jumpseat-lite

An open-sourced fork of Jumpseat, built using Typescript, Next.js, and Prisma ORM.

At its core, Jumpseat is a wrapper around Google Flights that automatically pairs and includes shuttle costs and times in itinerary ranking. Thus, Jumpseat might be useful if your school has one or more of the following:

• Multiple airports to fly to/from (for Purdue, this is IND and ORD)
• Infrequent airport transit that requires planning around (shuttles every 3 hours from ORD)
• Cost / time tradeoffs between airports (shuttles are faster + cheaper to IND, but flights are much more expensive)
• Multiple shuttle providers with different schedules

Adding a new school

Adding schools to Jumpseat is easy: first, add the school to the metadata map in ./lib/schools.ts. The type is:

export type SchoolConfig = {
    name: string,
    nameLong: string,
    coordinates: [lat: number, long: number],
    email: string,
    emailPlaceholder: string,
    passwordPlaceholder: string,
    excludedAirports: string[]
}

(look at the Purdue entry for an example); this will add the school to the home-page dropdown.

You will also need to insert the school into the database: create scripts upsert-[school]-breaks.ts and upsert-[school]-shuttles.ts that add the school's breaks and shuttles to the DB, respectively (see the Purdue scripts for an example). You will need to provide

• The airport IATAs Jumpseat should associate with your school.
• A list of (current year) school breaks, for Jumpseat to autofill as options on the search page.
• A list of shuttle providers and their routes; see the expected format in the upsert-shuttles scripts.

Updating school information

To update a school's breaks or shuttle information, simply make a pull request editing the appropriate upsert- script. The data will automatically be updated on merge.

Running locally

Create a .env in the project root exporting your Postgres database / serpapi key, Google Maps key, and Next Auth secret. This should look something like

DATABASE_URL="prisma+postgres://..."
AUTH_SECRET=...
SERP_KEY=...
NEXT_PUBLIC_GOOGLE_MAPS_KEY=...

Run

npm i

to install dependencies,

npx prisma generate

to generate typings for the Prisma client, and

npm run dev

to start the development server on localhost:3000.

The database schema is stored in /prisma/schema.prisma. After editing this schema, run

npx prisma migrate dev

to migrate your changes to the remote database and regenerate Prisma's client typings.

Google Maps setup

To generate a Google Maps key, go to Google Cloud console and create a new project. Under APIs & Services, enable the Maps Javascript API:

Database setup

Note: The below instructions assume a traditional Postgres database hosted on AWS RDS. Currently, for cost reasons, this project uses a Postgres database on Prisma using Prisma accelerate. To switch back to a traditional Postgres connection, add the pg adapter and remove the withAccelerate() plugin on all Prisma clients.

To set up the database on AWS, create an Aurora Postgres database through RDS.

Since the frontend is not deployed directly on AWS through EC2, we need to configure the database to accept external connections. Follow this guide to add a security group in the VPC dashboard that allows inbound connections to :5432.

Once the Aurora database is configured with the above security rule, we'll also need to modify the database instance to allow public connections: under Connectivity > Additional configuration in the instance settings,

Finally, you should be able to connect to your database externally, e.g. via psql with

psql --host=jumpseat.cluster-[...].us-east-2.rds.amazonaws.com --port=5432 --username=postgres

and your database master password (if you didn't create the database with one, you can always add one via the Modify menu).

(note: the psql host is the the Aurora writer endpoint as seen in the console here:)

Set the DATABASE_URL in the .env to

postgres://postgres:[master password]@[aurora writer endpoint]:5432/postgres

as mentioned above, and run

npx prisma migrate dev

to set up the database with the initial migrations.

Scripts

In the case that you must drop all data in the production database, you can run

npx tsx ./scripts/backup.ts

to create JSON backups of all tables in the database. Then, drop the tables as usual.

Restoring the database from the JSON backups is slightly more complex; in general, you want to restore "base" tables before restoring tables with relations that reference them (for example, you'll need to restore User before Friend).

Manually edit ./scripts/recover.ts to facilitate this, then run

npx tsx ./scripts/recover.ts

to redeploy your saved data.