From f0d0222c339d8a61f0076763ea05e67907f60169 Mon Sep 17 00:00:00 2001
From: malog <2153315236@qq.com>
Date: Mon, 1 Dec 2025 15:11:53 +0800
Subject: [PATCH 1/3] =?UTF-8?q?feat:=20=E8=BF=81=E7=A7=BB=E7=94=A8?=
=?UTF-8?q?=E6=88=B7=E6=95=B0=E6=8D=AE=E8=B7=9F=E8=B8=AA=E5=88=86=E6=9E=90?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
---
pom.xml | 18 +-
refine-app/pom.xml | 6 +-
.../java/com/achobeta/config/RagConfig.java | 161 +----
.../achobeta/config/RagConfigProperties.java | 18 +-
.../com/achobeta/config/WeaviateConfig.java | 86 +++
.../ocr/service/impl/DefaultOcrService.java | 30 +-
.../service/impl/LearningAnalysisService.java | 12 +-
.../impl/LearningDynamicsServiceImpl.java | 179 +++++-
.../domain/user/event/UserBehaviorEvent.java | 32 +
.../domain/user/event/UserMistakeEvent.java | 14 +
.../domain/user/event/UserReviewEvent.java | 14 +
.../domain/user/event/UserUploadEvent.java | 14 +
refine-infrastructure/pom.xml | 4 +
.../repository/LearningDataRepository.java | 338 +++++++++--
.../repository/WeaviateSchemaInitializer.java | 156 +++++
.../repository/WeaviateVectorRepository.java | 573 ++++++++++++++++++
.../event/UserBehaviorEventListener.java | 162 +++++
.../http/LearningAnalysisController.java | 2 +
.../achobeta/trigger/http/OcrController.java | 2 -
19 files changed, 1573 insertions(+), 248 deletions(-)
create mode 100644 refine-app/src/main/java/com/achobeta/config/WeaviateConfig.java
create mode 100644 refine-domain/src/main/java/com/achobeta/domain/user/event/UserBehaviorEvent.java
create mode 100644 refine-domain/src/main/java/com/achobeta/domain/user/event/UserMistakeEvent.java
create mode 100644 refine-domain/src/main/java/com/achobeta/domain/user/event/UserReviewEvent.java
create mode 100644 refine-domain/src/main/java/com/achobeta/domain/user/event/UserUploadEvent.java
create mode 100644 refine-infrastructure/src/main/java/com/achobeta/infrastructure/adapter/repository/WeaviateSchemaInitializer.java
create mode 100644 refine-infrastructure/src/main/java/com/achobeta/infrastructure/adapter/repository/WeaviateVectorRepository.java
create mode 100644 refine-trigger/src/main/java/com/achobeta/trigger/event/UserBehaviorEventListener.java
diff --git a/pom.xml b/pom.xml
index bdebeaf..29ce182 100644
--- a/pom.xml
+++ b/pom.xml
@@ -76,11 +76,11 @@
postgresql
42.7.1
-
+
- com.pgvector
- pgvector
- 0.1.4
+ io.weaviate
+ client
+ 4.8.0
@@ -181,7 +181,7 @@
org.apache.commons
commons-lang3
- 3.9
+ 3.12.0
com.google.guava
@@ -235,14 +235,6 @@
langchain4j-reactor
${langchain4j.version}
-
-
- dev.langchain4j
- langchain4j-pgvector
- ${langchain4j.version}
-
-
-
com.achobeta
diff --git a/refine-app/pom.xml b/refine-app/pom.xml
index 94e2d03..02bf62e 100644
--- a/refine-app/pom.xml
+++ b/refine-app/pom.xml
@@ -113,10 +113,10 @@
dev.langchain4j
langchain4j
-
+
- dev.langchain4j
- langchain4j-pgvector
+ io.weaviate
+ client
diff --git a/refine-app/src/main/java/com/achobeta/config/RagConfig.java b/refine-app/src/main/java/com/achobeta/config/RagConfig.java
index 9f7d305..6cd9ebf 100644
--- a/refine-app/src/main/java/com/achobeta/config/RagConfig.java
+++ b/refine-app/src/main/java/com/achobeta/config/RagConfig.java
@@ -15,7 +15,6 @@
import dev.langchain4j.store.embedding.EmbeddingStoreIngestor;
import dev.langchain4j.store.embedding.filter.Filter;
import dev.langchain4j.store.embedding.filter.MetadataFilterBuilder;
-import dev.langchain4j.store.embedding.pgvector.PgVectorEmbeddingStore;
import jakarta.annotation.PostConstruct;
import jakarta.annotation.Resource;
import lombok.extern.slf4j.Slf4j;
@@ -24,10 +23,6 @@
import org.springframework.context.annotation.Configuration;
import java.io.File;
-import java.sql.Connection;
-import java.sql.DriverManager;
-import java.sql.PreparedStatement;
-import java.sql.SQLException;
import java.util.List;
import java.util.stream.Collectors;
@@ -46,21 +41,22 @@ public class RagConfig {
@PostConstruct
public void initialize() {
- this.embeddingStore = PgVectorEmbeddingStore.builder()
- .host(ragConfigProperties.getHost())
- .port(ragConfigProperties.getPort())
- .database(ragConfigProperties.getDatabase())
- .user(ragConfigProperties.getUser())
- .password(ragConfigProperties.getPassword())
- .table("knowledge_embeddings")
- .dimension(qwenEmbeddingModel.dimension())
- .build();
- log.info("EmbeddingStore初始化完成");
+ try {
+ log.info("开始初始化Weaviate EmbeddingStore");
+
+ // 暂时跳过EmbeddingStore初始化，因为我们使用自定义的向量存储
+ log.info("RagConfig初始化完成，使用自定义向量存储服务");
+
+ } catch (Exception e) {
+ log.error("RagConfig初始化失败", e);
+ }
}
@Bean
public EmbeddingStore embeddingStore() {
- return this.embeddingStore;
+ // 返回null，因为我们使用自定义的向量存储服务
+ log.info("使用自定义向量存储服务，不创建EmbeddingStore Bean");
+ return null;
}
@Bean
@@ -75,29 +71,8 @@ public ContentRetriever contentRetriever() {
// 获取目录下的文件（排除子目录，只考虑文件）
File[] files = docDir.listFiles(File::isFile);
if (files != null && files.length > 0) {
- // 先清空向量表的所有旧数据
- truncateEmbeddingTable(); // 全删逻辑（替换原来的按文件名删）
-
- // 加载文档并重新入库
- List documents = FileSystemDocumentLoader.loadDocuments(docPath);
- log.info("待录入文档总数：{} 个，开始分段并生成向量", documents.size());
-
-
- // 文档分段
- DocumentByParagraphSplitter documentSplitter = new DocumentByParagraphSplitter(800, 350);
- // 自定义文档加载器，把文档转换成向量并存储到向量数据库中
- EmbeddingStoreIngestor ingestor = EmbeddingStoreIngestor.builder()
- .documentSplitter(documentSplitter)
- .textSegmentTransformer(textSegment -> TextSegment.from(
- textSegment.metadata().getString("file_name") + "\n" + textSegment.text(),
- textSegment.metadata()
- ))
- .embeddingModel(qwenEmbeddingModel)
- .embeddingStore(embeddingStore)
- .build();
- // 加载文档
- ingestor.ingest(documents);
- log.info("文档全量入库完成! 共加载 {} 个文件，已清空旧数据，当前表中为最新全量数据", files.length);
+ // TODO: 使用Weaviate实现文档向量化和存储
+ log.warn("RagConfig暂未实现Weaviate文档向量化，跳过文档加载");
} else {
log.info("文档目录 {} 下无可用文件，跳过文档加载（表中数据保持不变）", docPath);
}
@@ -109,100 +84,30 @@ public ContentRetriever contentRetriever() {
return new ContentRetriever() {
@Override
public List retrieve(Query query) {
- // 1. 从查询元数据中动态提取业务传入的学科（target）
- Object obj = query.metadata().invocationContext().chatMemoryId();
- if (null == obj) {
- log.warn("RAG检索 - 未传递 file_name 筛选条件，返回空结果");
+ try {
+ log.info("ContentRetriever收到查询请求: {}", query.text());
+
+ // 暂时返回空结果，因为我们主要关注学习行为记录
+ // 后续可以集成Weaviate进行内容检索
+ log.info("ContentRetriever暂时返回空结果，专注于学习行为记录功能");
+ return List.of();
+
+ } catch (Exception e) {
+ log.error("ContentRetriever查询失败", e);
return List.of();
}
- String target = String.valueOf(obj);
- log.info("RAG检索 - 动态过滤 file_name（模糊匹配）: {}", target);
-
- // 2. 生成查询向量
- Embedding embeddedQuery = qwenEmbeddingModel.embed(query.text()).content();
-
- // 3. 用自定义的Filter构建动态过滤条件
- //Filter filter = MetadataFilterBuilder.metadataKey("subject").isEqualTo(subject);
-
- //模糊匹配，自定义的Filter构建动态过滤条件
- Filter filter = MetadataFilterBuilder.metadataKey("file_name").containsString(target);
-
- // 4. 执行带动态过滤的检索
- EmbeddingSearchRequest searchRequest = EmbeddingSearchRequest.builder()
- .queryEmbedding(embeddedQuery)
- .maxResults(5) // 最多返回5条
- .minScore(0.75) // 相关度阈值
- .filter(filter) // 动态过滤条件
- .build();
-
- EmbeddingSearchResult searchResult = embeddingStore.search(searchRequest);
-
- // 5. 转换结果为Content列表
- return searchResult.matches().stream()
- .map(match -> Content.from(match.embedded()))
- .collect(Collectors.toList());
}
};
}
- private void truncateEmbeddingTable() {
- // 方案1：TRUNCATE（推荐）- 快速清空表，不记录单行删除日志，效率极高
- String sql = String.format("TRUNCATE TABLE %s CASCADE", "knowledge_embeddings");
- // CASCADE：若表有外键关联，同时清空关联表（无外键可去掉CASCADE）
-
- // 方案2：DELETE（备选，无TRUNCATE权限时使用）
- // String sql = String.format("DELETE FROM %s", VECTOR_TABLE);
-
- try (
- // 复用配置类的连接信息，避免硬编码
- Connection conn = DriverManager.getConnection(
- String.format("jdbc:postgresql://%s:%d/%s",
- ragConfigProperties.getHost(),
- ragConfigProperties.getPort(),
- ragConfigProperties.getDatabase()),
- ragConfigProperties.getUser(),
- ragConfigProperties.getPassword()
- );
- PreparedStatement pstmt = conn.prepareStatement(sql)
- ) {
- log.info("开始清空向量表：{}", "knowledge_embeddings");
- pstmt.executeUpdate();
- log.info("向量表 {} 清空成功！", "knowledge_embeddings");
-
- } catch (SQLException e) {
- // 若TRUNCATE失败（如无权限），自动降级为DELETE重试（可选逻辑）
- if (e.getMessage().contains("permission denied for table") || e.getMessage().contains("TRUNCATE")) {
- log.warn("TRUNCATE权限不足，尝试用DELETE清空表", e);
- deleteAllEmbeddingData(); // 调用DELETE全删方法
- return;
- }
-
- log.error("清空向量表 {} 失败", "knowledge_embeddings", e);
- throw new RuntimeException("清空旧数据异常，终止文档入库", e);
- }
- }
+ // 注释掉PgVector相关的表操作方法
+ // private void truncateEmbeddingTable() {
+ // // TODO: 使用Weaviate实现数据清理
+ // log.warn("RagConfig暂未实现Weaviate数据清理");
+ // }
- /**
- * 备选方法：用DELETE清空全表（无TRUNCATE权限时触发）
- */
- private void deleteAllEmbeddingData() {
- String sql = String.format("DELETE FROM %s", "knowledge_embeddings");
- try (
- Connection conn = DriverManager.getConnection(
- String.format("jdbc:postgresql://%s:%d/%s",
- ragConfigProperties.getHost(),
- ragConfigProperties.getPort(),
- ragConfigProperties.getDatabase()),
- ragConfigProperties.getUser(),
- ragConfigProperties.getPassword()
- );
- PreparedStatement pstmt = conn.prepareStatement(sql)
- ) {
- int deletedRows = pstmt.executeUpdate();
- log.info("用DELETE清空表 {} 成功，共删除 {} 条记录", "knowledge_embeddings", deletedRows);
- } catch (SQLException e) {
- log.error("DELETE清空表 {} 失败", "knowledge_embeddings", e);
- throw new RuntimeException("删除旧数据异常，终止文档入库", e);
- }
- }
+ // private void deleteAllEmbeddingData() {
+ // // TODO: 使用Weaviate实现数据删除
+ // log.warn("RagConfig暂未实现Weaviate数据删除");
+ // }
}
\ No newline at end of file
diff --git a/refine-app/src/main/java/com/achobeta/config/RagConfigProperties.java b/refine-app/src/main/java/com/achobeta/config/RagConfigProperties.java
index 6c84c24..1701596 100644
--- a/refine-app/src/main/java/com/achobeta/config/RagConfigProperties.java
+++ b/refine-app/src/main/java/com/achobeta/config/RagConfigProperties.java
@@ -4,20 +4,20 @@
import org.springframework.boot.context.properties.ConfigurationProperties;
@Data
-@ConfigurationProperties(prefix = "pgvector.config", ignoreInvalidFields = true)
+@ConfigurationProperties(prefix = "weaviate.config", ignoreInvalidFields = true)
public class RagConfigProperties {
/** host:ip */
private String host;
/** 端口 */
private int port;
- /** 数据库名 */
- private String database;
- /** 用户名 */
- private String user;
- /** 账密 */
- private String password;
- /** 向量维度 */
- //private String vectorDimension;
+ /** scheme */
+ private String scheme;
+ /** api-key */
+ private String apiKey;
+ /** timeout */
+ private int timeout;
