How alstn113/fluffy works
Based on our analysis of the file tree and README, this project is an online exam platform (“플러피”) that enables users to create, manage, and take exams online. It’s implemented as a full-stack application with a clear separation between a backend server (built with Kotlin and Spring Boot) and a frontend web application (built with React using Vite, Tailwind CSS, and related libraries).
Project’s Main Components
• Backend Server: This component is found under the “server” directory. It includes:
– API controllers grouped by functionalities (e.g., authentication, exam management, comments, questions, submissions, reactions, OAuth2 integration, etc.).
– Domain, application, infrastructure, and UI (controller) layers following a layered, possibly Hexagonal or Domain-Driven design.
– Supporting infrastructure: persistence via PostgreSQL (with Flyway for migration), caching/locking with Redis/Redisson, AWS S3 for storage, and integration with external OAuth2 providers (GitHub, Google).
– Build and deployment scripts (Dockerfile, Gradle scripts, Docker Compose files) that are used within CI/CD workflows defined in GitHub Actions.
• Frontend Web Application: Located under the “web” directory. It is a React application featuring:
– Component directories for dashboards, exams, questions, analytics, submission views, etc.
– API client modules that call backend endpoints.
– State management through Zustand and data fetching using Tanstack Query.
– Deployment configurations for Vercel and build configurations via Vite.
• External Services and Infrastructure: Note integrations with AWS (EC2, RDS, ElastiCache, S3, CloudFront), OAuth2 providers for authentication, and possibly other third-party tools.
Relationships and Interactions
• End Users (clients) interact with the web frontend.
• The React-based frontend communicates with the backend via REST API calls.
• The backend exposes well-structured endpoints (documented in the README link to the API documentation) and orchestrates business logic.
• The service layers within the backend coordinate with the database (PostgreSQL) and cache/storage subsystems (Redis, AWS S3).
• OAuth2 authentication integration involves flows where the backend interacts with external OAuth2 providers (for GitHub and Google).
• CI/CD systems (Github Actions, Docker, and Gradle builds) automate build, test, and deployment pipelines.
Architectural Patterns and Principles
• Layered Architecture: The backend neatly separates domain logic, application services, and controllers (UI layer), which is evident from the package structure.
• Domain-Driven Design: The presence of “domain,” “application,” and “infra” packages suggests encapsulation of business rules and data integrity.
• RESTful Architecture: Communication between frontend and backend is based on REST APIs.
• Deployment and Containerization: The Dockerfile, Docker Compose files, and GitHub workflows signify container-based builds and deployments.
• Integration Patterns: The project uses REST Docs for documentation and Testcontainers for integration testing, ensuring reliable system behavior.