gRPC Elixir

An Elixir implementation of gRPC.

Table of contents

• Installation
• Usage
  • Simple RPC
  • HTTP Transcoding
  • Start Application
• Features
• Benchmark
• Contributing

Installation

The package can be installed as:

def deps do
  [
    {:grpc, "~> 0.9"}
  ]
end

Usage

1. Write your protobuf file:

syntax = "proto3";

package helloworld;

// The request message containing the user's name.
message HelloRequest {
  string name = 1;
}

// The response message containing the greeting
message HelloReply {
  string message = 1;
}

// The greeting service definition.
service Greeter {
  // Greeting function
  rpc SayHello (HelloRequest) returns (HelloReply) {}
}

2. Then generate Elixir code from proto file as protobuf-elixir shows (especially the gRPC Support section) or using protobuf_generate hex package. Example using protobuf_generate lib:

mix protobuf.generate --output-path=./lib --include-path=./priv/protos helloworld.proto

In the following sections you will see how to implement gRPC server logic.

Simple RPC

1. Implement the server side code like below and remember to return the expected message types.

defmodule Helloworld.Greeter.Server do
  use GRPC.Server, service: Helloworld.Greeter.Service

  @spec say_hello(Helloworld.HelloRequest.t, GRPC.Server.Stream.t) :: Helloworld.HelloReply.t
  def say_hello(request, _stream) do
    Helloworld.HelloReply.new(message: "Hello #{request.name}")
  end
end

2. Define gRPC endpoints

# Define your endpoint
defmodule Helloworld.Endpoint do
  use GRPC.Endpoint

  intercept GRPC.Server.Interceptors.Logger
  run Helloworld.Greeter.Server
end

We will use this module in the gRPC server startup section.

Note: For other types of RPC call like streams see here.

HTTP Transcoding

1. Adding grpc-gateway annotations to your protobuf file definition:

import "google/api/annotations.proto";
import "google/protobuf/timestamp.proto";

package helloworld;

// The greeting service definition.
service Greeter {
  // Sends a greeting
  rpc SayHello (HelloRequest) returns (HelloReply) {
    option (google.api.http) = {
      get: "/v1/greeter/{name}"
    };
  }

  rpc SayHelloFrom (HelloRequestFrom) returns (HelloReply) {
    option (google.api.http) = {
      post: "/v1/greeter"
      body: "*"
    };
  }
}

2. Add protoc plugin dependency and compile your protos using protobuf_generate hex package:

In mix.exs:

def deps do
  [
    {:grpc, "~> 0.7"},
    {:protobuf_generate, "~> 0.1.1"}
  ]
end

And in your terminal:

mix protobuf.generate \
  --include-path=priv/proto \
  --include-path=deps/googleapis \
  --generate-descriptors=true \
  --output-path=./lib \
  --plugins=ProtobufGenerate.Plugins.GRPCWithOptions \
  google/api/annotations.proto google/api/http.proto helloworld.proto

3. Enable http_transcode option in your Server module

defmodule Helloworld.Greeter.Server do
  use GRPC.Server,
    service: Helloworld.Greeter.Service,
    http_transcode: true

  @spec say_hello(Helloworld.HelloRequest.t, GRPC.Server.Stream.t) :: Helloworld.HelloReply.t
  def say_hello(request, _stream) do
    %Helloworld.HelloReply{message: "Hello #{request.name}"}
  end
end

See full application code in helloworld_transcoding example.

Start Application

1. Start gRPC Server in your supervisor tree or Application module:

# In the start function of your Application
defmodule HelloworldApp do
  use Application
  def start(_type, _args) do
    children = [
      # ...
      {GRPC.Server.Supervisor, endpoint: Helloworld.Endpoint, port: 50051, start_server: true}
    ]

    opts = [strategy: :one_for_one, name: YourApp]
    Supervisor.start_link(children, opts)
  end
end

2. Call rpc:

iex> {:ok, channel} = GRPC.Stub.connect("localhost:50051")
iex> request = Helloworld.HelloRequest.new(name: "grpc-elixir")
iex> {:ok, reply} = channel |> Helloworld.Greeter.Stub.say_hello(request)

# With interceptors
iex> {:ok, channel} = GRPC.Stub.connect("localhost:50051", interceptors: [GRPC.Client.Interceptors.Logger])
...

Check the examples and interop directories in the project's source code for some examples.

Client Adapter and Configuration

The default adapter used by GRPC.Stub.connect/2 is GRPC.Client.Adapter.Gun. Another option is to use GRPC.Client.Adapters.Mint instead, like so:

GRPC.Stub.connect("localhost:50051",
  # Use Mint adapter instead of default Gun
  adapter: GRPC.Client.Adapters.Mint
)

The GRPC.Client.Adapters.Mint adapter accepts custom configuration. To do so, you can configure it from your mix application via:

# File: your application's config file.
config :grpc, GRPC.Client.Adapters.Mint, custom_opts

The accepted options for configuration are the ones listed on Mint.HTTP.connect/4

Features

• Various kinds of RPC:
  • Unary
  • Server-streaming
  • Client-streaming
  • Bidirectional-streaming
• HTTP Transcoding
• TLS Authentication
• Error handling
• Interceptors (See GRPC.Endpoint)
• Connection Backoff
• Data compression
• gRPC Reflection

Benchmark

1. Simple benchmark by using ghz

2. Benchmark followed by official spec

Contributing

Your contributions are welcome!

Please open issues if you have questions, problems and ideas. You can create pull requests directly if you want to fix little bugs, add small features and so on. But you'd better use issues first if you want to add a big feature or change a lot of code.