+ /** class-name */
+ private String className;
}
diff --git a/refine-app/src/main/java/com/achobeta/config/WeaviateConfig.java b/refine-app/src/main/java/com/achobeta/config/WeaviateConfig.java
new file mode 100644
index 0000000..a46d72c
--- /dev/null
+++ b/refine-app/src/main/java/com/achobeta/config/WeaviateConfig.java
@@ -0,0 +1,86 @@
+package com.achobeta.config;
+
+import io.weaviate.client.Config;
+import io.weaviate.client.WeaviateAuthClient;
+import io.weaviate.client.WeaviateClient;
+import lombok.extern.slf4j.Slf4j;
+import org.springframework.beans.factory.annotation.Value;
+import org.springframework.context.annotation.Bean;
+import org.springframework.context.annotation.Configuration;
+
+/**
+ * @Auth : Malog
+ * @Desc : Weaviate客户端配置
+ * @Time : 2025/11/28
+ */
+@Slf4j
+@Configuration
+public class WeaviateConfig {
+
+ @Value("${weaviate.config.host:localhost}")
+ private String host;
+
+ @Value("${weaviate.config.port:8080}")
+ private int port;
+
+ @Value("${weaviate.config.scheme:http}")
+ private String scheme;
+
+ @Value("${weaviate.config.api-key:}")
+ private String apiKey;
+
+ @Value("${weaviate.config.timeout:30000}")
+ private int timeout;
+
+ @Value("${weaviate.config.class-name:LearningVector}")
+ private String className;
+
+ /**
+ * 创建Weaviate客户端Bean
+ */
+ @Bean
+ public WeaviateClient weaviateClient() {
+ try {
+ // 构建连接URL - 对于云服务，不需要端口号
+ String url;
+ Config config;
+
+ if (port == 443 && "https".equals(scheme)) {
+ // Weaviate Cloud服务，不需要端口号
+ url = scheme + "://" + host;
+ config = new Config(scheme, host);
+ } else {
+ // 本地服务，需要端口号
+ url = scheme + "://" + host + ":" + port;
+ config = new Config(scheme, host + ":" + port);
+ }
+
+ WeaviateClient client;
+
+ // 如果配置了API Key，使用认证客户端
+ if (apiKey != null && !apiKey.trim().isEmpty()) {
+ log.info("使用API Key认证连接Weaviate");
+ client = WeaviateAuthClient.apiKey(config, apiKey);
+ } else {
+ // 无认证连接
+ log.info("使用无认证方式连接Weaviate");
+ client = new WeaviateClient(config);
+ }
+
+ log.info("Weaviate客户端初始化成功，连接地址: {}", url);
+ return client;
+
+ } catch (Exception e) {
+ log.error("Weaviate客户端初始化失败", e);
+ throw new RuntimeException("Failed to initialize Weaviate client", e);
+ }
+ }
+
+ /**
+ * 获取Weaviate类名配置
+ */
+ @Bean
+ public String weaviateClassName() {
+ return className;
+ }
+}
\ No newline at end of file
diff --git a/refine-domain/src/main/java/com/achobeta/domain/ocr/service/impl/DefaultOcrService.java b/refine-domain/src/main/java/com/achobeta/domain/ocr/service/impl/DefaultOcrService.java
index 08f4a33..069cfdd 100644
--- a/refine-domain/src/main/java/com/achobeta/domain/ocr/service/impl/DefaultOcrService.java
+++ b/refine-domain/src/main/java/com/achobeta/domain/ocr/service/impl/DefaultOcrService.java
@@ -7,10 +7,12 @@
import com.achobeta.domain.ocr.model.entity.QuestionEntity;
import com.achobeta.domain.ocr.service.IOcrService;
import com.achobeta.domain.rag.service.IVectorService;
-import com.achobeta.types.enums.ActionType;
+import com.achobeta.domain.user.event.UserUploadEvent;
import com.achobeta.types.exception.AppException;
import lombok.RequiredArgsConstructor;
import lombok.extern.slf4j.Slf4j;
+import org.springframework.beans.factory.annotation.Qualifier;
+import org.springframework.context.ApplicationEventPublisher;
import org.springframework.stereotype.Service;
import java.io.IOException;
@@ -29,6 +31,9 @@ public class DefaultOcrService implements IOcrService {
private final IFilePreprocessPort filePreprocessPort;
private final IOcrPort ocrPort;
private final IAiService aiService;
+ private final ApplicationEventPublisher eventPublisher;
+
+ @Qualifier("weaviateVectorRepository")
private final IVectorService vectorService;
/**
@@ -122,26 +127,21 @@ else if (lowerType.endsWith(".txt") || "txt".equals(lowerType)) {
questionEntity.setQuestionId(uuid);
questionEntity.setUserId(userId);
- // 存储学习行为向量到向量数据库
+ // 发布用户上传事件，由事件监听器处理向量存储
try {
- // 调用向量服务存储学习向量，行为类型为"upload"表示用户上传题目
- boolean vectorStored = vectorService.storeLearningVector(
+ UserUploadEvent uploadEvent = new UserUploadEvent(
+ this,
userId,
uuid,
recognizedText,
- ActionType.UPLOAD.getActionType(),
- null,
- null
+ null, // subject 暂时为空，可以从AI分析中获取
+ null // knowledgePointId 暂时为空，可以从AI分析中获取
);
-
- if (vectorStored) {
- log.info("成功存储题目向量到向量数据库，userId:{} questionId:{}", userId, uuid);
- } else {
- log.warn("存储题目向量到向量数据库失败，但不影响OCR主流程，userId:{} questionId:{}", userId, uuid);
- }
+ eventPublisher.publishEvent(uploadEvent);
+ log.info("用户上传题目事件已发布，userId:{}, questionId:{}", userId, uuid);
} catch (Exception e) {
- // 向量存储失败不应该影响OCR主流程，只记录日志
- log.error("存储题目向量到向量数据库时发生异常，userId:{} questionId:{}", userId, uuid, e);
+ // 事件发布失败不应该影响OCR主流程，只记录日志
+ log.error("发布用户上传题目事件时发生异常，userId:{} questionId:{}", userId, uuid, e);
}
// TODO 将题干存储到Redis中，通过uuid可以查询，设置24小时过期时间
diff --git a/refine-domain/src/main/java/com/achobeta/domain/rag/service/impl/LearningAnalysisService.java b/refine-domain/src/main/java/com/achobeta/domain/rag/service/impl/LearningAnalysisService.java
index 544db96..b697c3e 100644
--- a/refine-domain/src/main/java/com/achobeta/domain/rag/service/impl/LearningAnalysisService.java
+++ b/refine-domain/src/main/java/com/achobeta/domain/rag/service/impl/LearningAnalysisService.java
@@ -7,6 +7,7 @@
import com.achobeta.domain.ai.service.IAiService;
import lombok.extern.slf4j.Slf4j;
import org.springframework.beans.factory.annotation.Autowired;
+import org.springframework.beans.factory.annotation.Qualifier;
import org.springframework.scheduling.annotation.Async;
import org.springframework.stereotype.Service;
@@ -20,6 +21,7 @@
public class LearningAnalysisService {
@Autowired
+ @Qualifier("weaviateVectorRepository")
private IVectorService vectorService;
@Autowired
@@ -38,7 +40,15 @@ public void onUserLogin(String userId) {
// 分析用户最近7天的学习动态
List dynamics = learningDynamicsService.analyzeUserLearningDynamics(userId);
- log.info("用户登录学习动态分析完成，userId:{} 动态数量:{}", userId, dynamics.size());
+ if (dynamics != null && !dynamics.isEmpty()) {
+ log.info("用户登录学习动态分析完成，userId:{} 动态数量:{}", userId, dynamics.size());
+
+ // 可以在这里将分析结果存储到缓存或数据库中，供前端查询使用
+ // 例如：存储到Redis中，key为 "user_dynamics:" + userId
+
+ } else {
+ log.info("用户登录学习动态分析完成，但未生成有效动态，userId:{}", userId);
+ }
} catch (Exception e) {
log.error("用户登录学习动态分析失败，userId:{}", userId, e);
diff --git a/refine-domain/src/main/java/com/achobeta/domain/rag/service/impl/LearningDynamicsServiceImpl.java b/refine-domain/src/main/java/com/achobeta/domain/rag/service/impl/LearningDynamicsServiceImpl.java
index c3d9479..44cf5a5 100644
--- a/refine-domain/src/main/java/com/achobeta/domain/rag/service/impl/LearningDynamicsServiceImpl.java
+++ b/refine-domain/src/main/java/com/achobeta/domain/rag/service/impl/LearningDynamicsServiceImpl.java
@@ -15,6 +15,7 @@
import java.util.*;
import java.util.concurrent.CompletableFuture;
import java.util.concurrent.TimeUnit;
+import java.util.concurrent.TimeoutException;
import java.util.stream.Collectors;
@Slf4j
@@ -47,15 +48,24 @@ public List analyzeUserLearningDynamics(String userId) {
// 3. 使用AI分析学习动态
List dynamics = new ArrayList<>();
+ StringBuilder responseBuilder = new StringBuilder();
+
// 异步执行AI分析任务，解析题目并生成学习动态
CompletableFuture aiAnalysis = CompletableFuture.runAsync(() -> {
aiService.aiChat(analysisPrompt, response -> {
try {
- // 解析AI返回的JSON格式学习动态
- List parsedDynamics = parseAIResponse(response);
- dynamics.addAll(parsedDynamics);
+ // 收集所有响应片段
+ responseBuilder.append(response);
+
+ // 检查是否收到完整的JSON响应
+ String currentResponse = responseBuilder.toString();
+ if (isCompleteJsonResponse(currentResponse)) {
+ // 解析完整的AI返回的JSON格式学习动态
+ List parsedDynamics = parseAIResponse(currentResponse);
+ dynamics.addAll(parsedDynamics);
+ }
} catch (Exception e) {
- log.error("解析AI分析结果失败", e);
+ log.error("解析AI分析结果失败：{}", response, e);
}
});
});
@@ -64,9 +74,17 @@ public List analyzeUserLearningDynamics(String userId) {
// 等待AI分析完成，最多等待30秒
try {
aiAnalysis.get(30, TimeUnit.SECONDS);
+
+ // 如果AI分析没有产生结果，使用基于规则的分析
+ if (dynamics.isEmpty()) {
+ log.warn("AI分析未产生结果，使用基于规则的分析，userId:{}", userId);
+ return generateRuleBasedDynamics(userId);
+ }
+ } catch (TimeoutException e) {
+ log.error("AI分析超时，userId:{}", userId, e);
+ return generateRuleBasedDynamics(userId);
} catch (Exception e) {
- log.error("AI分析超时或失败", e);
- // 如果AI分析失败，返回基于规则的分析结果
+ log.error("AI分析失败，userId:{}", userId, e);
return generateRuleBasedDynamics(userId);
}
// 4. 限制返回最多3条动态
@@ -163,17 +181,26 @@ private String buildAnalysisPrompt(String learningDataSummary) {
学习数据：
%s
- 请按照以下JSON格式返回分析结果，每条动态都要有明确的类型、标题、描述和建议：
+ **重要要求：请严格按照以下JSON格式返回分析结果，不要添加任何额外的文字说明，只返回纯JSON数组：**
[
{
- "type": "progress|weakness|achievement",
+ "type": "progress",
"title": "动态标题",
"description": "详细描述",
"subject": "相关科目",
- "priority": 1-5,
+ "priority": 3,
"suggestion": "建议行动",
- "relatedQuestionCount": 数量
+ "relatedQuestionCount": 5
+ },
+ {
+ "type": "weakness",
+ "title": "另一个动态标题",
+ "description": "另一个详细描述",
+ "subject": "另一个科目",
+ "priority": 4,
+ "suggestion": "另一个建议行动",
+ "relatedQuestionCount": 2
}
]
@@ -182,6 +209,13 @@ private String buildAnalysisPrompt(String learningDataSummary) {
- weakness: 发现薄弱点，如某科目错误较多、学习频率低等
- achievement: 学习成就，如连续学习天数、完成学习目标等
+ **注意：**
+ 1. 必须返回有效的JSON数组格式
+ 2. priority必须是1-5之间的整数
+ 3. relatedQuestionCount必须是非负整数
+ 4. 所有字符串字段不能为null
+ 5. 不要在JSON前后添加任何解释文字
+
请确保返回的是有效的JSON格式，优先级高的动态排在前面。
""", learningDataSummary);
}
@@ -191,8 +225,16 @@ private String buildAnalysisPrompt(String learningDataSummary) {
*/
private List parseAIResponse(String aiResponse) {
try {
+ log.debug("开始解析AI响应，响应长度: {}", aiResponse != null ? aiResponse.length() : 0);
+
+ if (aiResponse == null || aiResponse.trim().isEmpty()) {
+ log.warn("AI响应为空，返回空列表");
+ return Collections.emptyList();
+ }
+
// 提取JSON部分（AI可能返回额外的文本）
String jsonPart = extractJsonFromResponse(aiResponse);
+ log.debug("提取的JSON部分: {}", jsonPart.length() > 500 ? jsonPart.substring(0, 500) + "..." : jsonPart);
// 解析JSON
List> dynamicMaps = objectMapper.readValue(
@@ -201,38 +243,125 @@ private List parseAIResponse(String aiResponse) {
List dynamics = new ArrayList<>();
for (Map map : dynamicMaps) {
- LearningDynamicVO dynamic = LearningDynamicVO.builder()
- .type(map.get("type").toString())
- .title(map.get("title").toString())
- .description(map.get("description").toString())
- .subject(map.get("subject") != null ? map.get("subject").toString() : null)
- .priority(Integer.parseInt(map.get("priority").toString()))
- .suggestion(map.get("suggestion") != null ? map.get("suggestion").toString() : null)
- .relatedQuestionCount(map.get("relatedQuestionCount") != null ?
