Go语言Grpc Stream

时间:2022-06-20 范闲人气:6

Stream Grpc

在我们单次投递的数据量很大的时候，比如传输一个二进制文件的时候，数据包过大，会造成瞬时传输压力。或者接收方接收到数据后，需要对数据做一系列的处理工作，

比如：数据过滤 -> 数据格式转换 -> 数据求和，这种场景非常适合使用stream grpc，

Stream Grpc演示

syntax = "proto3";

package book_stream;

option go_package = "/book_stream";

service HelloStreamService {
  rpc BookListStream(BookListStreamRequest) returns (stream BookListStreamResponse){};
  rpc CreateBookStream(stream CreateBookStreamRequest) returns (CreateBookStreamResponse){}
  rpc FindBookByIdStream(stream FindBookByIdStreamRequest) returns (stream FindBookByIdStreamResponse){}
}

message BookListStreamRequest{
}

message BookListStreamResponse{
  BookPoint book = 1;
}

message CreateBookStreamRequest{
  BookPoint book = 1;
}

message CreateBookStreamResponse{
  repeated BookIdPoint idx = 1;
}

message FindBookByIdStreamRequest{
  BookIdPoint idx = 1;
}
message FindBookByIdStreamResponse{
  BookPoint book = 1;
}

message BookIdPoint{
  uint64 idx = 1;
}

message BookPoint{
  uint64 idx = 1;
  string name = 2;
  float price = 3;
  string author = 4;
}

运行protoc --go_out=plugins=grpc:. *.proto生成脚手架文件

BookListStream服务端流式RPC
CreateBookStream客户端流式RPC
FindBookByIdStream双向流式RPC

注意，这里只是用作方便演示使用，演示方法都不是线程安全的

服务端server

var port = 8888

func main() {
   server := grpc.NewServer()
   book_stream.RegisterHelloStreamServiceServer(server, new(HelloStreamServiceImpl))
   lis, err := net.Listen("tcp", fmt.Sprintf(":%d", port))
   if err != nil {
      panic(err)
   }
   if err := server.Serve(lis); err != nil {
      panic(err)
   }
}

客户端

func main() {
   var port = 8888
   conn, err := grpc.Dial(fmt.Sprintf(":%d", port), grpc.WithInsecure())
   if err != nil {
      panic(err)
   }
   defer conn.Close()
   client := book_stream.NewHelloStreamServiceClient(conn)

   ctx := context.Background()
   if err := createBookStream(ctx, client); err != nil {
      panic(err)
   }
   if err := printBookList(ctx, client); err != nil {
      panic(err)
   }
   if err := getBookListById(ctx, client); err != nil {
      panic(err)
   }
}

BookListStream

服务器端流式 RPC，显然是单向流，并代指 Server 为 Stream 而 Client 为普通 RPC 请求

简单来讲就是客户端发起一次普通的 RPC 请求，服务端通过流式响应多次发送数据集，客户端 Recv 接收数据集。

server端实现

var bookStore = map[uint64]book_stream.BookPoint{
   1: {
      Idx:    1,
      Author: "程子",
      Price:  9.9,
      Name:   "游戏思维",
   },
   2: {
      Idx:    2,
      Author: "丁锐",
      Price:  9.9,
      Name:   "活出必要的锋芒",
   },
}


type HelloStreamServiceImpl struct{}

func (HelloStreamServiceImpl) BookListStream(_ *book_stream.BookListStreamRequest, streamServer book_stream.HelloStreamService_BookListStreamServer) error {
   for idx, bookPoint := range bookStore {
      err := streamServer.Send(&book_stream.BookListStreamResponse{Book: &book_stream.BookPoint{
         Idx:    idx,
         Name:   bookPoint.Name,
         Price:  bookPoint.GetPrice(),
         Author: bookPoint.Author,
      }})
      if err != nil {
         return err
      }
   }
   return nil
}

客户端实现

func printBookList(ctx context.Context, client book_stream.HelloStreamServiceClient) error {
   req := &book_stream.BookListStreamRequest{}
   listStream, err := client.BookListStream(ctx, req)
   if err != nil {
      return err
   }
   for true {
      resp, err := listStream.Recv()
      if err != nil {
         if err == io.EOF {
            return nil
         }
         return err
      }
      fmt.Printf("%v\n", *resp.Book)
   }
   return nil
}

CreateBookStream

客户端流式 RPC，单向流，客户端通过流式发起多次 RPC 请求给服务端，服务端发起一次响应给客户端

server端实现

func (HelloStreamServiceImpl) CreateBookStream(server book_stream.HelloStreamService_CreateBookStreamServer) error {
   var resList []*book_stream.BookIdPoint
   for {
      resp, err := server.Recv()
      if err == io.EOF {
         return server.SendAndClose(&book_stream.CreateBookStreamResponse{Idx: resList})
      }
      if err != nil {
         return err
      }
      bookStore[resp.Book.Idx] = *resp.Book
      resList = append(resList, &book_stream.BookIdPoint{Idx: resp.Book.Idx})
   }
}

客户端实现

var newBookStore = map[uint64]book_stream.BookPoint{
   3: {
      Idx:    3,
      Author: "程子1",
      Price:  9.9,
      Name:   "游戏思维1",
   },
   4: {
      Idx:    4,
      Author: "丁锐1",
      Price:  9.9,
      Name:   "活出必要的锋芒1",
   },
}

func createBookStream(ctx context.Context, client book_stream.HelloStreamServiceClient) error {
   stream, err := client.CreateBookStream(ctx)
   if err != nil {
      return err
   }
   for _, bookPoint := range newBookStore {
      if err := stream.Send(&book_stream.CreateBookStreamRequest{
         Book: &bookPoint,
      }); err != nil {
         return err
      }
   }
   recv, err := stream.CloseAndRecv()
   if err != nil {
      return err
   }
   fmt.Println(recv.Idx)
   return nil
}

stream.SendAndClose，它是做什么用的呢？

在这段程序中，我们对每一个 Recv 都进行了处理，当发现 io.EOF (流关闭) 后，需要将最终的响应结果发送给客户端，同时关闭正在另外一侧等待的 Recv

stream.CloseAndRecv 和 stream.SendAndClose 是配套使用的流方法，

FindBookByIdStream

服务端实现

func (HelloStreamServiceImpl) FindBookByIdStream(streamServer book_stream.HelloStreamService_FindBookByIdStreamServer) error {
   for {
      resp, err := streamServer.Recv()
      if err == io.EOF {
         return nil
      }
      if err != nil {
         return err
      }
      if book, ok := bookStore[resp.Idx.Idx]; ok {
         if err := streamServer.Send(&book_stream.FindBookByIdStreamResponse{Book: &book}); err != nil {
            return err
         }
      }
   }
}

客户端实现

func getBookListById(ctx context.Context, client book_stream.HelloStreamServiceClient) error {
   stream, err := client.FindBookByIdStream(ctx)
   if err != nil {
      return err
   }
   var findList = []uint64{1, 2}
   for _, idx := range findList {
      err := stream.Send(&book_stream.FindBookByIdStreamRequest{Idx: &book_stream.BookIdPoint{Idx: idx}})
      if err != nil {
         return err
      }
      recv, err := stream.Recv()
      if err != nil {
         return err
      }
      fmt.Printf("%v\n", recv.Book)
   }
   if err := stream.CloseSend(); err != nil {
      return err
   }
   return nil
}

加载全部内容