Inventory Management System

A high-performance, distributed Inventory Management System developed during my internship (PKL). This system is designed to handle heavy workloads through asynchronous processing, centralized object storage, and efficient caching strategies.

---

🌟 Key Features

• Automated Image Processing: High-quality image compression and resizing using Sharp to save storage costs without losing clarity.
• Message Queuing: Offloading intensive tasks (like image processing and reporting) to RabbitMQ background workers.
• S3-Compatible Storage: Integration with MinIO for scalable and secure product asset management.
• Distributed Caching: Redis implementation for session management and frequently accessed data to minimize database latency.
• Dockerized Environment: Fully containerized setup for consistent development and seamless production deployment.
• Security: CSRF protection, session encryption, and secure environment variable management.

---

🏗 System Architecture

The application follows a Producer-Consumer pattern to ensure the main API remains responsive.

1. Producer (Web Server): Receives product data and uploads images.
2. Queue (RabbitMQ): Stores image processing tasks in a durable queue.
3. Consumer (Worker): Picks up tasks, processes images via Sharp, and uploads the final results to MinIO.
4. Database (MongoDB): Stores structured metadata and stock information.

---

🛠 Tech Stack

• Runtime: Node.js
• Framework: Express.js
• Template Engine: EJS (Embedded JavaScript)
• Database: MongoDB (Mongoose ODM)
• Cache & Store: Redis
• Message Broker: RabbitMQ
• Object Storage: MinIO
• Utilities: Sharp (Image Processing), Tesseract.js (Planned/OCR)

---

🚦 Getting Started

Prerequisites

• Docker Desktop
• Git

---

Installation

1. Clone the repository:

   https://github.com/F1D02310092/pkl_inventaris.git
   npm install

2. Configure Environment Variables:

   Create a .env file in the root directory:

   # Server Configuration
   PORT=3000
   NODE_ENV=production
   BASE_URL=http://localhost:3000
   
   # Database (MongoDB)
   MONGO_URI=mongodb://admin:password@mongodb:27017/inventory?authSource=admin
   
   # Message Broker (RabbitMQ)
   RABBITMQ_URI=amqp://guest:guest@rabbitmq:5672
   
   # Distributed Cache (Redis)
   REDIS_URI=redis://redis:6379
   
   # Object Storage (MinIO)
   MINIO_ENDPOINT=minio
   MINIO_PORT=9000
   MINIO_ACCESS_KEY=your-key
   MINIO_SECRET_KEY=your-key
   
   # Security
   SESSION_SECRET=your_very_secure_random_string_here

3. Launch with Docker Compose:

   docker-compose up -d

---

📁 Project Structure

├── src/
│   ├── app.js          # Main entry point
│   ├── config/         # Database, Redis, MinIO, and RabbitMQ configs
│   ├── controllers/    # Business logic for each route
│   ├── helper/ 
│   ├── models/         # Mongoose schemas (Items, Categories, Users)
│   ├── middleware/     # Authentication & error handling
│   ├── public/         # Static assets (CSS, JS)
│   ├── security/       # Security implementations 
│   └── views/          # EJS templates for the dashboard
│
├── workers/        # RabbitMQ consumers for Sharp processing
├── docker-compose.yml
├── Dockerfile
└── README.md

---

🛡 Security Practices

• Environment Isolation: Credentials are never hardcoded and are managed via .env.
• Statelessness: Files are stored in MinIO, not on the server's local disk, allowing for horizontal scaling.
• Input Validation: All incoming data is sanitized and validated before hitting the database.