hexastack-grpc

High-performance gRPC presentation adapter, interceptors, and CQRS service integration for Hexastack.

1. Overview & Capabilities

hexastack-grpc provides a binary RPC presentation adapter for Hexastack applications:

• Declarative Service Registration (@grpc_service): Mounts generated protobuf RPC servicers onto the grpc.Server with automatic DI resolution.
• gRPC Health Checking Protocol: Standard grpc.health.v1.Health servicer (GrpcHealthServicer) supporting Kubernetes liveness, readiness, and stream health probes.
• Bidirectional & Server-Streaming RPC Dispatch: dispatch_rpc_stream_query and dispatch_rpc_bidirectional_stream streaming CQRS command and query pipelines over HTTP/2.
• Cross-Cutting Interceptor Pipeline:
  • CorrelationServerInterceptor: Propagates x-correlation-id from incoming metadata into ContextVar.
  • LoggingServerInterceptor: Structured telemetry for RPC method invocations.
  • TimingServerInterceptor: Measures RPC execution latency in milliseconds.
  • MetricsServerInterceptor: Records RPC request rates, statuses, and handling durations to MetricsPort.
• Single-Pass Reflection: Automatically discovers and registers decorated servicers in Phase 3 module scanning via create_grpc_visitor.
• gRPC Server Reflection: Automatically enables gRPC Server Reflection Protocol when grpcio-reflection is installed.

2. Package Anatomy & Key Components

hexastack_grpc/
├── domain/          # GrpcError, ServiceRegistrationError, RpcExecutionError
├── adapters/        # create_async_grpc_server, run_grpc_server, GrpcHealthServicer
└── infra/
    ├── bootstrap.py # GrpcBootstrapper (order=40)
    ├── config.py    # HexastackGrpcConfig
    ├── decorators.py# @grpc_service
    ├── dispatch.py  # dispatch_rpc_command, dispatch_rpc_query, dispatch_rpc_stream_query, dispatch_rpc_bidirectional_stream
    ├── autodiscovery.py # create_grpc_visitor, autodiscover_grpc_services
    ├── interceptors/# correlation, logging, timing interceptors
    └── registries/  # service.py (GrpcServiceRegistry)

3. Monorepo & Sibling Relationships

graph TD
    subgraph Clients ["gRPC Clients (Microservices, Mobile, Gateway)"]
        CLIENT["gRPC Protobuf Client"]
    end

    subgraph GrpcAdapter ["hexastack-grpc"]
        SERVER["grpc.Server"]
        INTERCEPTORS["Interceptors (Correlation, Logging, Timing)"]
        REG["GrpcServiceRegistry"]
        SERVICER["CQRS Protobuf Servicer"]
    end

    subgraph CQRSExecution ["hexastack-cqrs"]
        CBUS["CommandBusPort"]
        QBUS["QueryBusPort"]
    end

    CLIENT --> SERVER
    SERVER --> INTERCEPTORS
    INTERCEPTORS --> SERVICER
    SERVICER -->|dispatches commands| CBUS
    SERVICER -->|dispatches queries| QBUS

Explicit Dependencies (Direct)

• hexastack-core: DI container (rodi), configuration registry, context variables.
• hexastack-cqrs: CommandBusPort and QueryBusPort.
• grpcio>=1.68.0, protobuf>=5.29.0.

Optional Integrations (Extras)

• [reflection]: Installs grpcio-reflection>=1.68.0 for runtime service reflection tools (grpcurl, Postman).

4. Installation

# Standalone installation
pip install hexastack-grpc

# With Server Reflection support
pip install "hexastack-grpc[reflection]"

# Via umbrella package
pip install "hexastack[grpc]"

5. Configuration Reference

[hexastack.grpc]
host = "0.0.0.0"
port = 50051
max_workers = 10
enable_reflection = true
auto_start = false

6. Quickstart Example

from hexastack_core.infra.bootstrap import bootstrap
from hexastack_grpc.infra.decorators import grpc_service


# 1. Define Servicer implementing generated protobuf class
class GreeterServicer:
    def SayHello(self, request, context):
        return HelloReply(message=f"Hello, {request.name}!")


# 2. Register Servicer with generated add hook
@grpc_service(add_GreeterServicer_to_server)
class GreeterService(GreeterServicer):
    pass


# 3. Bootstrap Runtime and Start Server
runtime = bootstrap(packages_to_scan=[__name__])
grpc_server = runtime.get("grpc_server")
grpc_server.start()