feat(data-management): 添加数据集父子层级结构功能

- 在OpenAPI规范中新增parentDatasetId字段用于层级过滤
- 实现数据集父子关系的创建、更新和删除逻辑
- 添加数据集移动时的路径重命名和文件路径前缀更新
- 增加子数据集数量验证防止误删父数据集
- 更新前端界面支持选择父数据集和导航显示
- 优化Python后端自动标注任务的路径处理逻辑
- 修改数据库表结构添加外键约束确保数据一致性

backend/openapi/specs/data-management.yaml

@@ -61,12 +61,17 @@ paths:
           schema:
             type: string
           description: 关键词搜索（名称、描述）
-        - name: status
+        - name: status
-          in: query
+          in: query
-          schema:
+          schema:
-            type: string
+            type: string
-            enum: [DRAFT, ACTIVE, PROCESSING, ARCHIVED, PUBLISHED, DEPRECATED]
+            enum: [DRAFT, ACTIVE, PROCESSING, ARCHIVED, PUBLISHED, DEPRECATED]
-          description: 数据集状态过滤
+          description: 数据集状态过滤
+        - name: parentDatasetId
+          in: query
+          schema:
+            type: string
+          description: 父数据集ID过滤（传空字符串表示根数据集）
       responses:
         '200':
           description: 成功
@@ -538,15 +543,18 @@ components:
           type: boolean
           description: 是否为最后一页
 
-    DatasetResponse:
+    DatasetResponse:
-      type: object
+      type: object
-      properties:
+      properties:
-        id:
+        id:
-          type: string
+          type: string
-          description: 数据集ID
+          description: 数据集ID
-        name:
+        parentDatasetId:
-          type: string
+          type: string
-          description: 数据集名称
+          description: 父数据集ID
+        name:
+          type: string
+          description: 数据集名称
         description:
           type: string
           description: 数据集描述
@@ -590,29 +598,32 @@ components:
           type: string
           description: 创建者
 
-    CreateDatasetRequest:
+    CreateDatasetRequest:
-      type: object
+      type: object
-      required:
+      required:
-        - name
+        - name
-        - type
+        - type
-      properties:
+      properties:
-        name:
+        name:
-          type: string
+          type: string
-          description: 数据集名称
+          description: 数据集名称
-          minLength: 1
+          minLength: 1
-          maxLength: 100
+          maxLength: 100
-        description:
+        description:
-          type: string
+          type: string
-          description: 数据集描述
+          description: 数据集描述
-          maxLength: 500
+          maxLength: 500
-        type:
+        type:
-          type: string
+          type: string
-          description: 数据集类型
+          description: 数据集类型
-        tags:
+        parentDatasetId:
-          type: array
+          type: string
-          items:
+          description: 父数据集ID
-            type: string
+        tags:
-          description: 标签列表
+          type: array
+          items:
+            type: string
+          description: 标签列表
         dataSource:
           type: string
           description: 数据源
@@ -620,22 +631,25 @@ components:
           type: string
           description: 目标位置
 
-    UpdateDatasetRequest:
+    UpdateDatasetRequest:
-      type: object
+      type: object
-      properties:
+      properties:
-        name:
+        name:
-          type: string
+          type: string
-          description: 数据集名称
+          description: 数据集名称
-          maxLength: 100
+          maxLength: 100
-        description:
+        description:
-          type: string
+          type: string
-          description: 数据集描述
+          description: 数据集描述
-          maxLength: 500
+          maxLength: 500
-        tags:
+        parentDatasetId:
-          type: array
+          type: string
-          items:
+          description: 父数据集ID
-            type: string
+        tags:
-          description: 标签列表
+          type: array
+          items:
+            type: string
+          description: 标签列表
         status:
           type: string
           enum: [DRAFT, ACTIVE, PROCESSING, ARCHIVED, PUBLISHED, DEPRECATED]

backend/services/data-management-service/src/main/java/com/datamate/datamanagement/application/DatasetApplicationService.java

@@ -6,6 +6,9 @@ import com.datamate.common.domain.utils.ChunksSaver;
 import com.datamate.common.setting.application.SysParamApplicationService;
 import com.datamate.datamanagement.interfaces.dto.*;
 import com.datamate.common.infrastructure.exception.BusinessAssert;
+import com.datamate.common.infrastructure.exception.BusinessException;
+import com.datamate.common.infrastructure.exception.CommonErrorCode;
+import com.datamate.common.infrastructure.exception.SystemErrorCode;
 import com.datamate.common.interfaces.PagedResponse;
 import com.datamate.datamanagement.domain.model.dataset.Dataset;
 import com.datamate.datamanagement.domain.model.dataset.DatasetFile;
@@ -27,6 +30,7 @@ import org.springframework.stereotype.Service;
 import org.springframework.transaction.annotation.Transactional;
 import org.springframework.util.StringUtils;
 
