definitions, while the interpretation service evaluates the rules against project artifacts. Detected smells are stored and disseminated through client interfaces . Developers receive notifications in the IDE and through a mobile interface that summarizes smell occurrences by severity, type, and location. This workflow enables teams to detect and address code smells while maintaining software in parallel, reducing disruption to ongoing development and supporting continuous maintenance. Considering the implementation of SmellHunter , it was built upon widely adopted technologies and frameworks that ensure scalability and interoperability with modern development infrastructures. The tool

Read the complete text