- Integer.parseInt(map.get("relatedQuestionCount").toString()) : 0)
- .build();
- dynamics.add(dynamic);
+ try {
+ LearningDynamicVO dynamic = LearningDynamicVO.builder()
+ .type(getStringValue(map, "type"))
+ .title(getStringValue(map, "title"))
+ .description(getStringValue(map, "description"))
+ .subject(getStringValue(map, "subject"))
+ .priority(getIntValue(map, "priority", 3))
+ .suggestion(getStringValue(map, "suggestion"))
+ .relatedQuestionCount(getIntValue(map, "relatedQuestionCount", 0))
+ .build();
+ dynamics.add(dynamic);
+ log.debug("成功解析学习动态: type={}, title={}", dynamic.getType(), dynamic.getTitle());
+ } catch (Exception e) {
+ log.warn("解析单个学习动态失败，跳过该项: {}", map, e);
+ }
}
+ log.info("成功解析AI响应，获得{}条学习动态", dynamics.size());
return dynamics;
} catch (Exception e) {
- log.error("解析AI分析结果失败：{}", aiResponse, e);
+ log.error("解析AI分析结果失败，响应内容: {}",
+ aiResponse != null && aiResponse.length() > 200 ?
+ aiResponse.substring(0, 200) + "..." : aiResponse, e);
return Collections.emptyList();
}
}
+
+ /**
+ * 安全获取字符串值
+ */
+ private String getStringValue(Map map, String key) {
+ Object value = map.get(key);
+ return value != null ? value.toString() : null;
+ }
+
+ /**
+ * 安全获取整数值
+ */
+ private int getIntValue(Map map, String key, int defaultValue) {
+ Object value = map.get(key);
+ if (value == null) {
+ return defaultValue;
+ }
+ try {
+ if (value instanceof Number) {
+ return ((Number) value).intValue();
+ }
+ return Integer.parseInt(value.toString());
+ } catch (NumberFormatException e) {
+ log.warn("无法解析整数值: key={}, value={}, 使用默认值: {}", key, value, defaultValue);
+ return defaultValue;
+ }
+ }
+
+ /**
+ * 检查响应是否包含完整的JSON
+ */
+ private boolean isCompleteJsonResponse(String response) {
+ if (response == null || response.trim().isEmpty()) {
+ return false;
+ }
+
+ String trimmed = response.trim();
+
+ // 检查是否包含JSON数组的开始和结束标记
+ int startIndex = trimmed.indexOf('[');
+ int endIndex = trimmed.lastIndexOf(']');
+
+ if (startIndex == -1 || endIndex == -1 || endIndex <= startIndex) {
+ return false;
+ }
+
+ // 简单的括号匹配检查
+ int openBrackets = 0;
+ int closeBrackets = 0;
+
+ for (char c : trimmed.toCharArray()) {
+ if (c == '[') openBrackets++;
+ if (c == ']') closeBrackets++;
+ }
+
+ return openBrackets == closeBrackets && openBrackets > 0;
+ }
/**
* 从AI响应中提取JSON部分
*/
private String extractJsonFromResponse(String response) {
+ if (response == null || response.trim().isEmpty()) {
+ throw new IllegalArgumentException("AI响应为空");
+ }
+
+ String trimmed = response.trim();
+ log.debug("尝试从响应中提取JSON，响应长度: {}", trimmed.length());
+
// 查找JSON数组的开始和结束
- int start = response.indexOf('[');
- int end = response.lastIndexOf(']');
+ int start = trimmed.indexOf('[');
+ int end = trimmed.lastIndexOf(']');
if (start != -1 && end != -1 && end > start) {
- return response.substring(start, end + 1);
+ String jsonPart = trimmed.substring(start, end + 1);
+ log.debug("成功提取JSON部分，长度: {}", jsonPart.length());
+ return jsonPart;
+ }
+
+ // 如果没有找到完整的JSON数组，尝试查找JSON对象
+ start = trimmed.indexOf('{');
+ end = trimmed.lastIndexOf('}');
+
+ if (start != -1 && end != -1 && end > start) {
+ // 包装单个对象为数组
+ String jsonPart = "[" + trimmed.substring(start, end + 1) + "]";
+ log.debug("找到JSON对象，包装为数组，长度: {}", jsonPart.length());
+ return jsonPart;
}
+
+ log.error("无法从AI响应中提取有效的JSON，响应内容: {}",
+ trimmed.length() > 200 ? trimmed.substring(0, 200) + "..." : trimmed);
throw new IllegalArgumentException("无法从AI响应中提取有效的JSON");
}
diff --git a/refine-domain/src/main/java/com/achobeta/domain/user/event/UserBehaviorEvent.java b/refine-domain/src/main/java/com/achobeta/domain/user/event/UserBehaviorEvent.java
new file mode 100644
index 0000000..cf39c3e
--- /dev/null
+++ b/refine-domain/src/main/java/com/achobeta/domain/user/event/UserBehaviorEvent.java
@@ -0,0 +1,32 @@
+package com.achobeta.domain.user.event;
+
+import lombok.Getter;
+import org.springframework.context.ApplicationEvent;
+
+/**
+ * @Auth : Malog
+ * @Desc : 用户行为事件基类
+ * @Time : 2025/11/28
+ */
+@Getter
+public abstract class UserBehaviorEvent extends ApplicationEvent {
+
+ private final String userId;
+ private final String actionType;
+ private final String questionId;
+ private final String questionContent;
+ private final String subject;
+ private final Integer knowledgePointId;
+
+ public UserBehaviorEvent(Object source, String userId, String actionType,
+ String questionId, String questionContent,
+ String subject, Integer knowledgePointId) {
+ super(source);
+ this.userId = userId;
+ this.actionType = actionType;
+ this.questionId = questionId;
+ this.questionContent = questionContent;
+ this.subject = subject;
+ this.knowledgePointId = knowledgePointId;
+ }
+}
\ No newline at end of file
diff --git a/refine-domain/src/main/java/com/achobeta/domain/user/event/UserMistakeEvent.java b/refine-domain/src/main/java/com/achobeta/domain/user/event/UserMistakeEvent.java
new file mode 100644
index 0000000..c59e118
--- /dev/null
+++ b/refine-domain/src/main/java/com/achobeta/domain/user/event/UserMistakeEvent.java
@@ -0,0 +1,14 @@
+package com.achobeta.domain.user.event;
+
+/**
+ * @Auth : Malog
+ * @Desc : 用户错题事件
+ * @Time : 2025/11/28
+ */
+public class UserMistakeEvent extends UserBehaviorEvent {
+
+ public UserMistakeEvent(Object source, String userId, String questionId,
+ String questionContent, String subject, Integer knowledgePointId) {
+ super(source, userId, "mistake", questionId, questionContent, subject, knowledgePointId);
+ }
+}
\ No newline at end of file
diff --git a/refine-domain/src/main/java/com/achobeta/domain/user/event/UserReviewEvent.java b/refine-domain/src/main/java/com/achobeta/domain/user/event/UserReviewEvent.java
new file mode 100644
index 0000000..37bd77f
--- /dev/null
+++ b/refine-domain/src/main/java/com/achobeta/domain/user/event/UserReviewEvent.java
@@ -0,0 +1,14 @@
+package com.achobeta.domain.user.event;
+
+/**
+ * @Auth : Malog
+ * @Desc : 用户复习错题事件
+ * @Time : 2025/11/28
+ */
+public class UserReviewEvent extends UserBehaviorEvent {
+
+ public UserReviewEvent(Object source, String userId, String questionId,
+ String questionContent, String subject, Integer knowledgePointId) {
+ super(source, userId, "review", questionId, questionContent, subject, knowledgePointId);
+ }
+}
\ No newline at end of file
diff --git a/refine-domain/src/main/java/com/achobeta/domain/user/event/UserUploadEvent.java b/refine-domain/src/main/java/com/achobeta/domain/user/event/UserUploadEvent.java
new file mode 100644
index 0000000..553b13e
--- /dev/null
+++ b/refine-domain/src/main/java/com/achobeta/domain/user/event/UserUploadEvent.java
@@ -0,0 +1,14 @@
+package com.achobeta.domain.user.event;
+
+/**
+ * @Auth : Malog
+ * @Desc : 用户上传题目事件
+ * @Time : 2025/11/28
+ */
+public class UserUploadEvent extends UserBehaviorEvent {
+
+ public UserUploadEvent(Object source, String userId, String questionId,
+ String questionContent, String subject, Integer knowledgePointId) {
+ super(source, userId, "upload", questionId, questionContent, subject, knowledgePointId);
+ }
+}
\ No newline at end of file
diff --git a/refine-infrastructure/pom.xml b/refine-infrastructure/pom.xml
index 904eecc..60f4b2a 100644
--- a/refine-infrastructure/pom.xml
+++ b/refine-infrastructure/pom.xml
@@ -11,6 +11,10 @@
refine-infrastructure
+
+ io.weaviate
+ client
+
org.apache.pdfbox
diff --git a/refine-infrastructure/src/main/java/com/achobeta/infrastructure/adapter/repository/LearningDataRepository.java b/refine-infrastructure/src/main/java/com/achobeta/infrastructure/adapter/repository/LearningDataRepository.java
index 8107137..bbbfb0f 100644
--- a/refine-infrastructure/src/main/java/com/achobeta/infrastructure/adapter/repository/LearningDataRepository.java
+++ b/refine-infrastructure/src/main/java/com/achobeta/infrastructure/adapter/repository/LearningDataRepository.java
@@ -3,14 +3,17 @@
import com.achobeta.domain.rag.adapter.port.ILearningDataRepository;
import com.achobeta.domain.rag.model.entity.LearningVectorEntity;
import com.achobeta.domain.rag.model.valobj.LearningStatisticsVO;
-import com.achobeta.infrastructure.dao.vector.IVectorDao;
-import com.achobeta.infrastructure.dao.po.LearningVector;
+import com.fasterxml.jackson.databind.ObjectMapper;
+import io.weaviate.client.WeaviateClient;
+import io.weaviate.client.base.Result;
+import io.weaviate.client.v1.graphql.model.GraphQLResponse;
+import io.weaviate.client.v1.graphql.query.fields.Field;
import lombok.extern.slf4j.Slf4j;
import org.springframework.beans.factory.annotation.Autowired;
import org.springframework.stereotype.Repository;
-import org.springframework.util.CollectionUtils;
import java.time.LocalDateTime;
+import java.time.format.DateTimeFormatter;
import java.util.*;
import java.util.stream.Collectors;
@@ -24,7 +27,12 @@
public class LearningDataRepository implements ILearningDataRepository {
@Autowired
- private IVectorDao vectorDao;
+ private WeaviateClient weaviateClient;
+
+ @Autowired
+ private String weaviateClassName;
+
+ private final ObjectMapper objectMapper = new ObjectMapper();
@Override
public List getUserRecentLearningData(String userId, int days) {
@@ -41,22 +49,16 @@ public List getUserRecentLearningData(String userId, int d
return Collections.emptyList();
}
- // 从数据库查询用户最近N天的学习数据
- List vectors = vectorDao.getUserRecentLearningData(userId, days);
-
- if (CollectionUtils.isEmpty(vectors)) {
- log.info("用户最近{}天无学习数据，userId:{}", days, userId);
- return Collections.emptyList();
- }
+ // 使用Weaviate查询用户最近N天的学习数据
+ List> rawData = queryUserDataWithTimeFilter(userId, days);
- // 转换为领域实体
- List result = vectors.stream()
- .map(this::convertToEntity)
+ // 转换为LearningVectorEntity
+ return rawData.stream()
+ .map(this::convertToLearningVectorEntity)
+ .filter(Objects::nonNull)
+ .sorted((a, b) -> b.getCreatedAt().compareTo(a.getCreatedAt())) // 按时间倒序
.collect(Collectors.toList());
- log.info("成功获取用户学习向量数据，userId:{}, 数据量:{}", userId, result.size());
- return result;
-
} catch (Exception e) {
log.error("获取用户学习向量数据失败，userId:{}, days:{}", userId, days, e);
return Collections.emptyList();