+import java.io.File;
 import java.io.IOException;
 import java.nio.file.Files;
 import java.nio.file.Path;
@@ -62,7 +66,9 @@ public class DatasetApplicationService {
         BusinessAssert.isTrue(datasetRepository.findByName(createDatasetRequest.getName()) == null, DataManagementErrorCode.DATASET_ALREADY_EXISTS);
         // 创建数据集对象
         Dataset dataset = DatasetConverter.INSTANCE.convertToDataset(createDatasetRequest);
-        dataset.initCreateParam(datasetBasePath);
+        Dataset parentDataset = resolveParentDataset(createDatasetRequest.getParentDatasetId(), dataset.getId());
+        dataset.setParentDatasetId(parentDataset == null ? null : parentDataset.getId());
+        dataset.initCreateParam(datasetBasePath, parentDataset == null ? null : parentDataset.getPath());
         // 处理标签
         Set<Tag> processedTags = Optional.ofNullable(createDatasetRequest.getTags())
             .filter(CollectionUtils::isNotEmpty)
@@ -98,6 +104,9 @@ public class DatasetApplicationService {
         if (Objects.nonNull(updateDatasetRequest.getStatus())) {
             dataset.setStatus(updateDatasetRequest.getStatus());
         }
+        if (updateDatasetRequest.getParentDatasetId() != null) {
+            handleParentChange(dataset, updateDatasetRequest.getParentDatasetId());
+        }
         if (StringUtils.hasText(updateDatasetRequest.getDataSource())) {
             // 数据源id不为空，使用异步线程进行文件扫盘落库
             processDataSourceAsync(dataset.getId(), updateDatasetRequest.getDataSource());
@@ -112,10 +121,11 @@ public class DatasetApplicationService {
     @Transactional
     public void deleteDataset(String datasetId) {
         Dataset dataset = datasetRepository.getById(datasetId);
+        BusinessAssert.notNull(dataset, DataManagementErrorCode.DATASET_NOT_FOUND);
+        long childCount = datasetRepository.countByParentId(datasetId);
+        BusinessAssert.isTrue(childCount == 0, DataManagementErrorCode.DATASET_HAS_CHILDREN);
         datasetRepository.removeById(datasetId);
-        if (dataset != null) {
+        ChunksSaver.deleteFolder(dataset.getPath());
-            ChunksSaver.deleteFolder(dataset.getPath());
-        }
     }
 
     /**
@@ -164,6 +174,73 @@ public class DatasetApplicationService {
         return tags;
     }
 
+    private Dataset resolveParentDataset(String parentDatasetId, String currentDatasetId) {
+        String normalized = normalizeParentId(parentDatasetId);
+        if (normalized == null) {
+            return null;
+        }
+        BusinessAssert.isTrue(!normalized.equals(currentDatasetId), CommonErrorCode.PARAM_ERROR);
+        Dataset parent = datasetRepository.getById(normalized);
+        BusinessAssert.notNull(parent, DataManagementErrorCode.DATASET_NOT_FOUND);
+        BusinessAssert.isTrue(parent.getParentDatasetId() == null, CommonErrorCode.PARAM_ERROR);
+        return parent;
+    }
+
+    private void handleParentChange(Dataset dataset, String parentDatasetId) {
+        String normalized = normalizeParentId(parentDatasetId);
+        if (Objects.equals(dataset.getParentDatasetId(), normalized)) {
+            return;
+        }
+        long childCount = datasetRepository.countByParentId(dataset.getId());
+        if (childCount > 0 && normalized != null) {
+            throw BusinessException.of(DataManagementErrorCode.DATASET_HAS_CHILDREN);
+        }
+        Dataset parent = normalized == null ? null : resolveParentDataset(normalized, dataset.getId());
+        String newPath = buildDatasetPath(parent == null ? datasetBasePath : parent.getPath(), dataset.getId());
+        moveDatasetPath(dataset, newPath);
+        dataset.setParentDatasetId(parent == null ? null : parent.getId());
+    }
+
+    private String normalizeParentId(String parentDatasetId) {
+        if (!StringUtils.hasText(parentDatasetId)) {
+            return null;
+        }
+        return parentDatasetId.trim();
+    }
+
+    private String buildDatasetPath(String basePath, String datasetId) {
+        String normalized = basePath;
+        while (normalized.endsWith(File.separator)) {
+            normalized = normalized.substring(0, normalized.length() - 1);
+        }
+        return normalized + File.separator + datasetId;
+    }
+
+    private void moveDatasetPath(Dataset dataset, String newPath) {
+        String oldPath = dataset.getPath();
+        if (Objects.equals(oldPath, newPath)) {
+            return;
+        }
+        Path sourcePath = Paths.get(oldPath);
+        Path targetPath = Paths.get(newPath);
+        try {
+            if (Files.exists(sourcePath)) {
+                if (Files.exists(targetPath)) {
+                    throw BusinessException.of(CommonErrorCode.PARAM_ERROR);
+                }
+                Files.createDirectories(targetPath.getParent());
+                Files.move(sourcePath, targetPath);
+            } else {
+                Files.createDirectories(targetPath);
+            }
+        } catch (IOException e) {
+            log.error("move dataset path error, from {} to {}", oldPath, newPath, e);
+            throw BusinessException.of(SystemErrorCode.FILE_SYSTEM_ERROR);
+        }
+        datasetFileRepository.updateFilePathPrefix(dataset.getId(), oldPath, newPath);
+        dataset.setPath(newPath);
+    }
+
     /**
      * 获取数据集统计信息
      */

