Patient management (Spring microservices)

Technology

• Java and Spring Boot services: API Gateway (Spring Cloud Gateway), Auth (Spring Security, JWT), Patient, Billing, and Analytics.
• PostgreSQL for persistent data; Apache Kafka for async domain events; gRPC and Protocol Buffers for synchronous service calls (e.g. patient ↔ billing).
• Docker-ready services; LocalStack-based AWS-style infrastructure in infrastructure/ for local cloud-style deployment and testing.
• Maven builds; SpringDoc OpenAPI for API documentation behind the gateway.

Methodology

The system is split by bounded context: each service owns its data and API. External traffic enters through the API Gateway, which routes to backend services and runs JWT validation on protected routes. Patient events are published to Kafka for Analytics; Billing is invoked via gRPC when patient flows require it. This keeps concerns separated while allowing integration tests across services (integration-tests).

Project startup (local)

Prerequisites: JDK 21, Maven, and Docker (for dependencies such as databases, Kafka, and peer services when you run in containers).

1. Infrastructure: Start the dependencies your application.properties / application.yml expect (PostgreSQL, Kafka, and any other containers your environment uses). Configure JDBC URLs, Kafka bootstrap servers, and gRPC host/ports to match that environment (see each service’s src/main/resources).

2. Service ports (defaults in repo): Auth 4005, Patient 4000, Billing 4001 (gRPC often on 9001), Gateway 4004, Analytics as configured. In Docker, hostnames in gateway routes match service names; locally you may use localhost and matching ports in overrides.

3. Build each service: from the service directory (e.g. auth-service/), run: mvn clean install (or mvn spring-boot:run to start). Repeat for patient-service, billing-service, analytics-service, and api-gateway in an order that satisfies dependencies: typically databases and Kafka first, then auth and domain services, then the gateway last.

4. HTTP tryouts: use the sample requests under api-requests/ and grpc-requests/ with your base URL (e.g. gateway on port 4004).

5. Optional LocalStack / CDK stack: with LocalStack available, you can use infrastructure/localstack-deploy.sh and the infrastructure project as documented in that module’s POM and scripts.

Adjust env vars and Spring profiles (e.g. application-prod.yml on the gateway) to match your deployment; there is no single docker-compose in this repository, so wire containers or IDE run configurations to the ports and hosts above.