GRPC与.NET Core

（给DotNet加星标，提升.Net技能）

转自：老六代码

cnblogs.com/chenyishi/p/11143375.html

概述

GRPC的数据交互模式有：

1、单项RPC，最简单的数据交换方式，客户端发出单个请求，收到单个响应

2、服务端流式RPC，是在服务端收到客户端的请求之后，返回一个应答流，客户端收到流之后处理。

3、客户端流式RPC，与单项类似，但客户端发送的是流式RPC

4、双向流式RPC，调用由客户端调用方法来初始化，而服务端则接收到客户端的元数据，方法名和截止时间。服务端可以选择发送回它的初始元数据或等待客户端发送请求。下一步怎样发展取决于应用，因为客户端和服务端能在任意顺序上读写 - 这些流的操作是完全独立的。例如服务端可以一直等直到它接收到所有客户端的消息才写应答，或者服务端和客户端可以像"乒乓球"一样：服务端后得到一个请求就回送一个应答，接着客户端根据应答来发送另一个请求，以此类推。

单项RPC较简单不做示例了。

首先在VS 2019中.NET Core 3.0中新建GRPC项目。然后定义响应的proto文件，根据proto文件生成响应的服务端与客户端代码。

一、服务端流式RPC

1、定义 protofile

syntax = "proto3";

option csharp_namespace = "GrpcGreeter";

package Greet;

// The greeting service definition.
service Greeter {
  // Sends a greeting
  rpc SayHello (HelloRequest) returns (HelloReply) {}
  rpc GetStreamContent (StreamRequest) returns (stream StreamContent) {}
}

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

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

message StreamRequest {
  string fileName = 1;
}
message StreamContent {
  bytes content = 1;
}

2、实现服务端Service

重新生成项目，然后实现GetStreamContent，简单的读取文件内容，并将内容返回给Client

using System;
using System.Collections.Generic;
using System.IO;
using System.Linq;
using System.Threading.Tasks;
using Google.Protobuf;
using Grpc.Core;
namespace GrpcGreeter
{
    public class GreeterService : Greeter.GreeterBase
    {
        public override Task<HelloReply> SayHello(HelloRequest request, ServerCallContext context)
        {
            return Task.FromResult(new HelloReply
            {
                Message = "Hello " + request.Name
            });
        }
        public override Task GetStreamContent(StreamRequest request, IServerStreamWriter<StreamContent> responseStream, ServerCallContext context)
        {
            return Task.Run(async () =>
            {
                 using (var fs = File.Open(request.FileName, FileMode.Open)) // 从 request 中读取文件名并打开文件流
                 {
                     var remainingLength = fs.Length; // 剩余长度
                     var buff = new byte[1048576]; // 缓冲区，这里我们设置为 1 Mb
                     while (remainingLength > 0) // 若未读完则继续读取
                     {
                         var len = await fs.ReadAsync(buff); // 异步从文件中读取数据到缓冲区中
                         remainingLength -= len; // 剩余长度减去刚才实际读取的长度
                         // 向流中写入我们刚刚读取的数据
                         await responseStream.WriteAsync(new StreamContent
                         {
                             Content = ByteString.CopyFrom(buff, 0, len)
                         });
                     }
                 }
             });
        }
    }
}

3、实现Client

新建一个netcore 3.0的Console项目，并引入Nuget包

Install-Package Grpc.Net.Client -Version 0.1.22-pre1
Install-Package Google.Protobuf -Version 3.8.0
Install-Package Grpc.Tools -Version 1.22.0

编辑项目文件，修改如下节点

<Protobuf Include="Protos\greet.proto" GrpcServices="Client" />

重新生成项目，Client端主要实现发送请求，请求是一个服务器端的文件路径。然后实现接收服务端的流，并保存到Client本地。

using Grpc.Net.Client;
using GrpcGreeter;
using System;
using System.Collections.Generic;
using System.IO;
using System.Net.Http;

namespace GrpcGreeterClient
{
    class Program
    {
        static async System.Threading.Tasks.Task Main(string[] args)
        {
            AppContext.SetSwitch(          "System.Net.Http.SocketsHttpHandler.Http2UnencryptedSupport",
                true);
            var httpClient = new HttpClient();
            // The port number(50051) must match the port of the gRPC server.
            httpClient.BaseAddress = new Uri("http://localhost:50051");
            var client = GrpcClient.Create<Greeter.GreeterClient>(httpClient);
            //
            var reply = await client.SayHelloAsync(
                              new HelloRequest { Name = "GreeterClient" });
            Console.WriteLine("Greeting: " + reply.Message);
            Console.ReadKey();
            //
            var result = client.GetStreamContent(new StreamRequest { FileName = @"D:\Docs.zip" });  //发送请求
            var iter = result.ResponseStream;
            using (var fs = new FileStream(@"D:\Docs2.zip", FileMode.Create)) // 新建一个文件流用于存放我们获取到数据
            {
                while (await iter.MoveNext()) // 迭代
                {
                    iter.Current.Content.WriteTo(fs); // 将数据写入到文件流中
                }
            }
            Console.ReadKey();
        }
    }
}

