A realistic REST API example using Actix Web, SQLx, PostgreSQL, and cache-kit.

This example demonstrates:

• ✅ Multi-table relationships (User and Product entities)
• ✅ Async database access with SQLx and migrations
• ✅ REST API endpoints with Actix Web
• ✅ Cache-kit integration using Service Layer pattern
• ✅ Postcard serialization
• ✅ Clean separation of concerns (Routes → Services → Repositories)

• Rust 1.75+
• PostgreSQL 18+
• Docker (for running Postgres)

Using Docker:

make up

See available targets with make help. To build:

cargo build

cargo run

You should see:

Starting server at http://127.0.0.1:8080

The server runs on port 8080.

# Health check
curl http://localhost:8080/health

# Get user
curl http://localhost:8080/users/user_001

# Create user
curl -X POST http://localhost:8080/users \
  -H "Content-Type: application/json" \
  -d '{"id":"user_003","username":"charlie","email":"charlie@example.com","created_at":"2025-12-26T00:00:00Z"}'

This example uses the Service Layer architecture pattern:

• Routes: Handle HTTP requests/responses only, delegate to services
• Services: Contain business logic and cache integration
• Repositories: Pure database access, no cache awareness

This provides clear separation of concerns and makes the code testable and maintainable.

The example demonstrates:

1. Refresh (GET operations)

   • Check cache first
   • On miss, fetch from DB and store in cache
   • On hit, return from cache

2. Invalidate (PUT/DELETE operations)

   • Remove from cache
   • Force fresh fetch on next read

src/
├── main.rs              # Server setup
├── models.rs            # User, Product entities
├── repository.rs        # Database access (no cache)
├── routes.rs            # HTTP handlers
├── error.rs             # Error types
├── lib.rs               # Library exports
└── services/
    ├── mod.rs
    ├── user_service.rs      # User business logic + cache
    └── product_service.rs   # Product business logic + cache

migrations/
├── 001_create_users.sql
└── 002_create_products.sql

Run the complete test suite (includes database setup, migrations, and seeded data):

make test

This validates:

• Database migrations
• Cache hit/miss behavior
• CRUD cycles with cache invalidation

MIT