backend/services/data-management-service/src/main/java/com/datamate/datamanagement/application/DatasetFileApplicationService.java

@@ -323,10 +323,11 @@ public class DatasetFileApplicationService {
      * @return 请求id
      */
     @Transactional
-    public String preUpload(UploadFilesPreRequest chunkUploadRequest, String datasetId) {
+    public String preUpload(UploadFilesPreRequest chunkUploadRequest, String datasetId) {
-        if (Objects.isNull(datasetRepository.getById(datasetId))) {
+        Dataset dataset = datasetRepository.getById(datasetId);
-            throw BusinessException.of(DataManagementErrorCode.DATASET_NOT_FOUND);
+        if (Objects.isNull(dataset)) {
-        }
+            throw BusinessException.of(DataManagementErrorCode.DATASET_NOT_FOUND);
+        }
         
         // 构建上传路径，如果有 prefix 则追加到路径中
         String prefix = Optional.ofNullable(chunkUploadRequest.getPrefix()).orElse("").trim();
@@ -335,10 +336,13 @@ public class DatasetFileApplicationService {
             prefix = prefix.substring(1);
         }
         
-        String uploadPath = datasetBasePath + File.separator + datasetId;
+        String uploadPath = dataset.getPath();
-        if (!prefix.isEmpty()) {
+        if (uploadPath == null || uploadPath.isBlank()) {
-            uploadPath = uploadPath + File.separator + prefix.replace("/", File.separator);
+            uploadPath = datasetBasePath + File.separator + datasetId;
-        }
+        }
+        if (!prefix.isEmpty()) {
+            uploadPath = uploadPath + File.separator + prefix.replace("/", File.separator);
+        }
         
         ChunkUploadPreRequest request = ChunkUploadPreRequest.builder().build();
         request.setUploadPath(uploadPath);

backend/services/data-management-service/src/main/java/com/datamate/datamanagement/domain/model/dataset/Dataset.java

@@ -28,6 +28,10 @@ public class Dataset extends BaseEntity<String> {
      * 数据集描述
      */
     private String description;
+    /**
+     * 父数据集ID
+     */
+    private String parentDatasetId;
     /**
      * 数据集类型
      */
@@ -110,14 +114,23 @@ public class Dataset extends BaseEntity<String> {
         this.updatedAt = LocalDateTime.now();
     }
 
-    public void initCreateParam(String datasetBasePath) {
+    public void initCreateParam(String datasetBasePath, String parentPath) {
         this.id = UUID.randomUUID().toString();
-        this.path = datasetBasePath + File.separator + this.id;
+        String basePath = normalizeBasePath(parentPath != null && !parentPath.isBlank() ? parentPath : datasetBasePath);
+        this.path = basePath + File.separator + this.id;
         if (this.status == null) {
             this.status = DatasetStatusType.DRAFT;
         }
     }
 
+    private String normalizeBasePath(String basePath) {
+        String normalized = basePath;
+        while (normalized.endsWith(File.separator)) {
+            normalized = normalized.substring(0, normalized.length() - 1);
+        }
+        return normalized;
+    }
+
     public void updateBasicInfo(String name, String description, String category) {
         if (name != null && !name.isEmpty()) this.name = name;
         if (description != null) this.description = description;

backend/services/data-management-service/src/main/java/com/datamate/datamanagement/infrastructure/exception/DataManagementErrorCode.java

