Develop labeling module (#25)

* refactor: remove db table management from LS adapter (mv to scripts later); change adapter to use the same MySQL DB as other modules.

* refactor: Rename LS Adapter module to datamate-python

runtime/datamate-python/app/models/README.md

@@ -0,0 +1,138 @@
+# DataMate 数据模型结构
+
+本文档列出了根据 `scripts/db` 中的 SQL 文件创建的所有 Python 数据模型。
+
+## 模型组织结构
+
+```
+app/models/
+├── __init__.py                 # 主模块导出文件
+├── dm/                         # 数据管理 (Data Management) 模块
+│   ├── __init__.py
+│   ├── annotation_template.py  # 标注模板
+│   ├── labeling_project.py     # 标注项目
+│   ├── dataset.py              # 数据集
+│   ├── dataset_files.py        # 数据集文件
+│   ├── dataset_statistics.py   # 数据集统计
+│   ├── dataset_tag.py          # 数据集标签关联
+│   ├── tag.py                  # 标签
+│   └── user.py                 # 用户
+├── cleaning/                   # 数据清洗 (Data Cleaning) 模块
+│   ├── __init__.py
+│   ├── clean_template.py       # 清洗模板
+│   ├── clean_task.py           # 清洗任务
+│   ├── operator_instance.py    # 算子实例
+│   └── clean_result.py         # 清洗结果
+├── collection/                 # 数据归集 (Data Collection) 模块
+│   ├── __init__.py
+│   ├── task_execution.py       # 任务执行明细
+│   ├── collection_task.py      # 数据归集任务
+│   ├── task_log.py             # 任务执行记录
+│   └── datax_template.py       # DataX模板配置
+├── common/                     # 通用 (Common) 模块
+│   ├── __init__.py
+│   └── chunk_upload_request.py # 文件切片上传请求
+└── operator/                   # 算子 (Operator) 模块
+    ├── __init__.py
+    ├── operator.py             # 算子
+    ├── operator_category.py    # 算子分类
+    └── operator_category_relation.py  # 算子分类关联
+```
+
+## 模块详情
+
+### 1. Data Management (DM) 模块
+对应 SQL: `data-management-init.sql` 和 `data-annotation-init.sql`
+
+#### 模型列表：
+- **AnnotationTemplate** (`t_dm_annotation_templates`) - 标注模板
+- **LabelingProject** (`t_dm_labeling_projects`) - 标注项目
+- **Dataset** (`t_dm_datasets`) - 数据集（支持医学影像、文本、问答等多种类型）
+- **DatasetFiles** (`t_dm_dataset_files`) - 数据集文件
+- **DatasetStatistics** (`t_dm_dataset_statistics`) - 数据集统计信息
+- **Tag** (`t_dm_tags`) - 标签
+- **DatasetTag** (`t_dm_dataset_tags`) - 数据集标签关联
+- **User** (`users`) - 用户
+
+### 2. Data Cleaning 模块
+对应 SQL: `data-cleaning-init.sql`
+
+#### 模型列表：
+- **CleanTemplate** (`t_clean_template`) - 清洗模板
+- **CleanTask** (`t_clean_task`) - 清洗任务
+- **OperatorInstance** (`t_operator_instance`) - 算子实例
+- **CleanResult** (`t_clean_result`) - 清洗结果
+
+### 3. Data Collection (DC) 模块
+对应 SQL: `data-collection-init.sql`
+
+#### 模型列表：
+- **TaskExecution** (`t_dc_task_executions`) - 任务执行明细
+- **CollectionTask** (`t_dc_collection_tasks`) - 数据归集任务
+- **TaskLog** (`t_dc_task_log`) - 任务执行记录
+- **DataxTemplate** (`t_dc_datax_templates`) - DataX模板配置
+
+### 4. Common 模块
+对应 SQL: `data-common-init.sql`
+
+#### 模型列表：
+- **ChunkUploadRequest** (`t_chunk_upload_request`) - 文件切片上传请求
+
+### 5. Operator 模块
+对应 SQL: `data-operator-init.sql`
+
+#### 模型列表：
+- **Operator** (`t_operator`) - 算子
+- **OperatorCategory** (`t_operator_category`) - 算子分类
+- **OperatorCategoryRelation** (`t_operator_category_relation`) - 算子分类关联
+
+## 使用方式
+
+```python
+# 导入所有模型
+from app.models import (
+    # DM 模块
+    AnnotationTemplate,
+    LabelingProject,
+    Dataset,
+    DatasetFiles,
+    DatasetStatistics,
+    DatasetTag,
+    Tag,
+    User,
+    # Cleaning 模块
+    CleanTemplate,
+    CleanTask,
+    OperatorInstance,
+    CleanResult,
+    # Collection 模块
+    TaskExecution,
+    CollectionTask,
+    TaskLog,
+    DataxTemplate,
+    # Common 模块
+    ChunkUploadRequest,
+    # Operator 模块
+    Operator,
+    OperatorCategory,
+    OperatorCategoryRelation
+)
+
+# 或者按模块导入
+from app.models.dm import Dataset, DatasetFiles
+from app.models.collection import CollectionTask
+from app.models.operator import Operator
+```
+
+## 注意事项
+
+1. **UUID 主键**: 大部分表使用 UUID (String(36)) 作为主键
+2. **时间戳**: 使用 `TIMESTAMP` 类型，并配置自动更新
+3. **软删除**: 部分模型（如 AnnotationTemplate, LabelingProject）支持软删除，包含 `deleted_at` 字段和 `is_deleted` 属性
+4. **JSON 字段**: 配置信息、元数据等使用 JSON 类型存储
+5. **字段一致性**: 所有模型字段都严格按照 SQL 定义创建，确保与数据库表结构完全一致
+
+## 更新记录
+
+- 2025-10-25: 根据 `scripts/db` 中的 SQL 文件创建所有数据模型
+- 已更新现有的 `annotation_template.py`、`labeling_project.py`、`dataset_files.py` 以匹配 SQL 定义

