AI对话优化

ask-data-ai/ask-data-ai-biz/src/main/java/com/ask/config/CommonConfiguration.java

@@ -2,11 +2,13 @@ package com.ask.config;
 
 import org.springframework.ai.chat.client.ChatClient;
 import org.springframework.ai.chat.client.advisor.MessageChatMemoryAdvisor;
+import org.springframework.ai.chat.client.advisor.PromptChatMemoryAdvisor;
 import org.springframework.ai.chat.client.advisor.api.Advisor;
 import org.springframework.ai.chat.memory.ChatMemory;
 import org.springframework.ai.chat.memory.ChatMemoryRepository;
 import org.springframework.ai.chat.memory.MessageWindowChatMemory;
 import org.springframework.ai.chat.memory.repository.jdbc.JdbcChatMemoryRepository;
+import org.springframework.ai.chat.prompt.PromptTemplate;
 import org.springframework.ai.openai.OpenAiChatModel;
 import org.springframework.ai.rag.advisor.RetrievalAugmentationAdvisor;
 import org.springframework.ai.rag.generation.augmentation.ContextualQueryAugmenter;
@@ -24,41 +26,48 @@ public class CommonConfiguration {
 
 
     @Bean
-    public ChatMemory chatMemory (JdbcTemplate jdbcTemplate,PostgresChatMemoryDialect postgresChatMemoryDialect) {
-		ChatMemoryRepository chatMemoryRepository = JdbcChatMemoryRepository.builder()
-				.jdbcTemplate(jdbcTemplate)
-				.dialect(postgresChatMemoryDialect)
-				.build();
+    public ChatMemory chatMemory(JdbcTemplate jdbcTemplate, PostgresChatMemoryDialect postgresChatMemoryDialect) {
+        ChatMemoryRepository chatMemoryRepository = JdbcChatMemoryRepository.builder()
+                .jdbcTemplate(jdbcTemplate)
+                .dialect(postgresChatMemoryDialect)
+                .build();
 
-		return MessageWindowChatMemory.builder()
-				.chatMemoryRepository(chatMemoryRepository)
-				.maxMessages(10)
-				.build();
+        return MessageWindowChatMemory.builder()
+                .chatMemoryRepository(chatMemoryRepository)
+                .maxMessages(5)
+                .build();
     }
 
     @Bean
     public ChatClient chatClient(OpenAiChatModel model, ChatMemory chatMemory) {
-        List<Advisor> advisors = new ArrayList<>();
-        Advisor messageChatMemoryAdvisor =MessageChatMemoryAdvisor.builder(chatMemory).build();
-        advisors.add(messageChatMemoryAdvisor);
         return ChatClient.builder(model)
-                .defaultAdvisors(advisors)
-                .defaultAdvisors().build();
+                .defaultAdvisors()
+                .build();
+    }
+    @Bean
+    public PromptChatMemoryAdvisor promptChatMemoryAdvisor(ChatMemory chatMemory){
+        return PromptChatMemoryAdvisor.builder(chatMemory).build();
     }
 
-	@Bean
-	public RetrievalAugmentationAdvisor retrievalAugmentationAdvisor(VectorStore vectorStore) {
+    @Bean
+    public MessageChatMemoryAdvisor messageChatMemoryAdvisor(ChatMemory chatMemory){
+        return MessageChatMemoryAdvisor.builder(chatMemory).build();
+    }
 
-		return RetrievalAugmentationAdvisor.builder()
-				.documentRetriever(VectorStoreDocumentRetriever.builder()
-						.similarityThreshold(0.60)
-						.topK(5)
-						.vectorStore(vectorStore)
-						.build())
-				.queryAugmenter(ContextualQueryAugmenter.builder()
-						.allowEmptyContext(true)
-						.build())
-				.build();
-	}
+    @Bean
+    public RetrievalAugmentationAdvisor retrievalAugmentationAdvisor(VectorStore vectorStore) {
+
+        return RetrievalAugmentationAdvisor.builder()
+                .documentRetriever(VectorStoreDocumentRetriever.builder()
+                        .similarityThreshold(0.70)
+                        .topK(5)
+                        .vectorStore(vectorStore)
+                        .build())
+                .documentPostProcessors(new MyDocumentPostProcessor())
+                .queryAugmenter(ContextualQueryAugmenter.builder()
+                        .allowEmptyContext(true)
+                        .build())
+                .build();
+    }
 
 }

ask-data-ai/ask-data-ai-biz/src/main/java/com/ask/config/CustomPromptAdvisor.java