@@ -37,10 +37,14 @@ public enum DataManagementErrorCode implements ErrorCode {
      * 数据集文件已存在
      */
     DATASET_FILE_ALREADY_EXISTS("data_management.0006", "数据集文件已存在"),
-    /**
+    /**
-     * 目录不存在
+     * 目录不存在
-     */
+     */
-    DIRECTORY_NOT_FOUND("data_management.0007", "目录不存在");
+    DIRECTORY_NOT_FOUND("data_management.0007", "目录不存在"),
+    /**
+     * 存在子数据集
+     */
+    DATASET_HAS_CHILDREN("data_management.0008", "存在子数据集，禁止删除或移动");
 
     private final String code;
     private final String message;

backend/services/data-management-service/src/main/java/com/datamate/datamanagement/infrastructure/persistence/mapper/DatasetFileMapper.java

@@ -26,4 +26,7 @@ public interface DatasetFileMapper extends BaseMapper<DatasetFile> {
 
     int update(DatasetFile file);
     int deleteById(@Param("id") String id);
+    int updateFilePathPrefix(@Param("datasetId") String datasetId,
+                             @Param("oldPrefix") String oldPrefix,
+                             @Param("newPrefix") String newPrefix);
 }

backend/services/data-management-service/src/main/java/com/datamate/datamanagement/infrastructure/persistence/repository/DatasetFileRepository.java

@@ -25,4 +25,6 @@ public interface DatasetFileRepository extends IRepository<DatasetFile> {
 
     IPage<DatasetFile> findByCriteria(String datasetId, String fileType, String status, String name,
                                       IPage<DatasetFile> page);
+
+    int updateFilePathPrefix(String datasetId, String oldPrefix, String newPrefix);
 }

backend/services/data-management-service/src/main/java/com/datamate/datamanagement/infrastructure/persistence/repository/DatasetRepository.java

@@ -26,4 +26,6 @@ public interface DatasetRepository extends IRepository<Dataset> {
     AllDatasetStatisticsResponse getAllDatasetStatistics();
 
     IPage<Dataset> findByCriteria(IPage<Dataset> page, DatasetPagingQuery query);
+
+    long countByParentId(String parentDatasetId);
 }

backend/services/data-management-service/src/main/java/com/datamate/datamanagement/infrastructure/persistence/repository/impl/DatasetFileRepositoryImpl.java

@@ -56,4 +56,9 @@ public class DatasetFileRepositoryImpl extends CrudRepository<DatasetFileMapper,
                 .eq(StringUtils.hasText(status), DatasetFile::getStatus, status)
                 .like(StringUtils.hasText(name), DatasetFile::getFileName, name));
     }
+
+    @Override
+    public int updateFilePathPrefix(String datasetId, String oldPrefix, String newPrefix) {
+        return datasetFileMapper.updateFilePathPrefix(datasetId, oldPrefix, newPrefix);
+    }
 }

backend/services/data-management-service/src/main/java/com/datamate/datamanagement/infrastructure/persistence/repository/impl/DatasetRepositoryImpl.java

@@ -54,6 +54,14 @@ public class DatasetRepositoryImpl extends CrudRepository<DatasetMapper, Dataset
                 .eq(query.getType() != null, Dataset::getDatasetType, query.getType())
                 .eq(query.getStatus() != null, Dataset::getStatus, query.getStatus());
 
+        if (query.getParentDatasetId() != null) {
+            if (StringUtils.isBlank(query.getParentDatasetId())) {
+                wrapper.isNull(Dataset::getParentDatasetId);
+            } else {
+                wrapper.eq(Dataset::getParentDatasetId, query.getParentDatasetId());
+            }
+        }
+
         if (StringUtils.isNotBlank(query.getKeyword())) {
             wrapper.and(w ->
                     w.like(Dataset::getName, query.getKeyword()).or()
@@ -74,4 +82,10 @@ public class DatasetRepositoryImpl extends CrudRepository<DatasetMapper, Dataset
         wrapper.orderByDesc(Dataset::getCreatedAt);
         return datasetMapper.selectPage(page, wrapper);
     }
+
+    @Override
+    public long countByParentId(String parentDatasetId) {
+        return datasetMapper.selectCount(new LambdaQueryWrapper<Dataset>()
+                .eq(Dataset::getParentDatasetId, parentDatasetId));
+    }
 }

backend/services/data-management-service/src/main/java/com/datamate/datamanagement/interfaces/dto/CreateDatasetRequest.java

