LEETCODE CLONE - Microservices Architecture

Overview

This project is a Leetcode-like application that allows users to submit coding solutions to problems, which are then evaluated and returned with results in real-time. The architecture is built on microservices, each responsible for a specific function in the workflow.

The services are designed to communicate through message queues to handle submissions, evaluations, and real-time feedback, ensuring a seamless user experience. The microservices involved are:

1. Problem Service: Manages and provides problem data. (CRUD problems)
2. Submission Service: Handles user submissions and coordinates processing.
3. Evaluation Service: Evaluates the submitted code in a secure environment.
4. Socket Service: Provides real-time updates to the frontend.
5. Frontend: The user interface where users interact with the platform.

Architecture and Workflow

Microservices Overview

1. Problem Service

• Function: Stores and serves coding problems and their metadata.
• Endpoints:
  • GET /api/v1/problems/: Retrieves a list of problems.
  • GET /api.v1/problems/:id: Retrieves details of a specific problem.
  • POST /api/v1/problems/: Adds a problem.
  • PATCH /api.v1/problems/:id: Updates details of a specific problem.
  • DELETE /api.v1/problems/:id: Deletes a specific problem.

2. Submission Service

• Function: Manages user submissions and coordinates processing.
• Endpoints:
  • POST /api/v1/submissions: creates a code submission.
• Queues:
  • submissionQueue: Sends submissions to the Evaluation Service.
  • evaluationQueue: Receives evaluation results and forwards them to the Socket Service.

3. Evaluation Service

• Function: Securely evaluates user-submitted code in a Docker container.
• Process:
  • Picks up submissions from the submissionQueue.
  • Runs the code against test cases.
  • Publishes results to the evaluationQueue.

4. Socket Service

• Function: Provides real-time updates to the client.
• Process:
  • Listens to the evaluationQueue for completed evaluations.
  • Uses WebSockets to push results to the client instantly.

5. Frontend

• Function: Provides an interactive interface for users to view problems, submit solutions, and see results.
• Technologies: React, WebSockets for real-time communication.

Installation and Setup

1. Clone the repository:

git clone https://github.com/HarshitBamotra/Leetcode-Project.git

2. Change directory

cd Leetcode-Project

3. Install Dependencies

cd leetcode-submission-service
npm i

cd ../leetcode-evaluator-service
npm i

cd ../leetcode-problem-service
npm i

cd ../leetcode-socket-service
npm i

cd ../leetcode-frontend
npm i

4. Set up environment variables

for leetcode-submission-service, leetcode-problem-service, and leetcode-evaluator-service create .env file in each directory

Sample .env file for submission service

PORT = 3000

REDIS_PORT = 6379
REDIS_HOST = '127.0.0.1'

DB_URL = your mongodb url

PROBLEM_SERVICE_URL = "http://localhost:3001"

Sample .env file for evaluator service

PORT = 3002
REDIS_PORT = 6379
REDIS_HOST = '127.0.0.1'
DOCKER_SOCKET_PATH = Your Docker Context (run docker context ls in terminal to find your path. eg: "/home/twobraincells/.docker/desktop/docker.sock") 

Sample .env file for problem service

PORT = 3001
DB_URL = MongoDB URL
LOG_DB_URL = MongoDB URL

5. Install docker desktop from https://www.docker.com/products/docker-desktop/

6. Install Redis: https://redis.io/docs/latest/operate/oss_and_stack/install/install-redis/

7. Start the Application

cd leetcode-submission-service
npm start

cd ../leetcode-evaluator-service
npm start

cd ../leetcode-problem-service
npm start

cd ../leetcode-socket-service
npm start

cd ../leetcode-frontend
npm run dev

8. Access the Application at localhost:5173/

PROJECT SCREENSHOTS

Problem List Page

Problem Description Page

Problem Submission