📚 CodexLibris – REST API for a Digital Library

CodexLibris is a RESTful API developed with Spring Boot, PostgreSQL, and JWT authentication, designed to manage a digital library of books, authors, genres, and users. The project models real-world library features like book availability, role-based access control, and personalized recommendations via a local LLM (Large Language Model).

Everything runs in Docker containers, making the setup and deployment seamless.

---

🧩 Tech Stack

• Java + Spring Boot (API, security, business logic)
• PostgreSQL (relational data persistence)
• JWT (token-based authentication)
• Docker + Docker Compose (containerized infrastructure)
• OpenAPI/Swagger (interactive API docs)
• Optional: Local LLM integration for book recommendations (Python + Transformers)

---

🎯 Features

• User management with role-based access (ADMIN, USER)
• CRUD operations for:
  • Books
  • Authors
  • Genres
  • Loads
  • Events
  • Users
  • Roles
• JWT-based login and secure endpoints
• Swagger UI for API exploration
• HTTPS enabled with self-signed cert
• Bonus: AI-powered recommendations with local LLM

---

🚀 Getting Started

🔧 1. Build Docker Containers

docker-compose build --no-cache

▶️ 2. Run the App

docker-compose up -d

✅ 3. Check Running Containers

docker ps

---

🔐 HTTPS Notes

• Access the API at: https://localhost
• Because of the self-signed certificate:
  • Use -k with curl to bypass SSL verification
  • In the browser, manually accept the certificate

---

🧪 API Documentation (Swagger)

https://localhost/swagger-ui/index.html

---

🔐 Authentication

Use the /auth/login endpoint to obtain a JWT token.

curl -X POST https://localhost/auth/login \
  -H "Content-Type: application/json" \
  -d '{"username": "admin", "password": "admin"}'

Example response:

{
  "token": "eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9...",
  "username": "admin",
  "roleId": 1
}

---

📚 API Endpoints

Users

GET /user/{id}
GET /users

Books

GET /books
GET /books/{id}
POST /books
PUT /books/{id}

Authors

GET /authors
GET /authors/{id}
POST /authors
PUT /authors/{id}

Genres

GET /genres
GET /genres/{id}
POST /genres
PUT /genres/{id}

All endpoints require Authorization header:

-H "Authorization: Bearer YOUR_JWT_TOKEN"

---

🤖 Optional: AI Book Recommendations (LLM)

This API includes an optional LLM-powered recommendation system, run locally via Docker.

Setup LLM

cd llm
docker-compose up --build

Use the Recommender Endpoint

curl -k -X POST https://localhost/api/llm/recomender \
  -H 'Authorization: Bearer YOUR_JWT_TOKEN' \
  -H 'Content-Type: application/json' \
  -d '{"text": "Harry Potter"}'

Example response:

{
  "results": [
    {
      "title": "Harry Potter and the Philosopher's Stone",
      "author": "J. K. Rowling",
      "year": 1997
    }
  ]
}

---

💡 Notes

• This project is part of a personal exploration of robust REST architecture and full-stack Java development.
• It serves as a base for testing different frontends, such as Vaadin or Angular, using the same backend.
• It’s also a playground for integrating lightweight AI systems into backend services.

---

🧑‍💻 Author

Jessica GitHub