From 3cfb185dded952414d37588a6831232f38137f3d Mon Sep 17 00:00:00 2001 From: evo <446796145@qq.com> Date: Wed, 1 Apr 2026 23:11:54 +0800 Subject: [PATCH] =?UTF-8?q?feat=EF=BC=9A=E5=A2=9E=E5=8A=A0=E5=8F=AF?= =?UTF-8?q?=E8=A7=82=E6=B5=8B=E6=80=A7=E7=9B=91=E5=90=AC=E5=99=A8=20?= =?UTF-8?q?=E8=B0=83=E6=95=B4=E6=80=9D=E8=80=83=E8=BE=93=E5=87=BA=E7=9B=91?= =?UTF-8?q?=E5=90=AC=E6=97=A5=E5=BF=97?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- .../ruoyi/observability/MyAgentListener.java | 15 ++++++ .../service/chat/impl/ChatServiceFacade.java | 53 ++++++++++++++++--- 2 files changed, 62 insertions(+), 6 deletions(-) diff --git a/ruoyi-modules/ruoyi-chat/src/main/java/org/ruoyi/observability/MyAgentListener.java b/ruoyi-modules/ruoyi-chat/src/main/java/org/ruoyi/observability/MyAgentListener.java index 7a548788..96223687 100644 --- a/ruoyi-modules/ruoyi-chat/src/main/java/org/ruoyi/observability/MyAgentListener.java +++ b/ruoyi-modules/ruoyi-chat/src/main/java/org/ruoyi/observability/MyAgentListener.java @@ -10,6 +10,7 @@ import dev.langchain4j.service.tool.ToolExecution; import lombok.extern.slf4j.Slf4j; import java.util.Map; +import java.util.concurrent.atomic.AtomicReference; /** * 自定义的 AgentListener 的监听器。 @@ -26,6 +27,13 @@ import java.util.Map; @Slf4j public class MyAgentListener implements dev.langchain4j.agentic.observability.AgentListener { + /** 最终捕获到的思考结果（主 Agent 完成后写入，供外部获取） */ + private final AtomicReference sharedOutputRef = new AtomicReference<>(); + + public String getCapturedResult() { + return sharedOutputRef.get(); + } + // ==================== Agent 调用生命周期 ==================== @Override @@ -72,12 +80,19 @@ public class MyAgentListener implements dev.langchain4j.agentic.observability.Ag AgentInstance agent = agentResponse.agent(); Map inputs = agentResponse.inputs(); Object output = agentResponse.output(); + String outputStr = output != null ? output.toString() : ""; log.info("【Agent调用后】Agent名称: {}", agent.name()); log.info("【Agent调用后】Agent ID: {}", agent.agentId()); log.info("【Agent调用后】Agent输入参数: {}", inputs); log.info("【Agent调用后】Agent输出结果: {}", output); log.info("【Agent调用后】是否为叶子节点: {}", agent.leaf()); + + // 捕获主 Agent 的最终输出，供外部获取 + if ("invoke".equals(agent.agentId()) && !outputStr.isEmpty()) { + sharedOutputRef.set(outputStr); + log.info("【Agent调用后】已捕获主Agent输出: {}", outputStr); + } } @Override diff --git a/ruoyi-modules/ruoyi-chat/src/main/java/org/ruoyi/service/chat/impl/ChatServiceFacade.java b/ruoyi-modules/ruoyi-chat/src/main/java/org/ruoyi/service/chat/impl/ChatServiceFacade.java index c0292871..62b90728 100644 --- a/ruoyi-modules/ruoyi-chat/src/main/java/org/ruoyi/service/chat/impl/ChatServiceFacade.java +++ b/ruoyi-modules/ruoyi-chat/src/main/java/org/ruoyi/service/chat/impl/ChatServiceFacade.java @@ -122,7 +122,7 @@ public class ChatServiceFacade implements IChatService { List contextMessages = buildContextMessages(chatRequest); // 3. 