🎬 HallyuHub – K-Drama & Movie Tracker (Microservices)

A cloud-native, microservices-based K-Drama and Movie tracking web application built with modern technologies and deployed on Azure Container Apps.

🌐 Live Site: HallyuHub on Azure

---

📌 Overview

HallyuHub allows users to:

• 📋 Browse a catalog of popular K-Dramas and movies
• ➕ Add titles to a personal watchlist with a rating
• ❌ Remove titles from the watchlist
• 🤖 Get AI-style recommendations based on genres from the user's watchlist

---

🏗️ Architecture

The application follows a microservices architecture with four independent services, each containerized with Docker and deployed to Azure Container Apps.

┌─────────────────────────────────────────────────┐
│                   Frontend (React)               │
│         Vite + TypeScript + Tailwind CSS         │
└──────────────┬──────────────────────────────────┘
               │  REST API Calls
  ┌────────────┼────────────────────┐
  ▼            ▼                    ▼
┌──────────┐ ┌──────────────┐ ┌───────────────────┐
│ Catalog  │ │  Watchlist   │ │   Recommender     │
│ Service  │ │   Service    │ │     Service       │
│  (Java / │ │ (Node.js /   │ │  (Python / Flask) │
│  Spring) │ │  Express.js) │ │                   │
└──────────┘ └──────────────┘ └───────────────────┘

---

🧩 Services

1. 🗂️ Catalog Service — Java / Spring Boot

• Provides a list of K-Dramas and movies with metadata (title, genre, release year, rating)
• Endpoint: GET /api/catalog
• Deployed at: https://catalog-app.icyhill-d8a50826.southeastasia.azurecontainerapps.io

2. 📝 Watchlist Service — Node.js / Express.js / TypeScript

• Manages a user's watchlist using an in-memory store
• Endpoints:
  • GET /api/watchlist/:userId – Get user's watchlist
  • POST /api/watchlist – Add a title to the watchlist
  • DELETE /api/watchlist/:userId/:catalogId – Remove a title from the watchlist
• Deployed at: https://watchlist-app.icyhill-d8a50826.southeastasia.azurecontainerapps.io

3. 🤖 Recommender Service — Python / Flask

• Fetches the user's watchlist and catalog, then recommends unwatched titles based on matching genres
• Endpoint: GET /api/recommendations/:userId
• Deployed at: https://recommender-app.icyhill-d8a50826.southeastasia.azurecontainerapps.io

4. 🖥️ Frontend — React / Vite / TypeScript / Tailwind CSS

• Single-page application (SPA) that consumes all three backend services
• Deployed at: https://frontend-app.icyhill-d8a50826.southeastasia.azurecontainerapps.io

---

🛠️ Tech Stack

🖥️ Frontend

⚙️ Backend Services

🐳 DevOps & Cloud

---

🚀 Running Locally

Prerequisites

• Docker Desktop installed and running

Clone the repository

git clone https://github.com/hasinduudara/HallyuHub-Microservices.git
cd HallyuHub-Microservices

Run each service individually

Catalog Service (Java / Spring Boot)

cd catalog-service
./mvnw spring-boot:run

Watchlist Service (Node.js / TypeScript)

cd watchlist-service
npm install
npm run dev

Recommender Service (Python / Flask)

cd recommender-service
pip install -r requirements.txt
python app.py

Frontend (React / Vite)

cd frontend
npm install
npm run dev

---

📁 Project Structure

HallyuHub-Microservices/
├── .github/
│   └── workflows/
│       └── docker-publish.yml  # CI/CD – Build & Push to Docker Hub
├── catalog-service/        # Spring Boot – Catalog API
├── watchlist-service/      # Node.js/Express – Watchlist API
├── recommender-service/    # Python/Flask – Recommender API
└── frontend/               # React/Vite – Web UI

---

⚙️ CI/CD Pipeline

This project uses GitHub Actions to automate the build and deployment pipeline.

Workflow: .github/workflows/docker-publish.yml

Trigger: Every push to the main or master branch automatically:

1. ✅ Checks out the source code
2. 🔐 Logs in to Docker Hub using GitHub Secrets (DOCKER_USERNAME, DOCKER_PASSWORD)
3. 🏗️ Builds Docker images for all four services
4. 🚀 Pushes images to Docker Hub with the latest tag

Service	Docker Hub Image
Catalog Service	<DOCKER_USERNAME>/catalog-service:latest
Watchlist Service	<DOCKER_USERNAME>/watchlist-service:latest
Recommender Service	<DOCKER_USERNAME>/recommender-service:latest
Frontend	<DOCKER_USERNAME>/frontend:latest

Required GitHub Secrets

Secret	Description
DOCKER_USERNAME	Your Docker Hub username
DOCKER_PASSWORD	Your Docker Hub access token

---

📬 Contact

Platform	Link
📧 Email	hasiduudara@gmail.com
💼 LinkedIn	linkedin.com/in/hasindu-udara
🐙 GitHub	github.com/hasinduudara
🌐 Portfolio	portfolio

---

Built with ❤️ by Hasindu Udara