JiaoTianBo/ylhp-ai-project-manager
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

feat(ai-knowledge-base): 实现AI知识库文档上传与管理功能

Browse Source 
- 新增AiDocument实体类，映射数据库ai_document表结构
- 添加AiDocumentMapper接口，提供文档增删改查及状态更新等数据库操作
- 实现AiKnowledgeBaseService接口及其实现类AiKnowledgeBaseServiceImpl，支持文件上传、文档列表查询、删除和重新索引
- 在AiKnowledgeBaseController中提供REST接口支持文件上传、文档管理和异步重新索引操作
- 实现DocumentProcessor组件，负责文档解析、文本切片、摘要生成和向量化存储
- 集成MinioService实现文件的上传、下载和删除操作
- 设计KbDocumentVO作为知识库文档视图对象，方便接口数据传输和展示
- 增加文件类型支持和上传文件校验，限制最大50MB文件大小
- 使用异步机制处理文档解析和向量化，提高系统处理性能和响应速度
- 实现文档状态管理和错误处理，确保文档处理流程的正确性和稳定性
This commit is contained in:
JiaoTianBo
2026-03-30 16:49:07 +08:00
parent d338490640
commit 9f972f5e30
7 changed files with 19 additions and 21 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
src/main/java/cn/yinlihupo/controller/ai/AiKnowledgeBaseController.java
View File

@@ -13,7 +13,6 @@ import org.springframework.web.bind.annotation.*;
import org.springframework.web.multipart.MultipartFile;
import java.util.List;
import java.util.UUID;
/**
* AI知识库控制器
@@ -98,7 +97,7 @@ public class AiKnowledgeBaseController {
*/
@DeleteMapping("/document/{docId}")
@Operation(summary = "删除文档", description = "删除指定的知识库文档")
public BaseResponse<Void> deleteDocument(@PathVariable UUID docId) {
public BaseResponse<Void> deleteDocument(@PathVariable String docId) {
Long userId = SecurityUtils.getCurrentUserId();
if (userId == null) {
return ResultUtils.error("用户未登录");
@@ -121,7 +120,7 @@ public class AiKnowledgeBaseController {
*/
@PostMapping("/document/{docId}/reindex")
@Operation(summary = "重新索引文档", description = "重新解析并索引指定的文档")
public BaseResponse<Void> reindexDocument(@PathVariable UUID docId) {
public BaseResponse<Void> reindexDocument(@PathVariable String docId) {
Long userId = SecurityUtils.getCurrentUserId();
if (userId == null) {
return ResultUtils.error("用户未登录");

3
src/main/java/cn/yinlihupo/domain/entity/AiDocument.java
View File

@@ -7,7 +7,6 @@ import lombok.Data;
import java.time.LocalDate;
import java.time.LocalDateTime;
import java.util.UUID;
/**
* AI文档向量实体
@@ -23,7 +22,7 @@ public class AiDocument {
/**
* 文档唯一标识(UUID)
*/
private UUID docId;
private String docId;
/**
* 关联项目ID

3
src/main/java/cn/yinlihupo/domain/vo/KbDocumentVO.java
View File

@@ -3,7 +3,6 @@ package cn.yinlihupo.domain.vo;
import lombok.Data;
import java.time.LocalDateTime;
import java.util.UUID;
/**
* 知识库文档VO
@@ -19,7 +18,7 @@ public class KbDocumentVO {
/**
* 文档UUID
*/
private UUID docId;
private String docId;
/**
* 文档标题