@@ -29,6 +29,8 @@ public class CreateDatasetRequest {
     /** 数据集类型 */
     @NotNull(message = "数据集类型不能为空")
     private DatasetType datasetType;
+    /** 父数据集ID */
+    private String parentDatasetId;
     /** 标签列表 */
     private List<String> tags;
     /** 数据源 */

backend/services/data-management-service/src/main/java/com/datamate/datamanagement/interfaces/dto/DatasetPagingQuery.java

@@ -39,4 +39,9 @@ public class DatasetPagingQuery extends PagingQuery {
      * 状态过滤
      */
     private DatasetStatusType status;
+
+    /**
+     * 父数据集ID过滤
+     */
+    private String parentDatasetId;
 }

backend/services/data-management-service/src/main/java/com/datamate/datamanagement/interfaces/dto/DatasetResponse.java

@@ -15,6 +15,8 @@ import java.util.Map;
 public class DatasetResponse {
     /** 数据集ID */
     private String id;
+    /** 父数据集ID */
+    private String parentDatasetId;
     /** 数据集名称 */
     private String name;
     /** 数据集描述 */

backend/services/data-management-service/src/main/java/com/datamate/datamanagement/interfaces/dto/UpdateDatasetRequest.java

@@ -23,6 +23,8 @@ public class UpdateDatasetRequest {
     private String description;
     /** 归集任务id */
     private String dataSource;
+    /** 父数据集ID */
+    private String parentDatasetId;
     /** 标签列表 */
     private List<String> tags;
     /** 数据集状态 */

backend/services/data-management-service/src/main/resources/mappers/DatasetFileMapper.xml

@@ -95,4 +95,11 @@
     <delete id="deleteById" parameterType="string">
         DELETE FROM t_dm_dataset_files WHERE id = #{id}
     </delete>
+
+    <update id="updateFilePathPrefix">
+        UPDATE t_dm_dataset_files
+        SET file_path = CONCAT(#{newPrefix}, SUBSTRING(file_path, LENGTH(#{oldPrefix}) + 1))
+        WHERE dataset_id = #{datasetId}
+          AND file_path LIKE CONCAT(#{oldPrefix}, '%')
+    </update>
 </mapper>

backend/services/data-management-service/src/main/resources/mappers/DatasetMapper.xml

@@ -4,12 +4,13 @@
 <mapper namespace="com.datamate.datamanagement.infrastructure.persistence.mapper.DatasetMapper">
 
   <sql id="Base_Column_List">
-    id, name, description, dataset_type, category, path, format, schema_info, size_bytes, file_count, record_count,
+    id, parent_dataset_id, name, description, dataset_type, category, path, format, schema_info, size_bytes, file_count, record_count,
         retention_days, tags, metadata, status, is_public, is_featured, version, created_at, updated_at, created_by, updated_by
   </sql>
 
   <sql id="Alias_D_Column_List">
         d.id AS id,
+        d.parent_dataset_id AS parent_dataset_id,
         d.name AS name,
         d.description AS description,
         d.dataset_type AS dataset_type,

frontend/src/pages/DataManagement/Create/CreateDataset.tsx

@@ -17,6 +17,7 @@ export default function DatasetCreate() {
     description: "",
     datasetType: DatasetType.TEXT,
     tags: [],
+    parentDatasetId: "",
   });
 
   const handleSubmit = async () => {

frontend/src/pages/DataManagement/Create/EditDataset.tsx

@@ -26,6 +26,7 @@ export default function EditDataset({
     description: "",
     datasetType: DatasetType.TEXT,
     tags: [],
+    parentDatasetId: "",
   });
   const fetchDataset = async () => {
     if (!open) return;
@@ -36,6 +37,7 @@ export default function EditDataset({
         ...newData,
         type: newData.type,
         tags: newData.tags.map((tag) => tag.name) || [],
+        parentDatasetId: newData.parentDatasetId || "",
       };
       setNewDataset(updatedDataset);
       form.setFieldsValue(updatedDataset);

frontend/src/pages/DataManagement/Create/components/BasicInformation.tsx

@@ -2,7 +2,7 @@ import RadioCard from "@/components/RadioCard";
 import { Input, Select, Form } from "antd";
 import { datasetTypes } from "../../dataset.const";
 import { useEffect, useState } from "react";
-import { queryDatasetTagsUsingGet } from "../../dataset.api";
+import { queryDatasetTagsUsingGet, queryDatasetsUsingGet } from "../../dataset.api";
 import {queryTasksUsingGet} from "@/pages/DataCollection/collection.apis.ts";
 
 export default function BasicInformation({
@@ -22,6 +22,9 @@ export default function BasicInformation({
     }[]
   >([]);
   const [collectionOptions, setCollectionOptions] = useState([]);
+  const [parentDatasetOptions, setParentDatasetOptions] = useState<
+    { label: string; value: string }[]
+  >([]);
 
