Royal Flush

Making using the bathroom, quick, simple, and insightful.

Royal Flush App Set Up

Frontend

cd client
npm install
npm run dev

Backend

To test on Cloud SQL instance

Install Cloud SQL Auth Proxy if you dont have it

OR install with Homebrew

brew install cloud-sql-proxy

Then run the following with the actual gcloud credentials

gcloud auth application-default login
cloud-sql-proxy YOUR_PROJECT_ID:YOUR_REGION:PROJECT_NAME

If this is successful it should expose Postgres on 127.0.0.1:5432 Then ensure you update the server/.env for the local development

PGHOST=127.0.0.1
PGPORT=5432
PGUSER=USERNAME_WE_SET
PGPASSWORD=PASSWORD_WE_SET
PGDATABASE=DATABASE_NAME_WE_SET

To test on local database

A local Postgres instance uses the same database engine as production and keeps test data separate from Cloud SQL, so preferably use this.

1. Start Postgres in Docker (from any directory):

docker run -d \
  --name royalflush-postgres \
  -e POSTGRES_USER=royalflush \
  -e POSTGRES_PASSWORD=royalflush \
  -e POSTGRES_DB=royalflush \
  -p 5433:5432 \
  postgres:16-alpine

2. Configure the server — copy server/.env.example to server/.env and set at least:

PGHOST=127.0.0.1
PGPORT=5433
PGUSER=royalflush
PGPASSWORD=royalflush
PGDATABASE=royalflush
JWT_SECRET=any-long-random-string-for-local-dev

Fill in GCS_BUCKET_NAME, GOOGLE_MAPS_API_KEY, and GOOGLE_APPLICATION_CREDENTIALS if you need uploads, address search, or other Google APIs locally.

3. Run the API — tables are created automatically when the server starts:

cd server
npm install
npm run dev

Useful commands

# Stop / start the DB container (data persists in the container until removed)
docker stop royalflush-postgres
docker start royalflush-postgres

# Wipe and recreate tables (destructive so pls dont do in prod cloud sql)
npm run reset

After local works

We can then deploy to Cloud Run

PGHOST=/cloudsql/....

NOTICE

• NO PGPORT
• SAME PGUSER, PGPASSWORD, PGDATABASE

Redeploying

# 1. Build & push server (from repo root or ./server)
docker build --platform linux/amd64 -t REGION-docker.pkg.dev/PROJECT/REPO/server:latest ./server
docker push REGION-docker.pkg.dev/PROJECT/REPO/server:latest

# 2. Redeploy API (same image tag or new tag)
gcloud run deploy YOUR-API-SERVICE --image=.../server:latest --region=REGION ...

# 3. Build & push client (no VITE_API_URL needed; still need Maps key if you use it)
docker build --platform linux/amd64 \
  --build-arg VITE_GOOGLE_MAPS_API_KEY="..." \
  -t REGION-docker.pkg.dev/PROJECT/REPO/client:latest ./client
docker push REGION-docker.pkg.dev/PROJECT/REPO/client:latest

# 4. Redeploy web WITH API_UPSTREAM
export API_URL="$(gcloud run services describe YOUR-API-SERVICE --region=REGION --format='value(status.url)')"
gcloud run deploy YOUR-WEB-SERVICE \
  --image=.../client:latest \
  --region=REGION \
  --set-env-vars="API_UPSTREAM=${API_URL}"

Resources for Christoper:

• To learn / Get started with docker: https://youtu.be/DQdB7wFEygo?si=esimDqHlx5G8EWE4

Contributors

• John Ortega
• Christopher Persaud

Structure of project

React (static frontend)
    ↓
Cloud CDN + Cloud Storage (or Firebase Hosting)
    ↓
Node.js API
    ↓
Cloud Run (containerized backend)
    ↓
Database (Cloud SQL / Firestore)

• Use redis to run multiple server instances