@@ -78,11 +80,10 @@ public LearningStatisticsVO getUserLearningStatistics(String userId, int days) {
return null;
}
- // 从数据库查询统计数据
- Map statisticsMap = vectorDao.getUserLearningStatistics(userId, days);
+ // 使用Weaviate查询用户学习统计数据
+ List> rawData = queryUserDataWithTimeFilter(userId, days);
- if (statisticsMap == null || statisticsMap.isEmpty()) {
- log.info("用户最近{}天无学习统计数据，userId:{}", days, userId);
+ if (rawData.isEmpty()) {
return LearningStatisticsVO.builder()
.totalActivities(0)
.subjectsCount(0)
@@ -91,11 +92,70 @@ public LearningStatisticsVO getUserLearningStatistics(String userId, int days) {
.build();
}
- // 转换为统计VO
- LearningStatisticsVO result = convertToStatisticsVO(statisticsMap);
+ // 计算统计数据
+ int totalActivities = rawData.size();
- log.info("成功获取用户学习统计数据，userId:{}, 总活动次数:{}", userId, result.getTotalActivities());
- return result;
+ Set subjects = rawData.stream()
+ .map(data -> (String) data.get("subject"))
+ .filter(Objects::nonNull)
+ .filter(s -> !s.trim().isEmpty())
+ .collect(Collectors.toSet());
+
+ Set actionTypes = rawData.stream()
+ .map(data -> (String) data.get("actionType"))
+ .filter(Objects::nonNull)
+ .collect(Collectors.toSet());
+
+ Set activeDates = rawData.stream()
+ .map(data -> (String) data.get("createdAt"))
+ .filter(Objects::nonNull)
+ .map(dateStr -> {
+ try {
+ LocalDateTime dateTime = LocalDateTime.parse(dateStr, DateTimeFormatter.ISO_LOCAL_DATE_TIME);
+ return dateTime.toLocalDate().toString();
+ } catch (Exception e) {
+ return null;
+ }
+ })
+ .filter(Objects::nonNull)
+ .collect(Collectors.toSet());
+
+ LocalDateTime firstActivity = rawData.stream()
+ .map(data -> (String) data.get("createdAt"))
+ .filter(Objects::nonNull)
+ .map(dateStr -> {
+ try {
+ return LocalDateTime.parse(dateStr, DateTimeFormatter.ISO_LOCAL_DATE_TIME);
+ } catch (Exception e) {
+ return null;
+ }
+ })
+ .filter(Objects::nonNull)
+ .min(LocalDateTime::compareTo)
+ .orElse(null);
+
+ LocalDateTime lastActivity = rawData.stream()
+ .map(data -> (String) data.get("createdAt"))
+ .filter(Objects::nonNull)
+ .map(dateStr -> {
+ try {
+ return LocalDateTime.parse(dateStr, DateTimeFormatter.ISO_LOCAL_DATE_TIME);
+ } catch (Exception e) {
+ return null;
+ }
+ })
+ .filter(Objects::nonNull)
+ .max(LocalDateTime::compareTo)
+ .orElse(null);
+
+ return LearningStatisticsVO.builder()
+ .totalActivities(totalActivities)
+ .subjectsCount(subjects.size())
+ .activeDays(activeDates.size())
+ .actionTypesCount(actionTypes.size())
+ .firstActivity(firstActivity)
+ .lastActivity(lastActivity)
+ .build();
} catch (Exception e) {
log.error("获取用户学习统计数据失败，userId:{}, days:{}", userId, days, e);
@@ -118,15 +178,43 @@ public List> getUserLearningDataBySubject(String userId, int
return Collections.emptyList();
}
- // 从数据库查询按科目分组的学习数据
- List> result = vectorDao.getUserLearningDataBySubject(userId, days);
+ // 使用Weaviate查询用户学习数据
+ List> rawData = queryUserDataWithTimeFilter(userId, days);
- if (CollectionUtils.isEmpty(result)) {
- log.info("用户最近{}天无按科目分组的学习数据，userId:{}", days, userId);
+ if (rawData.isEmpty()) {
return Collections.emptyList();
}
- log.info("成功获取用户按科目分组学习数据，userId:{}, 科目数量:{}", userId, result.size());
+ // 按科目分组统计
+ Map>> subjectGroups = rawData.stream()
+ .filter(data -> data.get("subject") != null && !data.get("subject").toString().trim().isEmpty())
+ .collect(Collectors.groupingBy(data -> data.get("subject").toString()));
+
+ // 转换为结果格式
+ List> result = new ArrayList<>();
+ for (Map.Entry>> entry : subjectGroups.entrySet()) {
+ String subject = entry.getKey();
+ List> subjectData = entry.getValue();
+
+ Set actionTypes = subjectData.stream()
+ .map(data -> (String) data.get("actionType"))
+ .filter(Objects::nonNull)
+ .collect(Collectors.toSet());
+
+ Map subjectStat = new HashMap<>();
+ subjectStat.put("subject", subject);
+ subjectStat.put("activity_count", subjectData.size());
+ subjectStat.put("action_list", String.join(",", actionTypes));
+ result.add(subjectStat);
+ }
+
+ // 按活动数量降序排序
+ result.sort((a, b) -> {
+ Integer countA = (Integer) a.get("activity_count");
+ Integer countB = (Integer) b.get("activity_count");
+ return countB.compareTo(countA);
+ });
+
return result;
} catch (Exception e) {
@@ -150,15 +238,44 @@ public List> getUserLearningDataByActionType(String userId,
return Collections.emptyList();
}
- // 从数据库查询按行为类型分组的学习数据
- List> result = vectorDao.getUserLearningDataByActionType(userId, days);
+ // 使用Weaviate查询用户学习数据
+ List> rawData = queryUserDataWithTimeFilter(userId, days);
- if (CollectionUtils.isEmpty(result)) {
- log.info("用户最近{}天无按行为类型分组的学习数据，userId:{}", days, userId);
+ if (rawData.isEmpty()) {
return Collections.emptyList();
}
- log.info("成功获取用户按行为类型分组学习数据，userId:{}, 行为类型数量:{}", userId, result.size());
+ // 按行为类型分组统计
+ Map>> actionGroups = rawData.stream()
+ .filter(data -> data.get("actionType") != null)
+ .collect(Collectors.groupingBy(data -> data.get("actionType").toString()));
+
+ // 转换为结果格式
+ List> result = new ArrayList<>();
+ for (Map.Entry>> entry : actionGroups.entrySet()) {
+ String actionType = entry.getKey();
+ List> actionData = entry.getValue();
+
+ Set subjects = actionData.stream()
+ .map(data -> (String) data.get("subject"))
+ .filter(Objects::nonNull)
+ .filter(s -> !s.trim().isEmpty())
+ .collect(Collectors.toSet());
+
+ Map actionStat = new HashMap<>();
+ actionStat.put("action_type", actionType);
+ actionStat.put("activity_count", actionData.size());
+ actionStat.put("subjects_count", subjects.size());
+ result.add(actionStat);
+ }
+
+ // 按活动数量降序排序
+ result.sort((a, b) -> {
+ Integer countA = (Integer) a.get("activity_count");
+ Integer countB = (Integer) b.get("activity_count");
+ return countB.compareTo(countA);
+ });
+
return result;
} catch (Exception e) {
@@ -168,26 +285,143 @@ public List> getUserLearningDataByActionType(String userId,
}
/**
- * 将数据库PO转换为领域实体
+ * 查询用户指定时间范围内的学习数据
*/
- private LearningVectorEntity convertToEntity(LearningVector vector) {
- if (vector == null) {
+ private List> queryUserDataWithTimeFilter(String userId, int days) {
+ try {
+ // 计算时间范围
+ LocalDateTime endTime = LocalDateTime.now();
+ LocalDateTime startTime = endTime.minusDays(days);
+
+ Field[] fields = {
+ Field.builder().name("userId").build(),
+ Field.builder().name("questionId").build(),
+ Field.builder().name("questionContent").build(),
+ Field.builder().name("actionType").build(),
+ Field.builder().name("subject").build(),
+ Field.builder().name("knowledgePointId").build(),
+ Field.builder().name("createdAt").build()
+ };
+
+ io.weaviate.client.v1.filters.WhereFilter whereFilter = io.weaviate.client.v1.filters.WhereFilter.builder()
+ .operator(io.weaviate.client.v1.filters.Operator.And)
+ .operands(new io.weaviate.client.v1.filters.WhereFilter[]{
+ io.weaviate.client.v1.filters.WhereFilter.builder()
+ .path(new String[]{"userId"})
+ .operator(io.weaviate.client.v1.filters.Operator.Equal)
+ .valueText(userId)
+ .build(),
+ io.weaviate.client.v1.filters.WhereFilter.builder()
+ .path(new String[]{"createdAt"})
+ .operator(io.weaviate.client.v1.filters.Operator.GreaterThanEqual)
+ .valueText(startTime.format(DateTimeFormatter.ISO_LOCAL_DATE_TIME))
+ .build()
+ })
+ .build();
+
+ Result result = weaviateClient.graphQL().get()
+ .withClassName(weaviateClassName)
+ .withFields(fields)
+ .withWhere(whereFilter)
+ .withLimit(1000)
+ .run();
+
+ if (result.hasErrors()) {
+ log.error("查询用户学习数据失败: {}", result.getError().getMessages());
+ return Collections.emptyList();
+ }
+
+ return parseQueryResults(result.getResult());
+
+ } catch (Exception e) {
+ log.error("查询用户学习数据失败，userId:{} days:{}", userId, days, e);
+ return Collections.emptyList();
+ }
+ }
+
+ /**
+ * 解析Weaviate查询结果
+ */
+ @SuppressWarnings("unchecked")
+ private List> parseQueryResults(GraphQLResponse response) {
+ try {
+ if (response == null || response.getData() == null) {
+ return Collections.emptyList();
+ }
+
+ // 安全地转换getData()的结果
+ Object dataObj = response.getData();
+ if (!(dataObj instanceof Map)) {
+ log.warn("Weaviate查询结果数据格式不正确，期望Map类型，实际类型: {}", dataObj.getClass().getSimpleName());
+ return Collections.emptyList();
+ }
+
+ Map data = (Map) dataObj;
+ Map get = (Map) data.get("Get");
+ if (get == null) {
+ return Collections.emptyList();
+ }
+
+ List> objects = (List>) get.get(weaviateClassName);
+ if (objects == null) {
+ return Collections.emptyList();
+ }
+
+ return objects;
+
+ } catch (Exception e) {
+ log.error("解析Weaviate查询结果失败", e);
+ return Collections.emptyList();
+ }
+ }
+
+ /**
+ * 将Weaviate查询结果转换为LearningVectorEntity
+ */
+ private LearningVectorEntity convertToLearningVectorEntity(Map data) {
+ try {
+ if (data == null) {
+ return null;
+ }
+
+ LocalDateTime createdAt = null;
+ String createdAtStr = (String) data.get("createdAt");
+ if (createdAtStr != null) {
+ try {
+ createdAt = LocalDateTime.parse(createdAtStr, DateTimeFormatter.ISO_LOCAL_DATE_TIME);
+ } catch (Exception e) {
+ log.warn("解析创建时间失败: {}", createdAtStr);
+ }
+ }
+
+ Integer knowledgePointId = null;
+ Object kpId = data.get("knowledgePointId");
+ if (kpId != null) {
+ if (kpId instanceof Number) {
+ knowledgePointId = ((Number) kpId).intValue();
+ } else {
+ try {
+ knowledgePointId = Integer.parseInt(kpId.toString());
+ } catch (NumberFormatException e) {
+ log.warn("解析知识点ID失败: {}", kpId);
+ }
+ }
+ }
+
+ return LearningVectorEntity.builder()
+ .userId((String) data.get("userId"))
+ .questionId((String) data.get("questionId"))
+ .questionContent((String) data.get("questionContent"))
+ .actionType((String) data.get("actionType"))
+ .subject((String) data.get("subject"))
+ .knowledgePointId(knowledgePointId)
+ .createdAt(createdAt)
+ .build();
+
+ } catch (Exception e) {
+ log.error("转换LearningVectorEntity失败", e);
return null;
}
-
- return LearningVectorEntity.builder()
- .id(vector.getId())
- .userId(vector.getUserId())
- .questionId(vector.getQuestionId())
- .actionType(vector.getActionType())
- .questionContent(vector.getQuestionContent())
- .subject(vector.getSubject())
- .knowledgePointId(vector.getKnowledgePointId())
- .embedding(vector.getEmbedding())
- .metadata(vector.getMetadata())