   // 获取标签
   const fetchTags = async () => {
@@ -52,10 +55,36 @@ export default function BasicInformation({
     }
   };
 
+  const fetchParentDatasets = async () => {
+    if (hidden.includes("parentDatasetId")) return;
+    try {
+      const { data: resData } = await queryDatasetsUsingGet({
+        parentDatasetId: "",
+        page: 1,
+        size: 1000,
+      });
+      const currentDatasetId = data?.id;
+      const rootDatasets = resData?.content || [];
+      const options = rootDatasets
+        .filter((dataset) => dataset.id !== currentDatasetId)
+        .map((dataset) => ({
+          label: dataset.name,
+          value: dataset.id,
+        }));
+      setParentDatasetOptions([
+        { label: "根数据集", value: "" },
+        ...options,
+      ]);
+    } catch (error) {
+      console.error("Error fetching parent datasets:", error);
+    }
+  };
+
   useEffect(() => {
     fetchTags();
     fetchCollectionTasks();
-  }, []);
+    fetchParentDatasets();
+  }, [data?.id, hidden.join(",")]);
   return (
     <>
       <Form.Item
@@ -70,6 +99,15 @@ export default function BasicInformation({
           <Input.TextArea placeholder="描述数据集的用途和内容" rows={3} />
         </Form.Item>
       )}
+      {!hidden.includes("parentDatasetId") && (
+        <Form.Item name="parentDatasetId" label="父数据集">
+          <Select
+            className="w-full"
+            options={parentDatasetOptions}
+            placeholder="选择父数据集（仅支持一层）"
+          />
+        </Form.Item>
+      )}
 
       {/* 数据集类型选择 - 使用卡片形式 */}
       {!hidden.includes("datasetType") && (

frontend/src/pages/DataManagement/Detail/DatasetDetail.tsx

@@ -37,29 +37,51 @@ const tabList = [
 export default function DatasetDetail() {
   const { id } = useParams(); // 获取动态路由参数
   const navigate = useNavigate();
-  const [activeTab, setActiveTab] = useState("overview");
+  const [activeTab, setActiveTab] = useState("overview");
-  const { message } = App.useApp();
+  const { message } = App.useApp();
-  const [showEditDialog, setShowEditDialog] = useState(false);
+  const [showEditDialog, setShowEditDialog] = useState(false);
-
+
-  const [dataset, setDataset] = useState<Dataset>({} as Dataset);
+  const [dataset, setDataset] = useState<Dataset>({} as Dataset);
-  const filesOperation = useFilesOperation(dataset);
+  const [parentDataset, setParentDataset] = useState<Dataset | null>(null);
+  const filesOperation = useFilesOperation(dataset);
 
   const [showUploadDialog, setShowUploadDialog] = useState(false);
-  const navigateItems = useMemo(
+  const navigateItems = useMemo(() => {
-    () => [
+    const items = [
-      {
+      {
-        title: <Link to="/data/management">数据管理</Link>,
+        title: <Link to="/data/management">数据管理</Link>,
-      },
+      },
-      {
+    ];
-        title: dataset.name || "数据集详情",
+    if (parentDataset) {
-      },
+      items.push({
-    ],
+        title: (
-    [dataset]
+          <Link to={`/data/management/detail/${parentDataset.id}`}>
-  );
+            {parentDataset.name}
-  const fetchDataset = async () => {
+          </Link>
-    const { data } = await queryDatasetByIdUsingGet(id as unknown as number);
+        ),
-    setDataset(mapDataset(data));
+      });
-  };
+    }
+    items.push({
+      title: dataset.name || "数据集详情",
+    });
+    return items;
+  }, [dataset, parentDataset]);
+  const fetchDataset = async () => {
+    if (!id) {
+      return;
+    }
+    const { data } = await queryDatasetByIdUsingGet(id);
+    const mapped = mapDataset(data);
+    setDataset(mapped);
+    if (data?.parentDatasetId) {
+      const { data: parentData } = await queryDatasetByIdUsingGet(
+        data.parentDatasetId
+      );
+      setParentDataset(mapDataset(parentData));
+    } else {
+      setParentDataset(null);
+    }
+  };
 