+//package com.ask.config;
+//
+//import org.springframework.ai.chat.client.advisor.api.Advisor;
+//import org.springframework.ai.chat.messages.UserMessage;
+//import org.springframework.ai.chat.prompt.Prompt;
+//
+//public class CustomPromptAdvisor implements Advisor {
+//
+//
+////        // 拼接自定义prompt
+////        String finalPrompt = String.format(
+////                "【系统提示】\n%s\n\n【知识库内容】\n%s\n\n【历史对话】\n%s\n\n【用户问题】\n%s\n\n请结合以上内容，专业、简明、准确地用中文回答用户问题。如果知识库内容无法回答，请如实说明。",
+////                instructions, this.context, this.memory, question
+////        );
+//}
\ No newline at end of file

ask-data-ai/ask-data-ai-biz/src/main/java/com/ask/config/MyDocumentPostProcessor.java

+package com.ask.config;
+
+import lombok.extern.slf4j.Slf4j;
+import org.springframework.ai.document.Document;
+import org.springframework.ai.rag.Query;
+import org.springframework.ai.rag.postretrieval.document.DocumentPostProcessor;
+
+import java.util.ArrayList;
+import java.util.List;
+import java.util.stream.Collectors;
+
+@Slf4j
+public class MyDocumentPostProcessor implements DocumentPostProcessor {
+
+
+    @Override
+    public List<Document> process(Query query, List<Document> documents) {
+        log.info("问题 {} 召回 {} 个文档",query,documents.size());
+
+        return documents.stream()
+                .collect(Collectors.collectingAndThen(
+                        Collectors.toMap(d -> d.getMetadata().get("id"), d -> d, (d1, d2) -> d1),
+                        m -> new ArrayList<>(m.values())
+                ));
+    }
+}
\ No newline at end of file

ask-data-ai/ask-data-ai-biz/src/main/java/com/ask/controller/ChatController.java

@@ -10,6 +10,8 @@ import io.swagger.v3.oas.annotations.tags.Tag;
 import lombok.RequiredArgsConstructor;
 import lombok.extern.slf4j.Slf4j;
 import org.springframework.ai.chat.client.ChatClient;
+import org.springframework.ai.chat.client.advisor.MessageChatMemoryAdvisor;
+import org.springframework.ai.chat.client.advisor.PromptChatMemoryAdvisor;
 import org.springframework.ai.chat.memory.ChatMemory;
 import org.springframework.ai.chat.messages.Message;
 import org.springframework.ai.chat.messages.SystemMessage;
@@ -40,6 +42,9 @@ public class ChatController {
 
 	private final RetrievalAugmentationAdvisor retrievalAugmentationAdvisor;
 
+	private final MessageChatMemoryAdvisor messageChatMemoryAdvisor;
+
+	private final PromptChatMemoryAdvisor promptChatMemoryAdvisor;
     /**
      * 获取会话ID
      * @return 新的会话ID
@@ -87,32 +92,34 @@ public class ChatController {
         // 创建提示，包含系统消息和用户消息
         Prompt prompt = new Prompt(Arrays.asList(systemMessage, userMessage));
         // 使用修改后的提示获取响应
-        return chatClient.prompt(prompt).advisors(a -> a.param(ChatMemory.CONVERSATION_ID, conversationId)).stream().content();
+        return chatClient.prompt(prompt).advisors(messageChatMemoryAdvisor).advisors(a -> a.param(ChatMemory.CONVERSATION_ID, conversationId)).stream().content();
     }
 
 	/**
-	 * 最基本的AI流式输出对话
+	 * 知识库对话
 	 *
 	 * * @param message
 	 * @return
 	 */
 	@Operation(summary = "知识库对话", description = "知识库对话")
-	@GetMapping(value = "/rag/chat", produces = "application/stream+json")
-	public Flux<ChatResponse> ragChat(String message, String conversationId) {
+	@GetMapping(value = "/rag/chat",  produces = "text/html;charset=utf-8")
+	public Flux<String> ragChat(String message, String conversationId) {
 		// 创建系统消息，告诉大模型只返回工具名和参数
-		Message systemMessage = new SystemMessage("你是一个AI客服助手。请严格按照以下格式回答每个问题：");
-
-		Message userMessage = new UserMessage(message);
-		// 创建提示，包含系统消息和用户消息
-		Prompt prompt = new Prompt(Arrays.asList(systemMessage, userMessage));
+//		Message systemMessage = new SystemMessage("你是一个AI客服助手。请严格按照以下格式回答每个问题：");
+//
+//		Message userMessage = new UserMessage(message);
+//		// 创建提示，包含系统消息和用户消息
+//		Prompt prompt = new Prompt(Arrays.asList(systemMessage, userMessage));
 		// 使用修改后的提示获取响应
 		FilterExpressionBuilder builder = new FilterExpressionBuilder();
 		Filter.Expression filter = builder.eq("isEnabled",1).build();
+//		List<Document> documents =  vectorStore.similaritySearch(SearchRequest.builder().query(message).filterExpression(filter).similarityThreshold(0.75).topK(5).build());
 