- .createdAt(vector.getCreatedAt())
- .updatedAt(vector.getUpdatedAt())
- .build();
}
/**
diff --git a/refine-infrastructure/src/main/java/com/achobeta/infrastructure/adapter/repository/WeaviateSchemaInitializer.java b/refine-infrastructure/src/main/java/com/achobeta/infrastructure/adapter/repository/WeaviateSchemaInitializer.java
new file mode 100644
index 0000000..18c2050
--- /dev/null
+++ b/refine-infrastructure/src/main/java/com/achobeta/infrastructure/adapter/repository/WeaviateSchemaInitializer.java
@@ -0,0 +1,156 @@
+package com.achobeta.infrastructure.adapter.repository;
+
+import io.weaviate.client.WeaviateClient;
+import io.weaviate.client.base.Result;
+import io.weaviate.client.v1.schema.model.DataType;
+import io.weaviate.client.v1.schema.model.Property;
+import io.weaviate.client.v1.schema.model.WeaviateClass;
+import lombok.extern.slf4j.Slf4j;
+import org.springframework.beans.factory.annotation.Autowired;
+import org.springframework.beans.factory.annotation.Value;
+import org.springframework.boot.CommandLineRunner;
+import org.springframework.stereotype.Component;
+
+import java.util.ArrayList;
+import java.util.List;
+
+/**
+ * @Auth : Malog
+ * @Desc : Weaviate Schema初始化器
+ * @Time : 2025/12/01
+ */
+@Slf4j
+@Component
+public class WeaviateSchemaInitializer implements CommandLineRunner {
+
+ @Autowired
+ private WeaviateClient weaviateClient;
+
+ @Value("${weaviate.class-name:LearningVector}")
+ private String className;
+
+ @Override
+ public void run(String... args) throws Exception {
+ try {
+ log.info("开始初始化Weaviate Schema，className: {}", className);
+
+ // 检查Schema是否已存在
+ if (isSchemaExists()) {
+ log.info("Weaviate Schema已存在，跳过初始化，className: {}", className);
+ return;
+ }
+
+ // 创建Schema
+ createSchema();
+ log.info("Weaviate Schema初始化完成，className: {}", className);
+
+ } catch (Exception e) {
+ log.error("Weaviate Schema初始化失败", e);
+ // 不抛出异常，避免影响应用启动
+ }
+ }
+
+ /**
+ * 检查Schema是否存在
+ */
+ private boolean isSchemaExists() {
+ try {
+ Result result = weaviateClient.schema().classGetter()
+ .withClassName(className)
+ .run();
+
+ return !result.hasErrors() && result.getResult() != null;
+ } catch (Exception e) {
+ log.warn("检查Schema是否存在时发生异常: {}", e.getMessage());
+ return false;
+ }
+ }
+
+ /**
+ * 创建Schema
+ */
+ private void createSchema() {
+ try {
+ // 定义属性
+ List properties = new ArrayList<>();
+
+ // 用户ID
+ properties.add(Property.builder()
+ .name("userId")
+ .dataType(List.of(DataType.TEXT))
+ .description("用户ID")
+ .build());
+
+ // 题目ID
+ properties.add(Property.builder()
+ .name("questionId")
+ .dataType(List.of(DataType.TEXT))
+ .description("题目ID")
+ .build());
+
+ // 题目内容
+ properties.add(Property.builder()
+ .name("questionContent")
+ .dataType(List.of(DataType.TEXT))
+ .description("题目内容")
+ .build());
+
+ // 行为类型
+ properties.add(Property.builder()
+ .name("actionType")
+ .dataType(List.of(DataType.TEXT))
+ .description("行为类型")
+ .build());
+
+ // 科目
+ properties.add(Property.builder()
+ .name("subject")
+ .dataType(List.of(DataType.TEXT))
+ .description("科目")
+ .build());
+
+ // 知识点ID
+ properties.add(Property.builder()
+ .name("knowledgePointId")
+ .dataType(List.of(DataType.INT))
+ .description("知识点ID")
+ .build());
+
+ // 创建时间
+ properties.add(Property.builder()
+ .name("createdAt")
+ .dataType(List.of(DataType.TEXT))
+ .description("创建时间")
+ .build());
+
+ // 更新时间
+ properties.add(Property.builder()
+ .name("updatedAt")
+ .dataType(List.of(DataType.TEXT))
+ .description("更新时间")
+ .build());
+
+ // 创建类定义
+ WeaviateClass weaviateClass = WeaviateClass.builder()
+ .className(className)
+ .description("学习行为向量数据")
+ .properties(properties)
+ .vectorizer("text2vec-cohere") // 使用Cohere向量化器，适合云实例
+ .build();
+
+ // 创建Schema
+ Result result = weaviateClient.schema().classCreator()
+ .withClass(weaviateClass)
+ .run();
+
+ if (result.hasErrors()) {
+ log.error("创建Weaviate Schema失败: {}", result.getError().getMessages());
+ } else {
+ log.info("成功创建Weaviate Schema，className: {}", className);
+ }
+
+ } catch (Exception e) {
+ log.error("创建Weaviate Schema时发生异常", e);
+ }
+ }
+}
\ No newline at end of file
diff --git a/refine-infrastructure/src/main/java/com/achobeta/infrastructure/adapter/repository/WeaviateVectorRepository.java b/refine-infrastructure/src/main/java/com/achobeta/infrastructure/adapter/repository/WeaviateVectorRepository.java
new file mode 100644
index 0000000..95c9a67
--- /dev/null
+++ b/refine-infrastructure/src/main/java/com/achobeta/infrastructure/adapter/repository/WeaviateVectorRepository.java
@@ -0,0 +1,573 @@
+package com.achobeta.infrastructure.adapter.repository;
+
+import com.achobeta.domain.rag.model.valobj.LearningInsightVO;
+import com.achobeta.domain.rag.model.valobj.SimilarQuestionVO;
+import com.achobeta.domain.rag.service.IVectorService;
+import com.achobeta.infrastructure.gateway.DashScopeEmbeddingService;
+import com.fasterxml.jackson.databind.ObjectMapper;
+import io.weaviate.client.WeaviateClient;
+import io.weaviate.client.base.Result;
+import io.weaviate.client.v1.data.model.WeaviateObject;
+import io.weaviate.client.v1.graphql.model.GraphQLResponse;
+import io.weaviate.client.v1.graphql.query.argument.NearVectorArgument;
+import io.weaviate.client.v1.graphql.query.fields.Field;
+import io.weaviate.client.v1.schema.model.DataType;
+import io.weaviate.client.v1.schema.model.Property;
+import io.weaviate.client.v1.schema.model.WeaviateClass;
+import jakarta.annotation.PostConstruct;
+import lombok.extern.slf4j.Slf4j;
+import org.springframework.beans.factory.annotation.Autowired;
+import org.springframework.beans.factory.annotation.Qualifier;
+import org.springframework.stereotype.Repository;
+
+import java.time.LocalDateTime;
+import java.time.format.DateTimeFormatter;
+import java.util.*;
+import java.util.stream.Collectors;
+
+/**
+ * @Auth : Malog
+ * @Desc : Weaviate向量数据库适配器实现
+ * @Time : 2025/11/28
+ */
+@Slf4j
+@Repository
+@Qualifier("weaviateVectorRepository")
+public class WeaviateVectorRepository implements IVectorService {
+
+ @Autowired
+ private WeaviateClient weaviateClient;
+
+ @Autowired
+ private String weaviateClassName;
+
+ @Autowired
+ private DashScopeEmbeddingService embeddingService;
+
+ private final ObjectMapper objectMapper = new ObjectMapper();
+
+ /**
+ * 初始化Weaviate Schema
+ */
+ @PostConstruct
+ public void initializeSchema() {
+ try {
+ // 检查类是否已存在
+ Result existsResult = weaviateClient.schema().classGetter().withClassName(weaviateClassName).run();
+
+ if (existsResult.hasErrors()) {
+ // 如果获取失败，可能是类不存在，尝试创建
+ log.info("Weaviate类 {} 不存在，开始创建", weaviateClassName);
+ createWeaviateClass();
+ log.info("Weaviate类 {} 创建成功", weaviateClassName);
+ } else {
+ log.info("Weaviate类 {} 已存在", weaviateClassName);
+ }
+ } catch (Exception e) {
+ log.error("初始化Weaviate Schema失败", e);
+ }
+ }
+
+ /**
+ * 创建Weaviate类定义
+ */
+ private void createWeaviateClass() {
+ WeaviateClass weaviateClass = WeaviateClass.builder()
+ .className(weaviateClassName)
+ .description("学习向量数据存储")
+ .vectorizer("none") // 使用自定义向量
+ .properties(Arrays.asList(
+ Property.builder()
+ .name("userId")
+ .dataType(Arrays.asList(DataType.TEXT))
+ .description("用户ID")
+ .build(),
+ Property.builder()
+ .name("questionId")
+ .dataType(Arrays.asList(DataType.TEXT))
+ .description("题目ID")
+ .build(),
+ Property.builder()
+ .name("actionType")
+ .dataType(Arrays.asList(DataType.TEXT))
+ .description("操作类型")
+ .build(),
+ Property.builder()
+ .name("questionContent")
+ .dataType(Arrays.asList(DataType.TEXT))
+ .description("题目内容")
+ .build(),
+ Property.builder()
+ .name("subject")
+ .dataType(Arrays.asList(DataType.TEXT))
+ .description("科目")
+ .build(),
+ Property.builder()
+ .name("knowledgePointId")
+ .dataType(Arrays.asList(DataType.INT))
+ .description("知识点ID")
+ .build(),
+ Property.builder()
+ .name("metadata")
+ .dataType(Arrays.asList(DataType.TEXT))
+ .description("元数据JSON")
+ .build(),
+ Property.builder()
+ .name("createdAt")
+ .dataType(Arrays.asList(DataType.TEXT))
+ .description("创建时间")
+ .build()
+ ))
+ .build();
+
+ Result result = weaviateClient.schema().classCreator().withClass(weaviateClass).run();
+ if (result.hasErrors()) {
+ throw new RuntimeException("创建Weaviate类失败: " + result.getError().getMessages());
+ }
+ }
+
+ @Override
+ public boolean storeLearningVector(String userId, String questionId, String questionContent,
+ String actionType, String subject, Integer knowledgePointId) {
+ try {
+ // 构建向量化文本
+ String vectorText = buildVectorText(actionType, questionContent, subject);
+
+ // 生成向量
+ float[] embedding = embeddingService.generateEmbedding(vectorText);
+
+ // 转换为Float[]数组
+ Float[] embeddingArray = new Float[embedding.length];
+ for (int i = 0; i < embedding.length; i++) {
+ embeddingArray[i] = embedding[i];
+ }
+
+ // 构建元数据
+ Map metadata = buildMetadata(actionType, subject, knowledgePointId);
+ String metadataJson = objectMapper.writeValueAsString(metadata);
+
+ // 构建Weaviate对象
+ Map properties = new HashMap<>();
+ properties.put("userId", userId);
+ properties.put("questionId", questionId);
+ properties.put("actionType", actionType);
+ properties.put("questionContent", questionContent);
+ properties.put("subject", subject);
+ properties.put("knowledgePointId", knowledgePointId);
+ properties.put("metadata", metadataJson);
+ properties.put("createdAt", LocalDateTime.now().format(DateTimeFormatter.ISO_LOCAL_DATE_TIME));
+
+ // 存储到Weaviate
+ Result result = weaviateClient.data().creator()
+ .withClassName(weaviateClassName)
+ .withProperties(properties)
+ .withVector(embeddingArray)
+ .run();
+
+ if (result.hasErrors()) {
+ log.error("=== Weaviate存储错误 ===");
+ log.error("错误信息: {}", result.getError().getMessages());
+ log.error("用户ID: {}", userId);
+ log.error("题目ID: {}", questionId);
+ log.error("======================");
+ return false;
+ }
+
+ log.info("=== 学习向量存储成功 ===");
+ log.info("用户ID: {}", userId);
+ log.info("题目ID: {}", questionId);
+ log.info("行为类型: {}", actionType);
+ log.info("科目: {}", subject != null ? subject : "未知");
+ log.info("题目内容: {}", questionContent != null ?