   useEffect(() => {
     fetchDataset();

frontend/src/pages/DataManagement/Detail/useFilesOperation.ts

@@ -21,7 +21,7 @@ export function useFilesOperation(dataset: Dataset) {
 
   // 文件相关状态
   const [fileList, setFileList] = useState<DatasetFile[]>([]);
-  const [selectedFiles, setSelectedFiles] = useState<number[]>([]);
+  const [selectedFiles, setSelectedFiles] = useState<string[]>([]);
   const [pagination, setPagination] = useState<{
     current: number;
     pageSize: number;

frontend/src/pages/DataManagement/Home/DataManagement.tsx

@@ -135,9 +135,9 @@ export default function DatasetManagementPage() {
     message.success("数据集下载成功");
   };
 
-  const handleDeleteDataset = async (id: number) => {
+  const handleDeleteDataset = async (id: string) => {
-    if (!id) return;
+    if (!id) return;
-    await deleteDatasetByIdUsingDelete(id);
+    await deleteDatasetByIdUsingDelete(id);
     fetchData({ pageOffset: 0 });
     message.success("数据删除成功");
   };

frontend/src/pages/DataManagement/dataset.model.ts

@@ -33,7 +33,7 @@ export enum DataSource {
 }
 
 export interface DatasetFile {
-  id: number;
+  id: string;
   fileName: string;
   size: string;
   uploadDate: string;
@@ -41,10 +41,10 @@ export interface DatasetFile {
 }
 
 export interface Dataset {
-  id: number;
+  id: string;
   name: string;
   description: string;
-  parentId?: number;
+  parentDatasetId?: string;
   datasetType: DatasetType;
   status: DatasetStatus;
   size?: string;
@@ -77,7 +77,7 @@ export interface ScheduleConfig {
 }
 
 export interface DatasetTask {
-  id: number;
+  id: string;
   name: string;
   description: string;
   type: string;

runtime/datamate-python/app/db/models/dataset_management.py

@@ -14,6 +14,7 @@ class Dataset(Base):
     __tablename__ = "t_dm_datasets"
     
     id = Column(String(36), primary_key=True, default=lambda: str(uuid.uuid4()), comment="UUID")
+    parent_dataset_id = Column(String(36), nullable=True, comment="父数据集ID（UUID）")
     name = Column(String(255), nullable=False, comment="数据集名称")
     description = Column(Text, nullable=True, comment="数据集描述")
     dataset_type = Column(String(50), nullable=False, comment="数据集类型：IMAGE/TEXT/QA/MULTIMODAL/OTHER")

runtime/python-executor/datamate/auto_annotation_worker.py

@@ -195,8 +195,8 @@ def _update_task_status(
         conn.execute(sql, params)
 
 
-def _load_dataset_files(dataset_id: str) -> List[Tuple[str, str, str]]:
+def _load_dataset_files(dataset_id: str) -> List[Tuple[str, str, str]]:
-    """加载指定数据集下的所有已完成文件。"""
+    """加载指定数据集下的所有已完成文件。"""
 
     sql = text(
         """
@@ -208,9 +208,45 @@ def _load_dataset_files(dataset_id: str) -> List[Tuple[str, str, str]]:
         """
     )
 
-    with SQLManager.create_connect() as conn:
+    with SQLManager.create_connect() as conn:
-        rows = conn.execute(sql, {"dataset_id": dataset_id}).fetchall()
+        rows = conn.execute(sql, {"dataset_id": dataset_id}).fetchall()
-        return [(str(r[0]), str(r[1]), str(r[2])) for r in rows]
+        return [(str(r[0]), str(r[1]), str(r[2])) for r in rows]
+
+
+def _load_dataset_meta(dataset_id: str) -> Optional[Dict[str, Any]]:
+    """加载数据集基础信息（含父ID与路径）。"""
+
+    sql = text(
+        """
+        SELECT id, name, parent_dataset_id, path
+        FROM t_dm_datasets
+        WHERE id = :dataset_id
+        """
+    )
+    with SQLManager.create_connect() as conn:
+        row = conn.execute(sql, {"dataset_id": dataset_id}).fetchone()
+        if not row:
+            return None
+        return dict(row._mapping)  # type: ignore[attr-defined]
+
+
+def _resolve_output_parent(source_dataset_id: str) -> Tuple[Optional[str], str]:
+    """根据源数据集确定产出数据集的父级与基路径（产出挂在父级下）。"""
+
+    base_path = DEFAULT_OUTPUT_ROOT.rstrip("/") or "/dataset"
+    source_meta = _load_dataset_meta(source_dataset_id)
+    if not source_meta:
+        return None, base_path
+
+    parent_dataset_id = source_meta.get("parent_dataset_id")
+    if not parent_dataset_id:
+        return None, base_path
+
+    parent_meta = _load_dataset_meta(str(parent_dataset_id))
+    parent_path = parent_meta.get("path") if parent_meta else None
+    if not parent_path:
+        return None, base_path
+    return str(parent_dataset_id), str(parent_path)
 
