现在AI的接口由于生成内容比较慢都是采用的流式输出的方式。这里将模拟一下流式输出。
后端接口
import cn.hutool.json.JSONUtil; import org.springframework.http.HttpHeaders; import org.springframework.http.HttpStatus; import org.springframework.http.MediaType; import org.springframework.http.ResponseEntity; import org.springframework.web.bind.annotation.*; import org.springframework.web.servlet.mvc.method.annotation.StreamingResponseBody; import java.io.PrintWriter; import java.util.HashMap; import java.util.Map; @RestController @RequestMapping("/test") public class TestController { @PostMapping("/stream") public ResponseEntitystreamData() { Map map = new HashMap<>(); map.put("content", "内容"); StreamingResponseBody responseBody = outputStream -> { try (PrintWriter writer = new PrintWriter(outputStream)) { for (int i = 0; i < 10; i++) { map.put("content", "内容:" + i); writer.println(JSONUtil.toJsonStr(map)); writer.flush(); // 模拟一些延迟 Thread.sleep(500); } } catch (InterruptedException e) { e.printStackTrace(); } }; HttpHeaders headers = new HttpHeaders(); headers.setContentType(MediaType.TEXT_PLAIN); // 指示这是一个流式响应 headers.setContentLength(-1L); return new ResponseEntity<>(responseBody, headers, HttpStatus.OK); } }
传统非流式-前端
url = 'http://127.0.0.1:8080/test/stream' async function getResp(){ const startTime = performance.now(); console.log("非流式输出") const resp = await fetch(url,{ method: 'POST', headers: { 'Content-Type': 'application/json' }, body: JSON.stringify({ content: '讲个笑话' }) }); const msg = await resp.text(); console.log(msg) const endTime = performance.now(); console.log(`执行耗时: ${endTime - startTime} ms`); } getResp()
测试结果如下,程序会等待所有内容都返回了,才输出内容
流式-前端
async function getRespStream(){ console.log("流式输出") const resp = await fetch(url,{ method: 'POST', headers: { 'Content-Type': 'application/json' }, body: JSON.stringify({ content: '讲个笑话' }) }); const reader = resp.body.getReader(); while(1){ // value 是类型化数组 const textDecoder = new TextDecoder() const {done,value} = await reader.read(); if(done){ break } const str = textDecoder.decode(value) console.log(str) } } getRespStream()
可以看到后端只要输出一段内容前端就会打印一段内容。