文件生成成功

二、客户端流式RPC

1、定义 protofile

syntax = "proto3";
option csharp_namespace = "GRPC.TEST";
package Greet;

// The greeting service definition.
service Greeter {
  rpc getResult (stream Value) returns (Result) {}
}

//定义Value消息类型，用于客户端消息
message Value {
    int32 value = 1;
}
//定义Result消息类型，包含总和，数字数量和平均值，用于服务端消息返回
message Result {
    int32 sum = 1;
    int32 cnt = 2;
    double avg = 3;
}

2、实现服务端Service

重新生成项目，并实现如下

using System;
using System.Collections.Generic;
using System.IO;
using System.Linq;
using System.Threading.Tasks;
using Google.Protobuf;
using Grpc.Core;

namespace GRPC.TEST
{
    public class GreeterService : Greeter.GreeterBase
    {
        public override async Task<Result> getResult(IAsyncStreamReader<Value> requestStream, ServerCallContext context)
        {
            while (await requestStream.MoveNext())
            {
                var point = requestStream.Current;
            }
            return new Result { Sum = 1 };
        }
    }
}

3、实现Client

新建一个netcore 3.0的Console项目，并引入Nuget包，安装nuget包与其他操作同上一个例子，实现代码如下

using Grpc.Net.Client;
using System;
using System.Collections.Generic;
using System.IO;
using System.Net.Http;

namespace GRPC.TEST.CLIENT
{
    class Program
    {
        static async System.Threading.Tasks.Task Main(string[] args)
        {
            AppContext.SetSwitch(               "System.Net.Http.SocketsHttpHandler.Http2UnencryptedSupport",
                 true);
            var httpClient = new HttpClient();
            // The port number(50051) must match the port of the gRPC server.
            httpClient.BaseAddress = new Uri("http://localhost:50051");
            var client = GrpcClient.Create<Greeter.GreeterClient>(httpClient);
            using (var call = client.getResult())
            {
                await call.RequestStream.WriteAsync(new Value { Value_ = 1 });
                await call.RequestStream.CompleteAsync();
                var response = await call.ResponseAsync;
            }
            Console.ReadKey();
        }
    }
}

三、双向流式RPC

1、定义proto

syntax = "proto3";
option csharp_namespace = "GRPC.TEST";
package Greet;

// The greeting service definition.
service Greeter {
  rpc getResult (stream Value) returns (stream Result) {}
}


//定义Value消息类型，用于客户端消息
message Value {
    int32 value = 1;
}
//定义Result消息类型，包含总和，数字数量和平均值，用于服务端消息返回
message Result {
    int32 sum = 1;
    int32 cnt = 2;
    double avg = 3;
}

2、服务端实现

重新生成项目，并实现如下

public override async Task getResult(IAsyncStreamReader<Value> requestStream, IServerStreamWriter<Result> responseStream, ServerCallContext context)
{
            while (await requestStream.MoveNext())
            {
                var note = requestStream.Current;
                await responseStream.WriteAsync(new Result { Sum = 100 });
            }
}

3、客户端代码

新建一个netcore 3.0的Console项目，并引入Nuget包，安装nuget包与其他操作同上一个例子，实现代码如下

using Grpc.Net.Client;
using System;
using System.Collections.Generic;
using System.IO;
using System.Net.Http;
using System.Threading.Tasks;
namespace GRPC.TEST.CLIENT
{
    class Program
    {
        static async System.Threading.Tasks.Task Main(string[] args)
        {
            AppContext.SetSwitch(
"System.Net.Http.SocketsHttpHandler.Http2UnencryptedSupport",

       true);
            var httpClient = new HttpClient();
            // The port number(50051) must match the port of the gRPC server.
            httpClient.BaseAddress = new Uri("http://localhost:50051");
            var client = GrpcClient.Create<Greeter.GreeterClient>(httpClient);
            using (var call = client.getResult())
            {
                var responseReaderTask = Task.Run(async () =>
                {
                    while (await call.ResponseStream.MoveNext())
                    {
                        var note = call.ResponseStream.Current;
                        Console.WriteLine("Received " + note);
                    }
                });
                await call.RequestStream.WriteAsync(new Value { Value_ = 12 });
                await call.RequestStream.CompleteAsync();
                await responseReaderTask;
            }
            Console.ReadKey();
        }
    }
}

至此，GRPC的几种数据交互分享完毕

推荐阅读

（点击标题可跳转阅读）

看完本文有收获？请转发分享给更多人

关注「DotNet」加星标，提升.Net技能 

好文章，我在看❤️