Nest Chat API

A backend real-time chat API built with NestJS, designed around WebSocket-based communication, session-based authentication, and a MongoDB-backed data layer.
The project focuses on real-time connections, authentication flows, testing, and production-style backend structure.

---

Overview

This project demonstrates how to build a real-time chat backend with:

• WebSocket-based connections for real-time messaging
• Session-based authentication
• OAuth authentication using Google and Twitter
• MongoDB for persisting users, sessions, and chat-related data
• Structured logging
• Full test coverage using Jest
• Dockerized NoSQL database for local development

The focus is on connection handling, authentication, and backend correctness, rather than frontend concerns.

---

Repository Separation

This repository contains only the backend service for the chat application.

• The backend (this repository) is responsible for:
  • Real-time WebSocket communication
  • Authentication and session management
  • Data persistence and business logic
• The frontend is implemented in a separate repository
  • Communicates with this backend via WebSockets and HTTP APIs
  • Handles UI, client-side state, and user interactions

This separation allows each part of the system to evolve independently and reflects real-world production architectures.

---

Tech Stack

Backend

• NestJS
• WebSockets
• MongoDB
• Session-based authentication
• OAuth (Google, Twitter)
• Structured logging with winston
• Testing with Jest

Infrastructure

• Docker
• Docker Compose

---

Architecture Overview

• The backend exposes a WebSocket server for handling real-time client connections.
• Clients connect over WebSockets and maintain persistent sessions.
• Authentication is handled using:
  • Traditional credential-based login with sessions
  • OAuth login via Google and Twitter
• Session data and user state are persisted in MongoDB.
• Logging is implemented to track connections, authentication events, and system errors.
• The system is designed to be modular and testable.

---

Real-Time Communication

• Real-time messaging is handled using WebSockets.
• Each client maintains a persistent connection to the server.
• Connection lifecycle events (connect, disconnect) are explicitly handled.
• Messages are routed through WebSocket gateways instead of HTTP endpoints.

---

Authentication Methods

• Session-based authentication
• OAuth authentication
  • Google
  • Twitter

Session Handling

• Sessions are persisted in MongoDB
• Authenticated users maintain state across WebSocket connections
• Authorization logic is enforced on protected WebSocket events

---

Database Design (MongoDB)

MongoDB is used to store:

• User accounts
• Authentication and session data
• Chat-related entities (e.g. conversations, messages)

---

Testing

• The project includes full test coverage using Jest
• Tests cover:
  • Authentication flows
  • WebSocket handlers
  • Core business logic
• The codebase is structured to allow isolated and repeatable tests

---

Logging

• Structured logging is implemented across the application
• Logs capture:
  • Connection lifecycle events
  • Authentication attempts
  • Errors and exceptional cases

---

Environment Configuration

The application uses a local .env file for configuration.

The .env file is used to configure:

• Database connection
• Session settings
• OAuth credentials
• Application ports

Environment files are created locally and are not committed to version control.

---

Running the Project Locally

Start MongoDB (Docker)

docker-compose up --build

Run the backend

yarn install
yarn start:dev