Stack Underflow | Backend

An attempt at building a Backend for Stackoverflow clone using type-graphql, mongoose, and others.

How to run this program?

You can choose either one of the below setup procedures:

Codesandbox Setup

• Step 1:

  

  Wait for VM bootup process....

• Step 2: Create a local branch on your codesandbox account.

• Step 3:
  Run yarn server start:dev

Local setup

Please Note: I have already loaded .env.dev file with required variables, But be sure to check packages/server/env.example and add required .env.dev or .env.test files in packages/server directory if something goes wrong. Run the commands in below order from project root directory.

# Install Dependencies
yarn install

# Check Environment variables

# Start MongoDB
yarn server start:mongo # check `packages/server/docker-compose.yml` for more information

# Build App
yarn server build

# Start Development Server
yarn server start:dev # development server runs with `.env.dev` variables
# or
yarn server start:test # test server runs with `.env.test` variables

# Cleanup: Kill MongoDB Process
yarn server kill:mongo

Built using

• Node.js - Runtime environment for JS
• Apollo Server - To build a self-documenting GraphQL API server
• Type-graphql - For defining the Graphql schema using classes and decorators and additional features like dependency injection, validation and auth guards to embrace code reusability
• MongoDB - Database to store document-based data
• Mongoose - MongoDB object modeling for Node.js
• Typegoose - Define Mongoose models using TypeScript classes
• JSON Web Token - A standard to secure/authenticate HTTP requests
• Bcrypt.js - For hashing passwords
• Mongoose Unique Validator - Plugin for better error handling of unique fields within Mongoose schema.
• Dotenv - To load environment variables from a .env file
• Docker

Changelog

Update 1: Migrated Server codebase to Typescript using type-graphql, typegoose.

Update 2: (27/10/2021) Automated Graphql Testing using jest, ts-jest, isomorphic-fetch.

Update 3: (29/10/2021) Replaced constant growing arrays with mongoose virtuals.

Update 4: (4/05/2022) Moved to Hybrid(poly repo + mono repo) structure to reduce code,build management issues, also I am not using tools like Nx,turborepo,bazel which would mitigate those issues.

Update 5: Improve code and data models.

TODO

1. Use transactions in mutation resolvers that involve multiple mutating database requests.