Cat Stories

Overview

Cat Stories (internally named “Platform Adopsi Kucing”) was a database-focused project for a university course with a tight 2-week deadline. The system managed cat adoption data — tracking cats, adopters, and adoption records. I deliberately stepped outside my PHP comfort zone and chose Node.js + Express for the backend, making this my first project in the JavaScript backend ecosystem.

Problem

The coursework required a complete database application with REST API, validation, and documentation — in 2 weeks
I had only worked with PHP before this project — the Node.js ecosystem was unfamiliar territory
The team needed API documentation that both frontend and backend developers could reference

Goal

Build a REST API with Express.js and Knex.js query builder for MSSQL
Implement schema validation with Joi and structured logging with Winston
Generate interactive API documentation with Swagger
Build a React frontend with Vite and Tailwind CSS

Role

I led the backend development while my teammate focused on the React frontend.

Team: Achmad Raihan Fahrezi Effendy (Backend Lead) + 1 teammate (Frontend)

Architecture

flowchart TD A["React 18 + Vite"] --> B["Express.js API"] B --> C["Joi Validation"] B --> D["Winston Logging"] B --> E["JWT Auth"] B --> F["Knex.js Query Builder"] F --> G[("Microsoft SQL Server")] H["Swagger"] --> B I["Tailwind CSS"] --> A

Layer	Technology	Purpose
Frontend	React 18, Vite	User interface with fast HMR
Styling	Tailwind CSS	Utility-first responsive design
Backend	Express.js	REST API
Query Builder	Knex.js	Database queries
Validation	Joi	Schema validation
Logging	Winston	Structured logging
Auth	JWT	Token-based authentication
Database	Microsoft SQL Server	Persistent storage
Docs	Swagger (swagger-jsdoc + swagger-ui-express)	Interactive API documentation
Testing	Jest + Supertest	Unit and integration tests
Uploads	multer	File upload handling

Key Features

REST API Architecture — Clean endpoint design with consistent error handling
Schema Validation — Joi validation for every request body and query parameter
Structured Logging — Winston logger for request tracking and debugging
JWT Authentication — Token-based auth for protected endpoints
Interactive API Docs — Swagger UI for testing and documenting every endpoint
File Uploads — Multer for handling multipart form data
SQL Injection Protection — Knex.js parameterized queries against MSSQL

Technical Decisions

Express was chosen to explore Node.js backend development beyond PHP
Knex.js provided a clean query builder without the overhead of a full ORM
Swagger was added early to keep the API contract visible and documented
Joi validation caught errors at the request layer before they reached business logic
Vite was used for fast frontend development with React

Challenges

Pivoting from PHP to Node.js under a 2-week deadline required rapid learning
Knex.js query syntax differs from raw SQL — translating complex queries took adjustment
Integrating Swagger with Express required understanding OpenAPI specification format

Outcome

The project was delivered on time with complete API documentation, validation, and logging. It demonstrated that backend concepts transfer across languages — the patterns I learned in PHP applied directly to Node.js.

Final Thoughts

Cat Stories taught me that the language matters less than the architecture. REST APIs, validation, logging, and documentation work the same way whether you use PHP, Node.js, or anything else. The 2-week deadline forced efficiency, and the Node.js experience opened doors to TypeScript and NestJS in later projects.