SpringBoot接入豆包AI 数据流式返回前端

1. maven引入依赖

		<dependency>
			<groupId>com.volcengine</groupId>
			<artifactId>volcengine-java-sdk-ark-runtime</artifactId>
			<version>LATEST</version>
		</dependency>

2. 代码

AI实现类

package com.hpp.common.AIChat;

import java.io.IOException;
import java.util.ArrayList;
import java.util.List;

import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import org.springframework.beans.factory.annotation.Autowired;
import org.springframework.stereotype.Component;
import org.springframework.web.servlet.mvc.method.annotation.SseEmitter;

import com.google.gson.Gson;
import com.hpp.common.redisson.RedisCache;
import com.volcengine.ark.runtime.model.completion.chat.ChatCompletionRequest;
import com.volcengine.ark.runtime.model.completion.chat.ChatMessage;
import com.volcengine.ark.runtime.model.completion.chat.ChatMessageRole;
import com.volcengine.ark.runtime.service.ArkService;

@Component
public class AIChatDouBao {
	private static final Logger LOGGER = LoggerFactory.getLogger(AIChatDouBao.class);
	@Autowired
	private RedisCache redisCache;
	private Gson gson = new Gson();
	private String apiKey = "xxxxxxxxxxxxxxxx";
	
	public SseEmitter conversation(String question) {
		// 创建ArkService实例
		ArkService arkService = ArkService.builder().apiKey(apiKey).build();
		// 初始化消息列表
		List<ChatMessage> chatMessages = new ArrayList<>();
		// 创建用户消息
		ChatMessage userMessage = ChatMessage.builder().role(ChatMessageRole.USER) // 设置消息角色为用户
				.content(question) // 设置消息内容
				.build();
		// 将用户消息添加到消息列表
		chatMessages.add(userMessage);
		// 创建聊天完成请求
		ChatCompletionRequest chatCompletionRequest = ChatCompletionRequest.builder().model("ep-xxxxxxxxxxxx")// 需要替换为您的推理接入点ID
				.messages(chatMessages) // 设置消息列表
				.build();
		// 发送聊天完成请求并打印响应
		// 获取响应并打印每个选择的消息内容
		SseEmitter emitter = new SseEmitter();
		try {
			new Thread(() -> {
				arkService.streamChatCompletion(chatCompletionRequest).doOnError(Throwable::printStackTrace)
						.doFinally(() -> {
							LOGGER.info("完成");
							emitter.complete();
						}).blockingForEach(choice -> {
							if (choice.getChoices().size() > 0) {
								ChatMessage message = choice.getChoices().get(0).getMessage();
								// 判断是否触发深度推理，触发则打印模型输出的思维链内容
								if (message.getReasoningContent() != null && !message.getReasoningContent().isEmpty()) {
									LOGGER.info(message.getReasoningContent());
									emitter.send(message.getReasoningContent());
								}
								// 打印模型输出的回答内容
								emitter.send(message.getContent().toString());
								LOGGER.info(message.getContent().toString());
							}
						});

			}).start();
		} catch (Exception e) {
			LOGGER.error("ai回复异常", e);
			emitter.completeWithError(e); // 发生错误时，通知客户端
		} finally {

		}
		return emitter;
	}

	public String getAccessToken() throws IOException {
		return "";
	}

}

controller层

@RestController
@RequestMapping(value = "chat")
public class ChatController {
	private static final Logger LOGGER = LoggerFactory.getLogger(ChatController.class);
	@Autowired
	private AIChatDouBao aiChatDouBao;

	@GetMapping("/event-stream")
	public SseEmitter streamEvents(HttpServletRequest req, @RequestParam String question) {
		return aiChatDouBao.conversation(question);
	}

}