+ (questionContent.length() > 100 ? questionContent.substring(0, 100) + "..." : questionContent) : "无");
+ log.info("向量维度: {}", embeddingArray.length);
+ log.info("存储时间: {}", LocalDateTime.now().format(DateTimeFormatter.ISO_LOCAL_DATE_TIME));
+ log.info("========================");
+ return true;
+
+ } catch (Exception e) {
+ log.error("=== 学习向量存储失败 ===");
+ log.error("用户ID: {}", userId);
+ log.error("题目ID: {}", questionId);
+ log.error("行为类型: {}", actionType);
+ log.error("错误信息: {}", e.getMessage());
+ log.error("========================");
+ return false;
+ }
+ }
+
+ @Override
+ public List searchSimilarQuestions(String userId, String queryText, int limit) {
+ try {
+ // 生成查询向量
+ float[] queryEmbedding = embeddingService.generateEmbedding(queryText);
+
+ // 转换为Float[]数组
+ Float[] queryEmbeddingArray = new Float[queryEmbedding.length];
+ for (int i = 0; i < queryEmbedding.length; i++) {
+ queryEmbeddingArray[i] = queryEmbedding[i];
+ }
+
+ // 构建GraphQL查询
+ Field[] fields = {
+ Field.builder().name("userId").build(),
+ Field.builder().name("questionId").build(),
+ Field.builder().name("questionContent").build(),
+ Field.builder().name("actionType").build(),
+ Field.builder().name("subject").build(),
+ Field.builder().name("createdAt").build(),
+ Field.builder().name("_additional").fields(
+ Field.builder().name("distance").build()
+ ).build()
+ };
+
+ NearVectorArgument nearVector = NearVectorArgument.builder()
+ .vector(queryEmbeddingArray)
+ .build();
+
+ io.weaviate.client.v1.filters.WhereFilter whereFilter = io.weaviate.client.v1.filters.WhereFilter.builder()
+ .path(new String[]{"userId"})
+ .operator(io.weaviate.client.v1.filters.Operator.Equal)
+ .valueText(userId)
+ .build();
+
+ Result result = weaviateClient.graphQL().get()
+ .withClassName(weaviateClassName)
+ .withFields(fields)
+ .withNearVector(nearVector)
+ .withWhere(whereFilter)
+ .withLimit(limit)
+ .run();
+
+ if (result.hasErrors()) {
+ log.error("Weaviate相似度搜索失败: {}", result.getError().getMessages());
+ return Collections.emptyList();
+ }
+
+ return parseSearchResults(result.getResult());
+
+ } catch (Exception e) {
+ log.error("搜索相似题目失败，userId:{} queryText:{}", userId, queryText, e);
+ return Collections.emptyList();
+ }
+ }
+
+ @Override
+ public List getUserWeaknesses(String userId) {
+ try {
+ // 查询用户的错题数据
+ List> mistakeData = queryUserDataByActionType(userId, "mistake");
+
+ // 分析薄弱点
+ return analyzeWeaknesses(mistakeData);
+
+ } catch (Exception e) {
+ log.error("获取用户薄弱点失败，userId:{}", userId, e);
+ return Collections.emptyList();
+ }
+ }
+
+ @Override
+ public List getUserRecommendations(String userId) {
+ try {
+ // 获取用户所有学习数据
+ List> allData = queryAllUserData(userId);
+
+ // 生成学习推荐
+ return generateRecommendations(allData);
+
+ } catch (Exception e) {
+ log.error("获取用户学习推荐失败，userId:{}", userId, e);
+ return Collections.emptyList();
+ }
+ }
+
+ @Override
+ public void analyzeUserLearningPatterns(String userId) {
+ try {
+ log.info("开始分析用户学习模式，userId:{}", userId);
+
+ // 获取用户所有学习数据
+ List> allData = queryAllUserData(userId);
+
+ if (allData.isEmpty()) {
+ log.info("用户暂无学习数据，userId:{}", userId);
+ return;
+ }
+
+ // 分析学习模式并生成洞察
+ List insights = generateLearningInsights(allData);
+
+ // 这里可以将洞察结果存储到其他地方，比如MySQL或Redis
+ log.info("用户学习模式分析完成，userId:{} 生成洞察数量:{}", userId, insights.size());
+
+ } catch (Exception e) {
+ log.error("分析用户学习模式失败，userId:{}", userId, e);
+ }
+ }
+
+ /**
+ * 查询用户特定行为类型的数据
+ */
+ private List> queryUserDataByActionType(String userId, String actionType) {
+ try {
+ Field[] fields = {
+ Field.builder().name("userId").build(),
+ Field.builder().name("questionId").build(),
+ Field.builder().name("questionContent").build(),
+ Field.builder().name("actionType").build(),
+ Field.builder().name("subject").build(),
+ Field.builder().name("knowledgePointId").build(),
+ Field.builder().name("createdAt").build()
+ };
+
+ io.weaviate.client.v1.filters.WhereFilter whereFilter = io.weaviate.client.v1.filters.WhereFilter.builder()
+ .operator(io.weaviate.client.v1.filters.Operator.And)
+ .operands(new io.weaviate.client.v1.filters.WhereFilter[]{
+ io.weaviate.client.v1.filters.WhereFilter.builder()
+ .path(new String[]{"userId"})
+ .operator(io.weaviate.client.v1.filters.Operator.Equal)
+ .valueText(userId)
+ .build(),
+ io.weaviate.client.v1.filters.WhereFilter.builder()
+ .path(new String[]{"actionType"})
+ .operator(io.weaviate.client.v1.filters.Operator.Equal)
+ .valueText(actionType)
+ .build()
+ })
+ .build();
+
+ Result result = weaviateClient.graphQL().get()
+ .withClassName(weaviateClassName)
+ .withFields(fields)
+ .withWhere(whereFilter)
+ .withLimit(1000)
+ .run();
+
+ if (result.hasErrors()) {
+ log.error("查询用户数据失败: {}", result.getError().getMessages());
+ return Collections.emptyList();
+ }
+
+ return parseQueryResults(result.getResult());
+
+ } catch (Exception e) {
+ log.error("查询用户数据失败，userId:{} actionType:{}", userId, actionType, e);
+ return Collections.emptyList();
+ }
+ }
+
+ /**
+ * 查询用户所有数据
+ */
+ private List> queryAllUserData(String userId) {
+ try {
+ Field[] fields = {
+ Field.builder().name("userId").build(),
+ Field.builder().name("questionId").build(),
+ Field.builder().name("questionContent").build(),
+ Field.builder().name("actionType").build(),
+ Field.builder().name("subject").build(),
+ Field.builder().name("knowledgePointId").build(),
+ Field.builder().name("createdAt").build()
+ };
+
+ io.weaviate.client.v1.filters.WhereFilter whereFilter = io.weaviate.client.v1.filters.WhereFilter.builder()
+ .path(new String[]{"userId"})
+ .operator(io.weaviate.client.v1.filters.Operator.Equal)
+ .valueText(userId)
+ .build();
+
+ Result result = weaviateClient.graphQL().get()
+ .withClassName(weaviateClassName)
+ .withFields(fields)
+ .withWhere(whereFilter)
+ .withLimit(1000)
+ .run();
+
+ if (result.hasErrors()) {
+ log.error("查询用户所有数据失败: {}", result.getError().getMessages());
+ return Collections.emptyList();
+ }
+
+ return parseQueryResults(result.getResult());
+
+ } catch (Exception e) {
+ log.error("查询用户所有数据失败，userId:{}", userId, e);
+ return Collections.emptyList();
+ }
+ }
+
+ /**
+ * 解析搜索结果
+ */
+ @SuppressWarnings("unchecked")
+ private List parseSearchResults(GraphQLResponse response) {
+ List results = new ArrayList<>();
+
+ try {
+ Object dataObj = response.getData();
+ if (!(dataObj instanceof Map)) {
+ log.warn("GraphQL响应数据格式不正确");
+ return results;
+ }
+ Map data = (Map) dataObj;
+ if (data != null && data.containsKey("Get")) {
+ Map get = (Map) data.get("Get");
+ if (get.containsKey(weaviateClassName)) {
+ List> objects = (List>) get.get(weaviateClassName);
+
+ for (Map obj : objects) {
+ SimilarQuestionVO vo = SimilarQuestionVO.builder()
+ .questionId((String) obj.get("questionId"))
+ .questionContent((String) obj.get("questionContent"))
+ .actionType((String) obj.get("actionType"))
+ .subject((String) obj.get("subject"))
+ .createdAt((String) obj.get("createdAt"))
+ .build();
+
+ // 获取相似度分数
+ if (obj.containsKey("_additional")) {
+ Map additional = (Map) obj.get("_additional");
+ if (additional.containsKey("distance")) {
+ Double distance = (Double) additional.get("distance");
+ // 将距离转换为相似度 (1 - distance)
+ vo.setSimilarity(1.0 - distance);
+ }
+ }
+
+ results.add(vo);
+ }
+ }
+ }
+ } catch (Exception e) {
+ log.error("解析搜索结果失败", e);
+ }
+
+ return results;
+ }
+
+ /**
+ * 解析查询结果
+ */
+ @SuppressWarnings("unchecked")
+ private List> parseQueryResults(GraphQLResponse response) {
+ List> results = new ArrayList<>();
+
+ try {
+ Object dataObj = response.getData();
+ if (!(dataObj instanceof Map)) {
+ log.warn("GraphQL响应数据格式不正确");
+ return results;
+ }
+ Map data = (Map) dataObj;
+ if (data != null && data.containsKey("Get")) {
+ Map get = (Map) data.get("Get");
+ if (get.containsKey(weaviateClassName)) {
+ List> objects = (List>) get.get(weaviateClassName);
+ results.addAll(objects);
+ }
+ }
+ } catch (Exception e) {
+ log.error("解析查询结果失败", e);
+ }
+
+ return results;
+ }
+
+ // 以下方法与原VectorRepository中的实现类似，但使用Map而不是LearningVector
+
+ private String buildVectorText(String actionType, String questionContent, String subject) {
+ StringBuilder textBuilder = new StringBuilder();
+ textBuilder.append("用户").append(getActionDescription(actionType));
+ if (subject != null && !subject.trim().isEmpty()) {
+ textBuilder.append("了一道").append(subject).append("题目：");
+ } else {
+ textBuilder.append("了一道题目：");
+ }
+ if (questionContent != null && !questionContent.trim().isEmpty()) {
+ String content = questionContent.trim();
+ if (content.length() > 500) {
+ content = content.substring(0, 500) + "...";
+ }
+ textBuilder.append(content);
+ }
+ return textBuilder.toString();
+ }
+
+ private String getActionDescription(String actionType) {
+ switch (actionType) {
+ case "upload": return "上传";
+ case "review": return "复习";
+ case "qa": return "问答";
+ case "mistake": return "做错";
+ default: return "学习";
+ }
+ }
+
+ private Map buildMetadata(String actionType, String subject, Integer knowledgePointId) {
+ Map metadata = new HashMap<>();
+ metadata.put("actionType", actionType);
+ if (subject != null) metadata.put("subject", subject);
+ if (knowledgePointId != null) metadata.put("knowledgePointId", knowledgePointId);
+ metadata.put("timestamp", System.currentTimeMillis());
+ metadata.put("embeddingModel", "dashscope-text-embedding-v1");
+ return metadata;
+ }
+
+ private List analyzeWeaknesses(List> vectors) {
+ List insights = new ArrayList<>();
+ if (vectors.isEmpty()) return insights;
+
+ // 按科目分组分析
+ Map>> subjectGroups = vectors.stream()
+ .filter(v -> v.get("subject") != null && !v.get("subject").toString().trim().isEmpty())
+ .collect(Collectors.groupingBy(v -> v.get("subject").toString()));
+
+ for (Map.Entry>> entry : subjectGroups.entrySet()) {
+ String subject = entry.getKey();
+ List> subjectVectors = entry.getValue();
+
+ if (subjectVectors.size() >= 3) {
+ insights.add(LearningInsightVO.builder()
+ .type("weakness")
+ .title(subject + "学科薄弱")
+ .description(String.format("在%s学科中错误%d道题目，建议加强练习", subject, subjectVectors.size()))
+ .confidenceScore(Math.min(0.9, 0.5 + subjectVectors.size() * 0.1))
+ .relatedQuestions(subjectVectors.stream()
+ .map(v -> v.get("questionId").toString())
+ .collect(Collectors.toList()))
+ .createdAt(LocalDateTime.now())
+ .isActive(true)
+ .build());
+ }
+ }
+
+ return insights;
+ }
+
+ private List generateRecommendations(List> vectors) {
+ List recommendations = new ArrayList<>();
+
+ if (vectors.isEmpty()) {
+ recommendations.add(LearningInsightVO.builder()
+ .type("recommendation")
+ .title("开始学习之旅")
+ .description("还没有学习记录，建议开始上传题目进行学习")
+ .confidenceScore(1.0)
+ .createdAt(LocalDateTime.now())
+ .isActive(true)
+ .build());
+ return recommendations;
+ }
+
+ // 分析学习频率等逻辑...
+ return recommendations;
+ }
+
+ private List generateLearningInsights(List> vectors) {
+ List insights = new ArrayList<>();
+ // 生成学习洞察的逻辑...
