Fish Fry Orders

A full-stack order management system built for parish fish fry events. It handles real-time order tracking, menu management, session-based daily operations, and historical reporting.

Architecture

• Backend: Go (Echo framework, PostgreSQL, WebSocket for real-time updates, gRPC/protobuf definitions)
• Frontend: React + TypeScript (Vite, deployed via Nginx)
• Infrastructure: Docker, Helm charts for Kubernetes deployment (Istio VirtualService)

Features

• Session-based operations -- sessions auto-create on first order and expire at end of day; admins can extend or close sessions manually
• Real-time order tracking via WebSocket -- all connected clients see order status changes instantly
• Menu management -- admins can create, update, reorder, and deactivate menu items
• Price capture at order time -- order items snapshot the menu item name and price, so historical data stays accurate even if prices change
• Role-based auth -- worker and admin roles with password-based login and session cookies
• Historical session comparison -- compare order counts, revenue, and item breakdowns across past sessions
• Daily order numbers -- human-friendly order numbers that reset per session (e.g. "Order #7")

Project Structure

cmd/
  server/         # Main HTTP/WebSocket server entrypoint
  migrate/        # Database migration CLI tool
internal/
  api/            # HTTP handlers (orders, menu, sessions)
  auth/           # Authentication service and middleware
  config/         # YAML config loading
  database/       # PostgreSQL repository, connection pooling, migrations
  logger/         # Structured logging
  metrics/        # Prometheus metrics
  types/          # Shared DB types and proto conversions
proto/            # Protobuf definitions and generated Go code
frontend/         # React + TypeScript SPA
testing/          # Integration tests with Docker Compose-managed Postgres
helm/             # Helm chart with per-environment value overrides
docs/             # Architecture and design documents

Getting Started

Prerequisites

• Go 1.24+
• PostgreSQL 15+
• Node.js 18+ (for frontend)

Backend

# Copy and edit config
cp config.yaml config.local.yaml
# Edit config.local.yaml with your DB credentials

# Run database migrations
go run cmd/migrate/main.go -command up

# Start the server
go run cmd/server/main.go

The API server listens on :8080 by default.

Frontend

cd frontend
npm ci
npm run dev

The dev server runs on http://localhost:5173 and proxies API requests to the backend.

Docker

# Build both images
make docker

# Or individually
make docker-api
make docker-frontend

Helm Deployment

The project ships with Helm charts supporting three environments: prod, demo, and dev.

# Deploy to production
make helm-deploy ENV=prod

# Deploy to demo
make helm-deploy ENV=demo

# Check status
make helm-status ENV=prod

Configuration

Configuration is loaded from config.yaml with environment variable substitution for secrets:

Key	Description	Default
database.host	PostgreSQL host	localhost
database.port	PostgreSQL port	5432
database.dbname	Database name	fish_fry_orders
http.address	HTTP listen address	:8080
auth.worker_password	Password for worker role	(from env AUTH_WORKER_PASSWORD)
auth.admin_password	Password for admin role	(from env AUTH_ADMIN_PASSWORD)
allowed_origins	CORS / WebSocket allowed origins	localhost:5173, localhost:8080

Testing

# Run all tests (requires Docker for integration tests)
make test

# Run integration/E2E tests only
make test-e2e

Integration tests use Docker Compose to spin up a temporary PostgreSQL instance, run migrations, seed test data, and tear down after each test.

License

Apache License 2.0. See LICENSE for details.