-		return chatClient.prompt(prompt)
+		return chatClient.prompt(message)
 				.advisors(a -> a.param(ChatMemory.CONVERSATION_ID, conversationId))
+				.advisors(promptChatMemoryAdvisor)  //会话记忆
 				.advisors(a -> a.param(VectorStoreDocumentRetriever.FILTER_EXPRESSION, filter))
-				.advisors(retrievalAugmentationAdvisor)
-				.stream().chatResponse();
+				.advisors(retrievalAugmentationAdvisor) //知识库召回
+				.stream().content();
 	}
 }
\ No newline at end of file

ask-data-ai/ask-data-ai-biz/src/main/java/com/ask/controller/KnowledgeBaseController.java

@@ -188,6 +188,8 @@ public class KnowledgeBaseController {
                             new Filter.Value(1)
                     )
             );
+
+
             // 执行向量搜索
             List<Document> searchResults = vectorStore.similaritySearch(SearchRequest.builder().query(content).filterExpression(filterExpression).similarityThreshold(similarityThreshold).topK(topK).build());
 

ask-data-ai/ask-data-ai-boot/target/classes/application.yml

-server:
-  port: 9999
-  servlet:
-    context-path: /admin  # 项目访问路径
-spring:
-  application:
-    name: ask-data-ai
-  # 文件上传配置
-  servlet:
-    multipart:
-      enabled: true
-      max-file-size: 100MB      # 单个文件最大大小
-      max-request-size: 500MB   # 请求最大大小
-      file-size-threshold: 0    # 文件写入磁盘的阈值
-  datasource:
-    url: jdbc:postgresql://81.70.183.25:25432/ask_data_ai_db
-    username: postgres
-    password: postgres123
-    driver-class-name: org.postgresql.Driver
-  ai:
-    vectorstore:
-      pgvector:
-        index-type: HNSW
-        distance-type: COSINE_DISTANCE
-        dimensions: 1024
-        max-document-batch-size: 10000 # Optional: Maximum number of documents per batch
-        schema-name: public
-        table-name: ask_vector_store
-    chat:
-      memory:
-        repository:
-          jdbc:
-            initialize-schema: never  # 开发环境可以使用 always，方便调试
-            platform: postgresql
-    openai:
-      base-url: https://dashscope.aliyuncs.com/compatible-mode
-      api-key: sk-ae96ff281ff644c992843c64a711a950
-      chat:
-        options:
-          model: qwen-plus
-      embedding:
-        base-url: https://dashscope.aliyuncs.com/compatible-mode
-        api-key: sk-ae96ff281ff644c992843c64a711a950
-        options:
-          model: text-embedding-v4
-
-mybatis-plus:
-  mapper-locations: classpath*:/mapper/*Mapper.xml # mapper文件位置
-  global-config:
-    banner: false # 是否打印 mybatis-plus banner
-    db-config:
-      table-prefix: ask_ # 表名前缀
-      logic-delete-field: delFlag # 全局逻辑删除字段
-      logic-delete-value: 1 # 逻辑已删除值（默认为 1）
-      logic-not-delete-value: 0 # 逻辑未删除值（默认为 0）
-      insert-strategy: not_null # 插入策略
-      update-strategy: not_null # 更新策略
-      where-strategy: not_null # 查询策略
-      id-type: assign_uuid # 主键策略
-    configuration:
-      map-underscore-to-camel-case: true # 开启驼峰命名
-      cache-enabled: false # 是否开启二级缓存
-      call-setters-on-nulls: true # 是否在字段为 null 时调用 setter 方法
-      jdbc-type-for-null: 'null' # 指定当结果集中值为 null 的时候如何处理
-      log-impl: org.apache.ibatis.logging.stdout.StdOutImpl # 日志实现
-  
-  # 启动时处理器配置
-  startup:
-    # 是否启用启动时向量化处理器
-    enabled: true
-    # 启动延迟时间（秒），避免启动时立即执行
-    delay_seconds: 30
-    # 每次提交任务间的延迟（毫秒）
-    task_interval_ms: 100
-    # 最大处理文档数量限制
-    max_documents: 100
-
-# springdoc-openapi项目配置
-springdoc:
-  swagger-ui:
-    enabled: true  # 开启swagger-ui
-    path: /swagger-ui.html  # 配置访问路径
-  api-docs:
-    enabled: true   # 开启api-docs
-    path: /v3/api-docs  # 配置访问路径
-  group-configs:
-    - group: 'default'
-      paths-to-match: '/**'
-      packages-to-scan: com.ask
-  default-produces-media-type: application/json
-  default-consumes-media-type: application/json
-
-# knife4j的增强配置，不需要增强可以不配
-knife4j:
-  enable: true
-  setting:
-    language: zh_cn
-    enable-swagger-models: true
-    enable-document-manage: true
-    swagger-model-name: 实体类列表
-    enable-version: false
-    enable-reload-cache-parameter: false
-    enable-after-script: false
-    enable-filter-multipart-api-method-type: POST
-    enable-filter-multipart-apis: false
-    enable-request-cache: true
-    enable-host: false
-    enable-host-text:
-
-
-# swagger配置
-swagger:
-  enabled: true
-  title: Ask Data AI接口文档
-  gateway: http://localhost:${server.port}/admin
-  token-url: ${swagger.gateway}/oauth2/token
-  scope: server
-
-# 日志配置
-logging:
-  level:
-    root: INFO
-    com.ask: DEBUG
-    org.springframework.ai: INFO
-    org.postgresql: WARN
-  pattern:
-    console: "%d{HH:mm:ss.SSS} [%thread] %-5level %logger{36} - %msg%n"
-
-# 本地文件系统
-file:
-  local:
-    enable: true
-    base-path: /app/upFiles