+ return insights;
+ }
+}
\ No newline at end of file
diff --git a/refine-trigger/src/main/java/com/achobeta/trigger/event/UserBehaviorEventListener.java b/refine-trigger/src/main/java/com/achobeta/trigger/event/UserBehaviorEventListener.java
new file mode 100644
index 0000000..86a97e4
--- /dev/null
+++ b/refine-trigger/src/main/java/com/achobeta/trigger/event/UserBehaviorEventListener.java
@@ -0,0 +1,162 @@
+package com.achobeta.trigger.event;
+
+import com.achobeta.domain.rag.service.IVectorService;
+import com.achobeta.domain.user.event.UserBehaviorEvent;
+import com.achobeta.domain.user.event.UserMistakeEvent;
+import com.achobeta.domain.user.event.UserReviewEvent;
+import com.achobeta.domain.user.event.UserUploadEvent;
+import lombok.extern.slf4j.Slf4j;
+import org.springframework.beans.factory.annotation.Autowired;
+import org.springframework.context.event.EventListener;
+import org.springframework.scheduling.annotation.Async;
+import org.springframework.stereotype.Component;
+
+/**
+ * @Auth : Malog
+ * @Desc : 用户行为事件监听器
+ * @Time : 2025/11/28
+ */
+@Slf4j
+@Component
+public class UserBehaviorEventListener {
+
+ @Autowired
+ private IVectorService vectorService;
+
+ /**
+ * 处理用户复习事件
+ */
+ @EventListener
+ @Async
+ public void handleUserReviewEvent(UserReviewEvent event) {
+ try {
+ log.info("检测到用户复习事件，开始记录向量数据，userId:{}, questionId:{}",
+ event.getUserId(), event.getQuestionId());
+
+ boolean success = vectorService.storeLearningVector(
+ event.getUserId(),
+ event.getQuestionId(),
+ event.getQuestionContent(),
+ event.getActionType(),
+ event.getSubject(),
+ event.getKnowledgePointId()
+ );
+
+ if (success) {
+ log.info("用户复习行为向量记录成功，userId:{}, questionId:{}",
+ event.getUserId(), event.getQuestionId());
+ } else {
+ log.warn("用户复习行为向量记录失败，userId:{}, questionId:{}",
+ event.getUserId(), event.getQuestionId());
+ }
+
+ } catch (Exception e) {
+ log.error("处理用户复习事件失败，userId:{}, questionId:{}",
+ event.getUserId(), event.getQuestionId(), e);
+ }
+ }
+
+ /**
+ * 处理用户错题事件
+ */
+ @EventListener
+ @Async
+ public void handleUserMistakeEvent(UserMistakeEvent event) {
+ try {
+ log.info("检测到用户错题事件，开始记录向量数据，userId:{}, questionId:{}",
+ event.getUserId(), event.getQuestionId());
+
+ boolean success = vectorService.storeLearningVector(
+ event.getUserId(),
+ event.getQuestionId(),
+ event.getQuestionContent(),
+ event.getActionType(),
+ event.getSubject(),
+ event.getKnowledgePointId()
+ );
+
+ if (success) {
+ log.info("用户错题行为向量记录成功，userId:{}, questionId:{}",
+ event.getUserId(), event.getQuestionId());
+ } else {
+ log.warn("用户错题行为向量记录失败，userId:{}, questionId:{}",
+ event.getUserId(), event.getQuestionId());
+ }
+
+ } catch (Exception e) {
+ log.error("处理用户错题事件失败，userId:{}, questionId:{}",
+ event.getUserId(), event.getQuestionId(), e);
+ }
+ }
+
+ /**
+ * 处理用户上传事件
+ */
+ @EventListener
+ @Async
+ public void handleUserUploadEvent(UserUploadEvent event) {
+ try {
+ log.info("检测到用户上传事件，开始记录向量数据，userId:{}, questionId:{}",
+ event.getUserId(), event.getQuestionId());
+
+ boolean success = vectorService.storeLearningVector(
+ event.getUserId(),
+ event.getQuestionId(),
+ event.getQuestionContent(),
+ event.getActionType(),
+ event.getSubject(),
+ event.getKnowledgePointId()
+ );
+
+ if (success) {
+ log.info("用户上传行为向量记录成功，userId:{}, questionId:{}",
+ event.getUserId(), event.getQuestionId());
+ } else {
+ log.warn("用户上传行为向量记录失败，userId:{}, questionId:{}",
+ event.getUserId(), event.getQuestionId());
+ }
+
+ } catch (Exception e) {
+ log.error("处理用户上传事件失败，userId:{}, questionId:{}",
+ event.getUserId(), event.getQuestionId(), e);
+ }
+ }
+
+ /**
+ * 处理通用用户行为事件
+ */
+ @EventListener
+ @Async
+ public void handleUserBehaviorEvent(UserBehaviorEvent event) {
+ try {
+ // 避免重复处理已经有专门监听器的事件
+ if (event instanceof UserReviewEvent || event instanceof UserMistakeEvent || event instanceof UserUploadEvent) {
+ return;
+ }
+
+ log.info("检测到用户行为事件，开始记录向量数据，userId:{}, actionType:{}, questionId:{}",
+ event.getUserId(), event.getActionType(), event.getQuestionId());
+
+ boolean success = vectorService.storeLearningVector(
+ event.getUserId(),
+ event.getQuestionId(),
+ event.getQuestionContent(),
+ event.getActionType(),
+ event.getSubject(),
+ event.getKnowledgePointId()
+ );
+
+ if (success) {
+ log.info("用户行为向量记录成功，userId:{}, actionType:{}, questionId:{}",
+ event.getUserId(), event.getActionType(), event.getQuestionId());
+ } else {
+ log.warn("用户行为向量记录失败，userId:{}, actionType:{}, questionId:{}",
+ event.getUserId(), event.getActionType(), event.getQuestionId());
+ }
+
+ } catch (Exception e) {
+ log.error("处理用户行为事件失败，userId:{}, actionType:{}, questionId:{}",
+ event.getUserId(), event.getActionType(), event.getQuestionId(), e);
+ }
+ }
+}
\ No newline at end of file
diff --git a/refine-trigger/src/main/java/com/achobeta/trigger/http/LearningAnalysisController.java b/refine-trigger/src/main/java/com/achobeta/trigger/http/LearningAnalysisController.java
index 664aa1e..f52c315 100644
--- a/refine-trigger/src/main/java/com/achobeta/trigger/http/LearningAnalysisController.java
+++ b/refine-trigger/src/main/java/com/achobeta/trigger/http/LearningAnalysisController.java
@@ -13,6 +13,7 @@
import com.achobeta.types.common.UserContext;
import lombok.extern.slf4j.Slf4j;
import org.springframework.beans.factory.annotation.Autowired;
+import org.springframework.beans.factory.annotation.Qualifier;
import org.springframework.web.bind.annotation.*;
import java.util.List;
@@ -33,6 +34,7 @@ public class LearningAnalysisController {
private LearningAnalysisService learningAnalysisService;
@Autowired
+ @Qualifier("weaviateVectorRepository")
private IVectorService vectorService;
/**
diff --git a/refine-trigger/src/main/java/com/achobeta/trigger/http/OcrController.java b/refine-trigger/src/main/java/com/achobeta/trigger/http/OcrController.java
index be21e90..158b7a0 100644
--- a/refine-trigger/src/main/java/com/achobeta/trigger/http/OcrController.java
+++ b/refine-trigger/src/main/java/com/achobeta/trigger/http/OcrController.java
@@ -75,8 +75,6 @@ public Response extractFirst(@RequestPart("file") Multi
// 调用OCR服务提取文件中的第一个题目
QuestionEntity questionEntity = ocrService.extractQuestionContent(userId, file.getBytes(), ft);
- //TODO
- // ai根据向量库（和mysql知识点表里的完全一致）（新建一个表）查询有没有相似知识点，再返回知识点名称
AiGenerationService.knowledgePoint knowledgePoint = aiGenerationService.knowledgeAnalysis(questionEntity.getQuestionText());
if (knowledgePoint == null || knowledgePoint.isEmpty()) {
log.warn("ai生成的知识点为空，请检查ai生成知识点的逻辑");
From 7b5e904e98902881153a657681b4fde78f731377 Mon Sep 17 00:00:00 2001
From: malog <2153315236@qq.com>
Date: Mon, 1 Dec 2025 15:24:49 +0800
Subject: [PATCH 2/3] =?UTF-8?q?feat:=20=E6=9B=B4=E6=96=B0=E9=85=8D?=
=?UTF-8?q?=E7=BD=AE=E6=96=87=E4=BB=B6?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
---
refine-app/src/main/resources/application-dev.yml | 14 ++++++++------
1 file changed, 8 insertions(+), 6 deletions(-)
diff --git a/refine-app/src/main/resources/application-dev.yml b/refine-app/src/main/resources/application-dev.yml
index a6fe14b..729ff98 100644
--- a/refine-app/src/main/resources/application-dev.yml
+++ b/refine-app/src/main/resources/application-dev.yml
@@ -125,13 +125,15 @@ redis:
ping-interval: 60000
keep-alive: true
-pgvector:
+# Weaviate 向量数据库配置
+weaviate:
config:
- host: ${PGVECTOR_IP:postgres}
- port: 5432
- database: postgres
- user: root
- password: 123456
+ host: ${WEAVIATE_HOST:fmamdtzprxs1hi3olvplxg.c0.asia-southeast1.gcp.weaviate.cloud}
+ port: ${WEAVIATE_PORT:443}
+ scheme: ${WEAVIATE_SCHEME:https}
+ api-key: ${WEAVIATE_API_KEY:YWtxam9aYkxzbFBkdWlCdV9IOThJWHdHdGNHSUplTE9VeERtSWk4a216emN3Z3pXVnUzNVZKSG9JeEFZPV92MjAw} # Weaviate Cloud API Key
+ timeout: 30000 # 连接超时时间（毫秒）
+ class-name: LearningVector # Weaviate中的类名
# MyBatis 配置【如需使用记得打开】
mybatis:
From 54b997657a2afe017a4066bdd6307863e13fdd2c Mon Sep 17 00:00:00 2001
From: malog <2153315236@qq.com>
Date: Mon, 1 Dec 2025 15:26:17 +0800
Subject: [PATCH 3/3] =?UTF-8?q?chore:=20=E9=87=8D=E6=9E=84=E9=A1=B9?=
=?UTF-8?q?=E7=9B=AE=E7=BB=93=E6=9E=84?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
---
.../resources/mybatis/mapper/VectorMapper.xml | 112 -----
.../adapter/repository/VectorRepository.java | 421 ------------------
.../infrastructure/dao/vector/IVectorDao.java | 73 ---
3 files changed, 606 deletions(-)
delete mode 100644 refine-app/src/main/resources/mybatis/mapper/VectorMapper.xml
delete mode 100644 refine-infrastructure/src/main/java/com/achobeta/infrastructure/adapter/repository/VectorRepository.java
delete mode 100644 refine-infrastructure/src/main/java/com/achobeta/infrastructure/dao/vector/IVectorDao.java
diff --git a/refine-app/src/main/resources/mybatis/mapper/VectorMapper.xml b/refine-app/src/main/resources/mybatis/mapper/VectorMapper.xml
deleted file mode 100644
index 43e0953..0000000
--- a/refine-app/src/main/resources/mybatis/mapper/VectorMapper.xml
+++ /dev/null
@@ -1,112 +0,0 @@
-
-
-
-
-
-
- INSERT INTO user_learning_vectors
- (user_id, question_id, action_type, question_content, subject, knowledge_point_id)
- VALUES
- (#{userId}, #{questionId}, #{actionType}, #{questionContent}, #{subject}, #{knowledgePointId})
-
-
-
-
-
-
-
-
-
-
-
-
-
- INSERT INTO learning_insights
- (user_id, insight_type, title, description, confidence_score)
- VALUES
- (#{userId}, #{insight.type}, #{insight.title}, #{insight.description}, #{insight.confidenceScore})
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
\ No newline at end of file
diff --git a/refine-infrastructure/src/main/java/com/achobeta/infrastructure/adapter/repository/VectorRepository.java b/refine-infrastructure/src/main/java/com/achobeta/infrastructure/adapter/repository/VectorRepository.java
deleted file mode 100644
index 4ea2adc..0000000
--- a/refine-infrastructure/src/main/java/com/achobeta/infrastructure/adapter/repository/VectorRepository.java
+++ /dev/null
@@ -1,421 +0,0 @@
-package com.achobeta.infrastructure.adapter.repository;
-
-import com.achobeta.domain.rag.model.valobj.LearningInsightVO;
-import com.achobeta.domain.rag.model.valobj.SimilarQuestionVO;
-import com.achobeta.domain.rag.service.IVectorService;
-import com.achobeta.infrastructure.dao.vector.IVectorDao;
-import com.achobeta.infrastructure.dao.po.LearningVector;
-import com.achobeta.infrastructure.gateway.DashScopeEmbeddingService;
-import lombok.extern.slf4j.Slf4j;
-import org.springframework.beans.factory.annotation.Autowired;
-import org.springframework.stereotype.Repository;
-
-import java.util.*;
-import java.util.stream.Collectors;
-
-import static com.achobeta.types.enums.ActionType.MISTAKE;
-
-/**
- * 向量数据库适配器实现
- */
-@Slf4j
-@Repository
-public class VectorRepository implements IVectorService {
-
- @Autowired
- private IVectorDao vectorDao;
-
- @Autowired
- private DashScopeEmbeddingService embeddingService; // 使用DashScope嵌入服务
-
- public VectorRepository(DashScopeEmbeddingService embeddingService) {
- this.embeddingService = embeddingService;
- }
-
- /**
- * 存储学习向量信息
- *
- * @param userId 用户ID
- * @param questionId 题目ID
- * @param questionContent 题目内容
- * @param actionType 操作类型
- * @param subject 科目
- * @param knowledgePointId 知识点ID
- * @return 存储成功返回true，失败返回false
- */
- @Override
- public boolean storeLearningVector(String userId, String questionId, String questionContent,
- String actionType, String subject, Integer knowledgePointId) {
- try {
- // 构建向量化文本
- String vectorText = buildVectorText(actionType, questionContent, subject);
