Mention API Documentation

Overview

Mention API is a robust backend service built with Express.js and TypeScript, providing functionality for social media interactions including posts, user management, authentication, and real-time communications.

Tech Stack

• Node.js with TypeScript
• Express.js for REST API
• MongoDB with Mongoose for data storage
• Socket.IO for real-time features
• JWT for authentication

Getting Started

Prerequisites

• Node.js (v14 or higher)
• MongoDB instance
• npm or yarn package manager

Installation

1. Clone the repository
2. Install dependencies:

npm install

3. Create a .env file in the root directory with the following variables:

MONGODB_URI=your_mongodb_connection_string
ACCESS_TOKEN_SECRET=your_jwt_access_token_secret
REFRESH_TOKEN_SECRET=your_jwt_refresh_token_secret
PORT=3000

Running the API

Development mode:

npm run dev

Production mode:

npm run build
npm start

Deployment

Docker Deployment

# Build the Docker image
docker build -t mention-api .

# Run the container
docker run -p 3000:3000 -e MONGODB_URI=your_mongodb_uri mention-api

Cloud Deployment (Vercel)

1. Configure vercel.json:

{
  "version": 2,
  "builds": [{
    "src": "dist/server.js",
    "use": "@vercel/node"
  }],
  "routes": [{
    "src": "/(.*)",
    "dest": "dist/server.js"
  }]
}

2. Deploy using Vercel CLI:

vercel --prod

API Endpoints

Authentication Flow

sequenceDiagram
    participant Client
    participant API
    participant Database
    
    Client->>API: POST /auth/signup
    API->>Database: Create User
    Database-->>API: User Created
    API-->>Client: Access Token
    
    Client->>API: POST /auth/login
    API->>Database: Verify Credentials
    Database-->>API: User Found
    API-->>Client: Access + Refresh Tokens

Loading

Authentication

POST /auth/signup

• Creates a new user account
• Body: { username: string, email: string, password: string }
• Returns:

{
  "user": {
    "id": "user_id",
    "username": "john_doe",
    "email": "[email protected]",
    "createdAt": "2023-01-01T00:00:00Z"
  },
  "accessToken": "eyJhbGciOiJIUzI1NiIs..."
}

POST /auth/login

• Authenticates existing user
• Body: { email: string, password: string }
• Returns:

{
  "accessToken": "eyJhbGciOiJIUzI1NiIs...",
  "refreshToken": "eyJhbGciOiJIUzI1NiIs..."
}

Posts

POST /posts

• Creates a new post
• Authentication: Bearer token required
• Body:

{
  "text": "Hello world!",
  "media": ["image1.jpg"],
  "location": {
    "longitude": -73.935242,
    "latitude": 40.730610
  }
}

• Returns:

{
  "id": "post_id",
  "text": "Hello world!",
  "media": ["image1.jpg"],
  "location": {
    "type": "Point",
    "coordinates": [-73.935242, 40.730610]
  },
  "createdAt": "2023-01-01T00:00:00Z",
  "user": {
    "id": "user_id",
    "username": "john_doe"
  }
}

GET /posts/explore

• Retrieves posts for exploration
• Query params: limit (default: 20), offset (default: 0)
• Returns:

{
  "posts": [
    {
      "id": "post_id",
      "text": "Post content",
      "user": {
        "id": "user_id",
        "username": "john_doe"
      },
      "createdAt": "2023-01-01T00:00:00Z"
    }
  ],
  "total": 100,
  "hasMore": true
}

Users

GET /users/:username

• Retrieves user profile
• Authentication: Bearer token required
• Returns:

{
  "id": "user_id",
  "username": "john_doe",
  "bio": "Hello, I'm John!",
  "followersCount": 1000,
  "followingCount": 500,
  "postsCount": 100,
  "isFollowing": true
}

Database Schema Relationships

erDiagram
    User ||--o{ Post : creates
    User ||--o{ Bookmark : has
    User ||--o{ Follow : follows
    Post ||--o{ Like : receives
    Post ||--o{ Comment : has
    
    User {
        string id PK
        string username
        string email
        string password
        string[] bookmarks
        datetime createdAt
    }
    
    Post {
        string id PK
        string userId FK
        string text
        string[] media
        object location
        datetime createdAt
    }

Loading

Performance Optimization

Caching Strategy

• Implement Redis caching for:
  • User profiles (TTL: 1 hour)
  • Popular posts (TTL: 15 minutes)
  • Trending hashtags (TTL: 5 minutes)

Database Indexing

// User Collection Indexes
db.users.createIndex({ "username": 1 }, { unique: true })
db.users.createIndex({ "email": 1 }, { unique: true })

// Post Collection Indexes
db.posts.createIndex({ "userId": 1, "createdAt": -1 })
db.posts.createIndex({ "location": "2dsphere" })

Monitoring and Logging

Health Check Endpoint

GET /health
Response: {
  "status": "healthy",
  "version": "1.0.0",
  "uptime": 1000,
  "mongoStatus": "connected"
}

Logging

• Use Winston for structured logging
• Log levels: error, warn, info, debug
• Include request ID in all logs

Troubleshooting Guide

Common Issues

1. Connection Timeouts

Error: MongoTimeoutError
Solution: Check MongoDB connection string and network connectivity

2. Authentication Failures

Error: JsonWebTokenError
Solution: Verify token expiration and secret keys

3. Rate Limit Exceeded

Error: 429 Too Many Requests
Solution: Implement exponential backoff in client

Contributing

1. Fork the repository
2. Create your feature branch
3. Commit your changes
4. Push to the branch
5. Create a new Pull Request

License

AGPL License