Train booking system - Group 7
Getting started
Dependencies:
- NodeJS
- React
- NestJS
- PostgresSQL
- TypeScript
Initializing the database
Execute docker-compose up
inside the root of /api to create a new docker-container with PostgresSQL v14.1.0.
Setting up environment variables
Create a new .env file inside /api and /client with data that matches their .env.example.
Setting up the projects
Execute npm install inside /api, /client and /shared respectively to install all packages & dependencies neccessary. And npm start to start the projects on your local machine.
Project structure
API
The API uses NestJS and utilizes the MVC pattern to structure its code. This means code is split up between controllers, entities and services.
Mockdata
In order to test and develop the application in a non-production environment you will need to add soem mockdata to the database. The database will automatically get filled up with mockdata from Trafikverket on the initial startup, this is done here.
-MAKE SURE YOU HAVE PROVIDED A TRAFIKVERKET API KEY FOR THIS TO WORK
Alternatively if you wish to provide your own testing data, you can do so by filling up the database BEFORE its initial launch.
Controllers
All controllers use a prefixed /api route.
BookingController
This controller is responsible for all logic regarding finding and booking trips.
- /booking/search (POST) - finds available trips
- /booking/reservation (POST) - books a trip
- /booking (POST) - find previously booked trip
- /booking/cancel (POST) - cancel a previously booked trip
- /booking/create-checkout-session (POST) - initializes a checkout session via Stripe
StationController
This controller is responsible for simple logic regarding station data.
- /station (GET) - finds all available trainstations
StripeController
This controller is responsible for fetching data about an existing/completed checkout session.
- /payment/:sessionId (GET) - fetches checkout data from Stripe sessionId
Client
The frontend/client is powered by React and its logic is split up by pages & components. Data is passed between "phases" of the booking via a BookingContext.
Deployment
Deployment is done automatically via CI/CD workflows found in ./github. All deployment & cloud logic is handled on Azure. Deployment is automatically triggered with pushes to the /main branch.
- API - deployed as a Serverless Function App
- client - deployed as a Static Web App
- database - deployed as a Azure Postgres Server