处理特殊聊天模式（工作流、人机交互恢复、思考模式） - SseEmitter specialResult = handleSpecialChatModes(chatRequest, contextMessages, chatModelVo, emitter); + SseEmitter specialResult = handleSpecialChatModes(chatRequest, contextMessages, chatModelVo, emitter, userId, tokenValue); if (specialResult != null) { return specialResult; } @@ -151,10 +151,13 @@ public class ChatServiceFacade implements IChatService { * @param contextMessages 上下文消息列表（可能被修改） * @param chatModelVo 聊天模型配置 * @param emitter SSE发射器 + * @param userId 用户ID + * @param tokenValue 会话令牌 * @return 如果需要提前返回则返回SseEmitter，否则返回null */ private SseEmitter handleSpecialChatModes(ChatRequest chatRequest, List contextMessages, - ChatModelVo chatModelVo, SseEmitter emitter) { + ChatModelVo chatModelVo, SseEmitter emitter, + Long userId, String tokenValue) { // 处理工作流对话 if (chatRequest.getEnableWorkFlow()) { log.info("处理工作流对话,会话: {}", chatRequest.getSessionId()); @@ -193,7 +196,16 @@ public class ChatServiceFacade implements IChatService { // 处理思考模式 if (chatRequest.getEnableThinking()) { - handleThinkingMode(chatRequest, contextMessages, chatModelVo); + String thinkingResult = handleThinkingMode(chatRequest, contextMessages, chatModelVo, userId, tokenValue); + // 思考模式产生了有效结果，通过 SSE 发送给前端后结束 + if (thinkingResult != null && !thinkingResult.isBlank()) { + SseMessageUtils.sendDone(userId); + SseMessageUtils.completeConnection(userId, tokenValue); + log.info("思考模式完成，结果已发送: {}", thinkingResult); + return emitter; + } + // 思考结果为空，继续走普通聊天流程 + log.warn("思考模式未产生有效结果，继续普通聊天"); } return null; @@ -205,8 +217,12 @@ public class ChatServiceFacade implements IChatService { * @param chatRequest 聊天请求 * @param contextMessages 上下文消息列表 * @param chatModelVo 聊天模型配置 + * @param userId 用户ID + * @param tokenValue 会话令牌 + * @return 思考结果字符串，如果无结果则返回空字符串 */ - private void handleThinkingMode(ChatRequest chatRequest, List contextMessages, ChatModelVo chatModelVo) { + private String handleThinkingMode(ChatRequest chatRequest, List contextMessages, + ChatModelVo chatModelVo, Long userId, String tokenValue) { // 步骤1: 配置MCP传输层 - 连接到bing-cn-mcp服务器 McpTransport transport = new StdioMcpTransport.Builder() .command(List.of("C:\\Program Files\\nodejs\\npx.cmd", "-y", "bing-cn-mcp")) @@ -263,13 +279,38 @@ public class ChatServiceFacade implements IChatService { // 构建监督者Agent SupervisorAgent supervisor = AgenticServices.supervisorBuilder() .chatModel(plannerModel) - .subAgents(sqlAgent, chartGenerationAgent) + .subAgents(sqlAgent, searchAgent, chartGenerationAgent) .responseStrategy(SupervisorResponseStrategy.LAST) .listener(new MyAgentListener()) .build(); + // 调用 supervisor String invoke = supervisor.invoke(chatRequest.getContent()); - contextMessages.add(AiMessage.from(invoke)); + log.info("【思考模式】supervisor.invoke() 返回: {}", invoke); + + // 如果有有效结果，通过 SSE 发送给前端并保存到数据库 + if (invoke != null && !invoke.isBlank()) { + try { + // 通过 SSE 实时发送思考结果 + SseMessageUtils.sendContent(userId, invoke); + log.info("【思考模式】结果已发送至SSE: {}", invoke); + + // 保存用户消息 + chatMessageService.saveChatMessage(userId, chatRequest.getSessionId(), + chatRequest.getContent(), RoleType.USER.getName(), chatRequest.getModel()); + + // 保存助手思考结果消息 + chatMessageService.saveChatMessage(userId, chatRequest.getSessionId(), + invoke, RoleType.ASSISTANT.getName(), chatRequest.getModel()); + + // 将思考结果添加到上下文，供后续流程使用（如果需要） + contextMessages.add(AiMessage.from(invoke)); + } catch (Exception e) { + log.error("【思考模式】发送结果或保存消息失败: {}", e.getMessage(), e); + } + } + + return invoke != null ? invoke : ""; } /**