Project description

This is the back-end server and database ORM for my mock project for the Angular Course at Softuni, client located here. The server accepts REST requests and sits between the Angular client and the database. Server works with JSON.

Engine used is Express. Controllers handle the different endpoints and call on services to perform database operations via Sequelize's ORM for PostgreSQL. Upon first run, the ORM will create the database and create the necessary relationships. Different (mostly bare-essential) middleware are attached to the server.

Technologies used

** Taken offline due to ongoing costs in AWS

• Front-end - Details
• Back-end - Javascript (Typescript), Sequelize ORM for PostreSQL, deployed via AWS CodeBuild
• Database Hosting - PostgreSQL hosted on AWS-RDS
• Server Hosting - Hosted via AWS Beanstalk, endpoint: http://node-typescript.eu-west-1.elasticbeanstalk.com
• Middlewares - helmet, compression, cors
• Frameworks/Libraries
  • Express
  • Sequelize ORM
  • bcrypt
  • jsonwebtoken
  • axios (pulling some data dynamically from API)
  • nodemon (for running locally)

Setup

NodeJS and Python 3.9 are necessary. To run the server locally, you will need a PostgreSQL server running locally. Create an .env file in the root dir and add your server configuration as per the src/config/database.ts naming conventions.

To run the project locally, you will need the back-end located here, as well as a PostgreSQL database to hook the back-end to, and run that one locally as well in parallel. NodeJS is necessary. Steps to follow for front-end:

• Install NodeJS (v16+)
• Install Python (v3.9+)
• Run npm i
• (Optional, loads device data into the DB) Run npm run load ***
• Run npm run dev:ts
• The server and database should now be up at the port of your choosing

*** This will pull data from the external API via Python workers and will populate the JSON files which will be generated in databaseLoad/rawData. The service will then clean the data and will load it into the database via Sequelize. Do NOTE that pulling the data from the API is a very time consuming processing, taking between 6 and 8 hours due to the volume of data and the endpoint needing requests to be done 1 by 1. The Python script will do this until all data is exhausted.

To host in the cloud, you will need an AWS-RDS hosted PostreSQL server and a CodeBuild project. You can deploy a pipeline from this repository directly once you have everything else configured. If using with the Angular client, the Angular client will also need your server endpoint to function.

Endpoints

• /data
  • GET /health -> server status
  • GET /recommended -> new/recommended devices
  • POST /news | with route:search and query in body -> news on device query
• /device
  • POST /list | optionally with limit and offset in body -> raw device data
  • POST /list/search | optionally with query, limit and offset -> queried device data
• /postings
  • POST /list | optionally with limit and offset in body -> raw postings data
  • POST /list/search | optionally with query, limit and offset -> queried postings data
  • GET /user/:id | user ID in params -> postings for specific user
  • POST /create | with posting payload in body -> create new posting
  • ** PUT /edit/:id | posting ID in params with price and comments in body -> edits specific posting
  • ** DELETE /delete/:id | posting ID in params -> deletes specific posting
• /users
  • POST /register | with register payload -> registers new user
  • POST /login | with login payload -> logs user in
  • POST /logout | with accessToken in body -> logs user out
  • POST /validate | with token in body -> validates user token
  • ** PATCH /edit/details/:id | user ID in params and profile payload in body -> changes user details
  • ** PATCH /edit/password/:id | user ID in params and password in body -> changes user password
  • ** DELETE /delete/:id | user ID in params -> deletes specific user

** Token of active user (if any) will be matched against back-end data for validation.

Credits

Thanks to ceo-py for assistance with the Python script

Architecture