 
 def _load_files_by_ids(file_ids: List[str]) -> List[Tuple[str, str, str]]:
@@ -245,34 +281,35 @@ def _ensure_output_dir(output_dir: str) -> str:
     return output_dir
 
 
-def _create_output_dataset(
+def _create_output_dataset(
-    source_dataset_id: str,
+    source_dataset_id: str,
-    source_dataset_name: str,
+    source_dataset_name: str,
-    output_dataset_name: str,
+    output_dataset_name: str,
-) -> Tuple[str, str]:
+) -> Tuple[str, str]:
-    """为自动标注结果创建一个新的数据集并返回 (dataset_id, path)。"""
+    """为自动标注结果创建一个新的数据集并返回 (dataset_id, path)。"""
-
+
-    new_dataset_id = str(uuid.uuid4())
+    new_dataset_id = str(uuid.uuid4())
-    dataset_base_path = DEFAULT_OUTPUT_ROOT.rstrip("/") or "/dataset"
+    parent_dataset_id, dataset_base_path = _resolve_output_parent(source_dataset_id)
-    output_dir = os.path.join(dataset_base_path, new_dataset_id)
+    output_dir = os.path.join(dataset_base_path, new_dataset_id)
 
     description = (
         f"Auto annotations for dataset {source_dataset_name or source_dataset_id}"[:255]
     )
 
-    sql = text(
+    sql = text(
-        """
+        """
-        INSERT INTO t_dm_datasets (id, name, description, dataset_type, path, status)
+        INSERT INTO t_dm_datasets (id, parent_dataset_id, name, description, dataset_type, path, status)
-        VALUES (:id, :name, :description, :dataset_type, :path, :status)
+        VALUES (:id, :parent_dataset_id, :name, :description, :dataset_type, :path, :status)
-        """
+        """
-    )
+    )
-    params = {
+    params = {
-        "id": new_dataset_id,
+        "id": new_dataset_id,
-        "name": output_dataset_name,
+        "parent_dataset_id": parent_dataset_id,
-        "description": description,
+        "name": output_dataset_name,
-        "dataset_type": "IMAGE",
+        "description": description,
-        "path": output_dir,
+        "dataset_type": "IMAGE",
-        "status": "ACTIVE",
+        "path": output_dir,
+        "status": "ACTIVE",
     }
 
     with SQLManager.create_connect() as conn:

scripts/db/data-management-init.sql

@@ -15,6 +15,7 @@ USE datamate;
 -- 数据集表（支持医学影像、文本、问答等多种类型）
 CREATE TABLE IF NOT EXISTS t_dm_datasets (
     id VARCHAR(36) PRIMARY KEY COMMENT 'UUID',
+    parent_dataset_id VARCHAR(36) NULL COMMENT '父数据集ID（UUID）',
     name VARCHAR(255) NOT NULL COMMENT '数据集名称',
     description TEXT COMMENT '数据集描述',
     dataset_type VARCHAR(50) NOT NULL COMMENT '数据集类型：IMAGE/TEXT/QA/MULTIMODAL/OTHER',
@@ -36,13 +37,15 @@ CREATE TABLE IF NOT EXISTS t_dm_datasets (
     updated_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP COMMENT '更新时间',
     created_by VARCHAR(255) COMMENT '创建者',
     updated_by VARCHAR(255) COMMENT '更新者',
+    INDEX idx_dm_parent_dataset_id (parent_dataset_id),
     INDEX idx_dm_dataset_type (dataset_type),
     INDEX idx_dm_category (category),
     INDEX idx_dm_format (format),
     INDEX idx_dm_status (status),
     INDEX idx_dm_public (is_public),
     INDEX idx_dm_featured (is_featured),
-    INDEX idx_dm_created_at (created_at)
+    INDEX idx_dm_created_at (created_at),
+    FOREIGN KEY (parent_dataset_id) REFERENCES t_dm_datasets(id) ON DELETE RESTRICT
 ) COMMENT='数据集表（UUID 主键）';
 
 -- 数据集文件表