9
src/main/java/cn/yinlihupo/mapper/AiDocumentMapper.java
View File

@@ -8,7 +8,6 @@ import org.apache.ibatis.annotations.Mapper;
import org.apache.ibatis.annotations.Param;
import java.util.List;
import java.util.UUID;
/**
* AI文档向量Mapper
@@ -30,7 +29,7 @@ public interface AiDocumentMapper extends BaseMapper<AiDocument> {
* @param docId 文档UUID
* @return 文档实体
*/
AiDocument selectByDocId(@Param("docId") UUID docId);
AiDocument selectByDocId(@Param("docId") String docId);
/**
* 根据docId删除文档
@@ -38,7 +37,7 @@ public interface AiDocumentMapper extends BaseMapper<AiDocument> {
* @param docId 文档UUID
* @return 影响行数
*/
int deleteByDocId(@Param("docId") UUID docId);
int deleteByDocId(@Param("docId") String docId);
/**
* 批量查询引用文档信息
@@ -63,7 +62,7 @@ public interface AiDocumentMapper extends BaseMapper<AiDocument> {
* @param status 状态
* @return 影响行数
*/
int updateStatus(@Param("docId") UUID docId, @Param("status") String status);
int updateStatus(@Param("docId") String docId, @Param("status") String status);
/**
* 更新文档错误信息
@@ -72,7 +71,7 @@ public interface AiDocumentMapper extends BaseMapper<AiDocument> {
* @param errorMessage 错误信息
* @return 影响行数
*/
int updateErrorMessage(@Param("docId") UUID docId, @Param("errorMessage") String errorMessage);
int updateErrorMessage(@Param("docId") String docId, @Param("errorMessage") String errorMessage);
/**
* 增加文档查看次数

5
src/main/java/cn/yinlihupo/service/ai/AiKnowledgeBaseService.java
View File

@@ -4,7 +4,6 @@ import cn.yinlihupo.domain.vo.KbDocumentVO;
import org.springframework.web.multipart.MultipartFile;
import java.util.List;
import java.util.UUID;
/**
* AI知识库服务接口
@@ -35,7 +34,7 @@ public interface AiKnowledgeBaseService {
* @param docId 文档UUID
* @param userId 用户ID
*/
void deleteDocument(UUID docId, Long userId);
void deleteDocument(String docId, Long userId);
/**
* 重新索引文档
@@ -43,7 +42,7 @@ public interface AiKnowledgeBaseService {
* @param docId 文档UUID
* @param userId 用户ID
*/
void reindexDocument(UUID docId, Long userId);
void reindexDocument(String docId, Long userId);
/**
* 处理文档（解析、切片、向量化）

7
src/main/java/cn/yinlihupo/service/ai/impl/AiKnowledgeBaseServiceImpl.java
View File

@@ -39,7 +39,7 @@ public class AiKnowledgeBaseServiceImpl implements AiKnowledgeBaseService {
validateFile(file);
// 2. 生成文档UUID
UUID docId = UUID.randomUUID();
String docId = UUID.randomUUID().toString();
// 3. 上传文件到MinIO
String originalFilename = file.getOriginalFilename();
@@ -63,6 +63,7 @@ public class AiKnowledgeBaseServiceImpl implements AiKnowledgeBaseService {
doc.setFileType(fileExtension);
doc.setFileSize(file.getSize());
doc.setFilePath(filePath);
doc.setContent("");
doc.setStatus("pending"); // 待处理状态
doc.setChunkTotal(0);
doc.setCreateBy(userId);
@@ -86,7 +87,7 @@ public class AiKnowledgeBaseServiceImpl implements AiKnowledgeBaseService {
}
@Override
public void deleteDocument(UUID docId, Long userId) {
public void deleteDocument(String docId, Long userId) {
// 1. 查询文档
AiDocument doc = documentMapper.selectByDocId(docId);
if (doc == null) {
@@ -111,7 +112,7 @@ public class AiKnowledgeBaseServiceImpl implements AiKnowledgeBaseService {
}
@Override
public void reindexDocument(UUID docId, Long userId) {
public void reindexDocument(String docId, Long userId) {
// 1. 查询文档
AiDocument doc = documentMapper.selectByDocId(docId);
if (doc == null) {

8
src/main/java/cn/yinlihupo/service/ai/rag/DocumentProcessor.java
View File

@@ -13,7 +13,9 @@ import org.springframework.scheduling.annotation.Async;
import org.springframework.stereotype.Component;
import java.io.InputStream;
import java.util.*;
import java.util.Collections;
import java.util.List;
import java.util.Map;
import java.util.stream.Collectors;
/**
@@ -167,7 +169,7 @@ public class DocumentProcessor {
* @param chunks 切片列表
*/
private void storeChunks(AiDocument parentDoc, List<String> chunks) {
UUID docId = parentDoc.getDocId();
String docId = parentDoc.getDocId();
Long parentId = parentDoc.getId();
for (int i = 0; i < chunks.size(); i++) {
@@ -206,7 +208,7 @@ public class DocumentProcessor {
*
* @param docId 文档UUID
*/
public void deleteDocumentVectors(UUID docId) {
public void deleteDocumentVectors(String docId) {
try {
// 查询所有相关切片
// 注意：pgvector store的删除需要根据metadata过滤