-
- // 生成向量
- float[] embedding = embeddingService.generateEmbedding(vectorText);
-
- // 构建数据库实体
- LearningVector vector = LearningVector.builder()
- .userId(userId)
- .questionId(questionId)
- .actionType(actionType)
- .questionContent(questionContent)
- .subject(subject)
- .knowledgePointId(knowledgePointId)
- .embedding(embedding)
- .metadata(buildMetadata(actionType, subject, knowledgePointId))
- .build();
-
- int result = vectorDao.insertLearningVector(vector);
-
- log.info("成功存储学习向量，userId:{} questionId:{} actionType:{}",
- userId, questionId, actionType);
- return result > 0;
-
- } catch (Exception e) {
- log.error("存储学习向量失败，userId:{} questionId:{}", userId, questionId, e);
- return false;
- }
- }
-
-
- /**
- * 搜索与给定文本相似的题目。
- *
- * @param userId 用户ID，用于限定搜索范围
- * @param queryText 查询文本，将被转换为向量进行相似度匹配
- * @param limit 返回结果的最大数量限制
- * @return 相似的题目列表，封装为 SimilarQuestionVO 对象；若发生异常则返回空列表
- */
- @Override
- public List searchSimilarQuestions(String userId, String queryText, int limit) {
- try {
- // 生成查询向量
- float[] queryEmbedding = embeddingService.generateEmbedding(queryText);
-
- // 执行向量相似度搜索
- List similarVectors = vectorDao.searchSimilarVectors(
- userId, queryEmbedding, limit);
-
- // 转换为值对象
- return similarVectors.stream()
- .map(this::convertToSimilarQuestionVO)
- .collect(Collectors.toList());
-
- } catch (Exception e) {
- log.error("搜索相似题目失败，userId:{} queryText:{}", userId, queryText, e);
- return Collections.emptyList();
- }
- }
-
- /**
- * 获取用户的薄弱知识点分析结果。
- *
- * @param userId 用户ID
- * @return 用户的薄弱点信息列表，封装为 LearningInsightVO 对象；若发生异常则返回空列表
- */
- @Override
- public List getUserWeaknesses(String userId) {
- try {
- // 分析用户错题模式
- List userVectors = vectorDao.getUserVectorsByActionType(userId, MISTAKE.getActionType());
-
- // 使用聚类分析找出薄弱点
- return analyzeWeaknesses(userVectors);
-
- } catch (Exception e) {
- log.error("获取用户薄弱点失败，userId:{}", userId, e);
- return Collections.emptyList();
- }
- }
-
- /**
- * 根据用户的学习行为数据生成个性化推荐内容。
- *
- * @param userId 用户ID
- * @return 推荐内容列表，封装为 LearningInsightVO 对象；若发生异常则返回空列表
- */
- @Override
- public List getUserRecommendations(String userId) {
- try {
- // 获取用户所有学习向量
- List allVectors = vectorDao.getUserAllVectors(userId);
-
- // 生成学习推荐
- return generateRecommendations(allVectors);
-
- } catch (Exception e) {
- log.error("获取用户学习推荐失败，userId:{}", userId, e);
- return Collections.emptyList();
- }
- }
-
- /**
- * 分析指定用户的学习行为模式，并将分析结果保存到数据库中。
- *
- * @param userId 用户ID
- */
- @Override
- public void analyzeUserLearningPatterns(String userId) {
- try {
- log.info("开始分析用户学习模式，userId:{}", userId);
-
- // 获取用户所有学习向量
- List allVectors = vectorDao.getUserAllVectors(userId);
-
- if (allVectors.isEmpty()) {
- log.info("用户暂无学习数据，userId:{}", userId);
- return;
- }
-
- // 分析学习模式并生成洞察
- List insights = generateLearningInsights(allVectors);
-
- // 保存分析结果
- for (LearningInsightVO insight : insights) {
- vectorDao.saveLearningInsight(userId, insight);
- }
- log.info("用户学习模式分析完成，userId:{} 生成洞察数量:{}", userId, insights.size());
-
- } catch (Exception e) {
- log.error("分析用户学习模式失败，userId:{}", userId, e);
- }
- }
-
-
- /**
- * 构建向量化文本
- */
- private String buildVectorText(String actionType, String questionContent, String subject) {
- StringBuilder textBuilder = new StringBuilder();
-
- // 添加行为描述
- textBuilder.append("用户").append(getActionDescription(actionType));
-
- // 添加科目信息
- if (subject != null && !subject.trim().isEmpty()) {
- textBuilder.append("了一道").append(subject).append("题目：");
- } else {
- textBuilder.append("了一道题目：");
- }
-
- // 添加题目内容
- if (questionContent != null && !questionContent.trim().isEmpty()) {
- // 限制题目内容长度，避免向量化文本过长
- String content = questionContent.trim();
- if (content.length() > 500) {
- content = content.substring(0, 500) + "...";
- }
- textBuilder.append(content);
- }
-
- return textBuilder.toString();
- }
-
- /**
- * 获取行为类型描述
- */
- private String getActionDescription(String actionType) {
- switch (actionType) {
- case "upload":
- return "上传";
- case "review":
- return "复习";
- case "qa":
- return "问答";
- case "mistake":
- return "做错";
- default:
- return "学习";
- }
- }
-
- /**
- * 构建元数据
- */
- private Map buildMetadata(String actionType, String subject, Integer knowledgePointId) {
- Map metadata = new HashMap<>();
- metadata.put("actionType", actionType);
- if (subject != null) metadata.put("subject", subject);
- if (knowledgePointId != null) metadata.put("knowledgePointId", knowledgePointId);
- metadata.put("timestamp", System.currentTimeMillis());
- metadata.put("embeddingModel", "dashscope-text-embedding-v1");
- return metadata;
- }
-
- /**
- * 转换为相似题目值对象
- */
- private SimilarQuestionVO convertToSimilarQuestionVO(LearningVector vector) {
- return SimilarQuestionVO.builder()
- .questionId(vector.getQuestionId())
- .questionContent(vector.getQuestionContent())
- .actionType(vector.getActionType())
- .subject(vector.getSubject())
- .similarity(vector.getSimilarity())
- .createdAt(vector.getCreatedAt() != null ? vector.getCreatedAt().toString() : null)
- .build();
- }
-
- /**
- * 分析薄弱点
- */
- private List analyzeWeaknesses(List vectors) {
- List insights = new ArrayList<>();
-
- if (vectors.isEmpty()) {
- return insights;
- }
-
- // 按科目分组分析
- Map> subjectGroups = vectors.stream()
- .filter(v -> v.getSubject() != null && !v.getSubject().trim().isEmpty())
- .collect(Collectors.groupingBy(LearningVector::getSubject));
-
- for (Map.Entry> entry : subjectGroups.entrySet()) {
- String subject = entry.getKey();
- List subjectVectors = entry.getValue();
-
- if (subjectVectors.size() >= 3) { // 至少3道错题才认为是薄弱点
- insights.add(LearningInsightVO.builder()
- .type("weakness")
- .title(subject + "学科薄弱")
- .description(String.format("在%s学科中错误%d道题目，建议加强练习",
- subject, subjectVectors.size()))
- .confidenceScore(Math.min(0.9, 0.5 + subjectVectors.size() * 0.1))
- .relatedQuestions(subjectVectors.stream()
- .map(LearningVector::getQuestionId)
- .collect(Collectors.toList()))
- .createdAt(java.time.LocalDateTime.now())
- .isActive(true)
- .build());
- }
- }
-
- // 分析总体错题频率
- if (vectors.size() > 10) {
- insights.add(LearningInsightVO.builder()
- .type("weakness")
- .title("错题频率较高")
- .description(String.format("总共有%d道错题，建议系统性复习基础知识", vectors.size()))
- .confidenceScore(0.8)
- .relatedQuestions(vectors.stream()
- .limit(10)
- .map(LearningVector::getQuestionId)
- .collect(Collectors.toList()))
- .createdAt(java.time.LocalDateTime.now())
- .isActive(true)
- .build());
- }
-
- return insights;
- }
-
- /**
- * 生成学习推荐
- */
- private List generateRecommendations(List vectors) {
- List recommendations = new ArrayList<>();
-
- if (vectors.isEmpty()) {
- recommendations.add(LearningInsightVO.builder()
- .type("recommendation")
- .title("开始学习之旅")
- .description("还没有学习记录，建议开始上传题目进行学习")
- .confidenceScore(1.0)
- .createdAt(java.time.LocalDateTime.now())
- .isActive(true)
- .build());
- return recommendations;
- }
-
- // 分析学习频率
- long recentWeekCount = vectors.stream()
- .filter(v -> v.getCreatedAt() != null &&
- v.getCreatedAt().isAfter(java.time.LocalDateTime.now().minusWeeks(1)))
- .count();
-
- if (recentWeekCount > 10) {
- recommendations.add(LearningInsightVO.builder()
- .type("recommendation")
- .title("保持学习节奏")
- .description("近一周学习活跃，建议继续保持良好的学习习惯")
- .confidenceScore(0.9)
- .createdAt(java.time.LocalDateTime.now())
- .isActive(true)
- .build());
- } else if (recentWeekCount < 3) {
- recommendations.add(LearningInsightVO.builder()
- .type("recommendation")
- .title("增加学习频率")
- .description("建议增加学习时间，每天至少学习30分钟")
- .confidenceScore(0.8)
- .createdAt(java.time.LocalDateTime.now())
- .isActive(true)
- .build());
- }
-
- // 分析科目分布
- Map subjectCounts = vectors.stream()
- .filter(v -> v.getSubject() != null && !v.getSubject().trim().isEmpty())
- .collect(Collectors.groupingBy(LearningVector::getSubject, Collectors.counting()));
-
- if (subjectCounts.size() == 1) {
- String subject = subjectCounts.keySet().iterator().next();
- recommendations.add(LearningInsightVO.builder()
- .type("recommendation")
- .title("扩展学习科目")
- .description(String.format("目前主要学习%s，建议尝试其他科目的学习", subject))
- .confidenceScore(0.7)
- .createdAt(java.time.LocalDateTime.now())
- .isActive(true)
- .build());
- }
-
- return recommendations;
- }
-
- /**
- * 生成学习洞察
- */
- private List generateLearningInsights(List vectors) {
- List insights = new ArrayList<>();
-
- // 分析学习频率
- long recentWeekCount = vectors.stream()
- .filter(v -> v.getCreatedAt() != null &&
- v.getCreatedAt().isAfter(java.time.LocalDateTime.now().minusWeeks(1)))
- .count();
-
- if (recentWeekCount > 15) {
- insights.add(LearningInsightVO.builder()
- .type("strength")
- .title("学习积极性高")
- .description("近一周学习非常活跃，保持良好的学习习惯")
- .confidenceScore(0.9)
- .createdAt(java.time.LocalDateTime.now())
- .isActive(true)
- .build());
- }
-
- // 分析学习连续性
- long activeDays = vectors.stream()
- .filter(v -> v.getCreatedAt() != null &&
- v.getCreatedAt().isAfter(java.time.LocalDateTime.now().minusWeeks(1)))
- .map(v -> v.getCreatedAt().toLocalDate())
- .distinct()
- .count();
-
- if (activeDays >= 5) {
- insights.add(LearningInsightVO.builder()
- .type("achievement")
- .title("学习坚持性优秀")
- .description(String.format("近一周有%d天进行了学习，坚持性很好", activeDays))
- .confidenceScore(0.8)
- .createdAt(java.time.LocalDateTime.now())
- .isActive(true)
- .build());
- }
-
- return insights;
- }
-}
\ No newline at end of file
diff --git a/refine-infrastructure/src/main/java/com/achobeta/infrastructure/dao/vector/IVectorDao.java b/refine-infrastructure/src/main/java/com/achobeta/infrastructure/dao/vector/IVectorDao.java
deleted file mode 100644
index 095f764..0000000
--- a/refine-infrastructure/src/main/java/com/achobeta/infrastructure/dao/vector/IVectorDao.java
+++ /dev/null
@@ -1,73 +0,0 @@
-package com.achobeta.infrastructure.dao.vector;
-
-import com.achobeta.infrastructure.dao.po.LearningVector;
-import com.achobeta.domain.rag.model.valobj.LearningInsightVO;
-import org.apache.ibatis.annotations.Mapper;
-import org.apache.ibatis.annotations.Param;
-
-import java.util.List;
-import java.util.Map;
-
-/**
- * @Auth : Malog
- * @Desc : 向量数据访问接口
- * @Time : 2025/11/10
- */
-@Mapper
-public interface IVectorDao {
-
- /**
- * 插入学习向量
- */
- int insertLearningVector(LearningVector vector);
-
- /**
- * 向量相似度搜索
- */
- List searchSimilarVectors(@Param("userId") String userId,
- @Param("queryVector") float[] queryVector,
- @Param("limit") int limit);
-
- /**
- * 根据行为类型获取用户向量
- */
- List getUserVectorsByActionType(@Param("userId") String userId,
- @Param("actionType") String actionType);
-
- /**
- * 获取用户所有向量
- */
- List getUserAllVectors(@Param("userId") String userId);
-
- /**
- * 保存学习洞察
- */
- int saveLearningInsight(@Param("userId") String userId,
- @Param("insight") LearningInsightVO insight);
-
- /**
- * 获取用户学习洞察
- */
- List getUserLearningInsights(@Param("userId") String userId,
- @Param("insightType") String insightType);
-
- /**
- * 获取用户最近N天的学习向量数据
- */
- List getUserRecentLearningData(@Param("userId") String userId, @Param("days") int days);
-
- /**
- * 获取用户最近N天的学习统计数据
- */
- Map getUserLearningStatistics(@Param("userId") String userId, @Param("days") int days);
-
- /**
- * 获取用户最近N天按科目分组的学习数据
- */
- List> getUserLearningDataBySubject(@Param("userId") String userId, @Param("days") int days);
-
- /**
- * 获取用户最近N天按行为类型分组的学习数据
- */
- List> getUserLearningDataByActionType(@Param("userId") String userId, @Param("days") int days);
-}
\ No newline at end of file