Setup: Running the Backend

Now for the fun part - actually running the course-catalog-api.

Run these commands in order:

1. cp template.env .env
   • This will copy our templated .env file for your own use. Some of these environment variables are required for our codebase to work. Make sure to fill them out!
2. yarn install
   • This command installs all of our dependencies, and it does so locally. In other words, these dependencies are only visible to this project.
3. yarn dev:docker
   • This creates two Docker containers for us, which we'll use for running the backend. One is Postgres, a relational database, which we use for storing data. The other is Elasticsearch, which helps us return results for search query.
4. yarn db:migrate
   • This does all the setup for our newly-created database, creating all the tables that we need to store our data
5. yarn db:refresh
   • This generated a custom Prisma client for our project
   • Prisma is an ORM - it allows us to communicate with our database in a Javascript-based fashion, instead of executing raw SQL statements

!> Important: If you are not on the Northeasern NUWave wifi (or don't want to wait for a scrape [~30 minutes]), please read the "Cache structure" page.

5. yarn scrape
   • Normally, this command would scrape Northeastern's course catalog for course data.
     • If you have installed the cache (see "Cache structure"), this command will just populate our database with the cached data. No network calls will be made.
     • If not, this will scrape Northeastern's live Banner API.
6. yarn dev
   • This command starts our GraphQL API. You should be able to access it at localhost:4000 to see the GraphQL playground

?> Mac users may get a message along the lines of Port 5000 already in use. There's a process called Control Center running on that port; it's the AirPlay server. This can be disabled by turning off "AirPlay Receiver" in the "Sharing" System Preference.