ask-data-ai/ask-data-ai-boot/target/classes/logback-spring.xml

-<?xml version="1.0" encoding="UTF-8"?>
-<configuration debug="false">
-	<!--定义日志文件的存储地址 -->
-	<property name="LOG_HOME" value="logs" />
-
-	<!--<property name="COLOR_PATTERN" value="%black(%contextName-) %red(%d{yyyy-MM-dd HH:mm:ss}) %green([%thread]) %highlight(%-5level) %boldMagenta( %replace(%caller{1}){'\t|Caller.{1}0|\r\n', ''})- %gray(%msg%xEx%n)" />-->
-	<!-- 控制台输出 -->
-	<appender name="STDOUT" class="ch.qos.logback.core.ConsoleAppender">
-		<encoder class="ch.qos.logback.classic.encoder.PatternLayoutEncoder">
-			<!--格式化输出：%d表示日期，%thread表示线程名，%-5level：级别从左显示5个字符宽度%msg：日志消息，%n是换行符
-			<pattern>%d{yyyy-MM-dd HH:mm:ss.SSS} [%thread] %-5level %logger{50}:%L - %msg%n</pattern>-->
-			<pattern>%d{yyyy-MM-dd HH:mm:ss.SSS} [%thread] %highlight(%-5level) %cyan(%logger{50}:%L) - %msg%n</pattern>
-		</encoder>
-	</appender>
-
-	<!-- 按照每天生成日志文件 -->
-	<appender name="FILE" class="ch.qos.logback.core.rolling.RollingFileAppender">
-		<rollingPolicy class="ch.qos.logback.core.rolling.SizeAndTimeBasedRollingPolicy">
-			<!--日志文件输出的文件名 -->
-			<FileNamePattern>${LOG_HOME}/ask-%d{yyyy-MM-dd}.%i.log</FileNamePattern>
-			<!--日志文件保留天数 -->
-			<MaxHistory>30</MaxHistory>
-			<maxFileSize>10MB</maxFileSize>
-		</rollingPolicy>
-		<encoder class="ch.qos.logback.classic.encoder.PatternLayoutEncoder">
-			<!--格式化输出：%d表示日期，%thread表示线程名，%-5level：级别从左显示5个字符宽度%msg：日志消息，%n是换行符 -->
-			<pattern>%d{yyyy-MM-dd HH:mm:ss.SSS} [%thread] %-5level %logger{50}:%L - %msg%n</pattern>
-		</encoder>
-	</appender>
-	<!-- 每天生成一个html格式的日志结束 -->
-
-	<!--myibatis log configure -->
-	<logger name="com.apache.ibatis" level="TRACE" />
-	<logger name="java.sql.Connection" level="DEBUG" />
-	<logger name="java.sql.Statement" level="DEBUG" />
-	<logger name="java.sql.PreparedStatement" level="DEBUG" />
-
-	<!-- 日志输出级别 -->
-	<root level="INFO">
-		<appender-ref ref="STDOUT" />
-		<appender-ref ref="FILE" />
-	</root>
-	<logger name="okhttp3" level="ERROR"/>
-</configuration>
\ No newline at end of file

ask-data-ai/ask-data-ai-common/target/classes/openapi-config.yaml

-# swagger 配置
-swagger:
-  enabled: true
-  title: Pig Swagger API
-  gateway: http://${GATEWAY-HOST:127.0.0.1}:${GATEWAY-PORT:9999}/admin
-  token-url: ${swagger.gateway}/admin/oauth2/token
-  scope: server