Recently I’ve been busy integrating third-party advertising platforms. During this process, I noticed that some API documentation was simply provided as Word documents with hundreds of parameters listed in tables, requiring manual parameter construction - a truly tedious process.

However, several providers stood out by offering .proto files, allowing automatic generation of Request and Response structures through protobuf - a developer-friendly approach worth applauding!

Key Takeaways

• When integrating APIs, always ask business partners if protobuf schema files are available
  • Generate Go code using: protoc pb\xxx.proto --go_out=.
  • This automatically creates ready-to-use request/response structures, enums, etc., significantly streamlining backend integration (a pattern worth adopting for future backend interfaces)

Sample Implementation

package main

import (
	"bytes"
	"io/ioutil"
	"log"
	"net/http"

	"github.com/golang/protobuf/proto"
	"xxxx/internal/pb"
)

func main() {
	getDataByAdServe()
}

func getDataByAdServe() {
    // Serialize request data
    r := &pb.AdProfRequest{}
    data, er := proto.Marshal(r)
    if er != nil {
        log.Fatal(er)
        return
    }
    
    // Send HTTP request (set proper timeout in production)
    reader := bytes.NewReader(data)
    resp, err := http.Post("http://127.0.0.1:8001/xxx", "application/x-protobuf", reader)
    if err != nil {
        log.Println(err)
        return
    }
    defer resp.Body.Close()
    
    // Read response
    body, err := ioutil.ReadAll(resp.Body)
    if err != nil {
        log.Println(err)
        return
    }
    
    // Deserialize response
    var res pb.AdProfResponse
    if err := proto.Unmarshal(body, &res); err != nil {
        log.Println(err)
        return
    }
    
    log.Println(&res)
}

This implementation demonstrates:

1. Protobuf serialization/deserialization
2. HTTP communication with protobuf payloads
3. Proper error handling and resource cleanup

The application/x-protobuf Content-Type header ensures servers correctly interpret the binary payload. For production use, consider adding request timeouts and retry mechanisms.