runtime/datamate-python/app/models/__init__.py

@@ -0,0 +1,69 @@
+# app/models/__init__.py
+
+# Data Management (DM) 模块
+from .dm import (
+    AnnotationTemplate,
+    LabelingProject,
+    Dataset,
+    DatasetFiles,
+    DatasetStatistics,
+    DatasetTag,
+    Tag,
+    User
+)
+
+# Data Cleaning 模块
+from .cleaning import (
+    CleanTemplate,
+    CleanTask,
+    OperatorInstance,
+    CleanResult
+)
+
+# Data Collection (DC) 模块
+from .collection import (
+    TaskExecution,
+    CollectionTask,
+    TaskLog,
+    DataxTemplate
+)
+
+# Common 模块
+from .common import (
+    ChunkUploadRequest
+)
+
+# Operator 模块
+from .operator import (
+    Operator,
+    OperatorCategory,
+    OperatorCategoryRelation
+)
+
+__all__ = [
+    # DM 模块
+    "AnnotationTemplate",
+    "LabelingProject",
+    "Dataset",
+    "DatasetFiles",
+    "DatasetStatistics",
+    "DatasetTag",
+    "Tag",
+    "User",
+    # Cleaning 模块
+    "CleanTemplate",
+    "CleanTask",
+    "OperatorInstance",
+    "CleanResult",
+    # Collection 模块
+    "TaskExecution",
+    "CollectionTask",
+    "TaskLog",
+    "DataxTemplate",
+    # Common 模块
+    "ChunkUploadRequest",
+    # Operator 模块
+    "Operator",
+    "OperatorCategory",
+    "OperatorCategoryRelation"
+]

runtime/datamate-python/app/models/cleaning/__init__.py

@@ -0,0 +1,13 @@
+# app/models/cleaning/__init__.py
+
+from .clean_template import CleanTemplate
+from .clean_task import CleanTask
+from .operator_instance import OperatorInstance
+from .clean_result import CleanResult
+
+__all__ = [
+    "CleanTemplate",
+    "CleanTask",
+    "OperatorInstance",
+    "CleanResult"
+]

runtime/datamate-python/app/models/cleaning/clean_result.py

@@ -0,0 +1,22 @@
+from sqlalchemy import Column, String, BigInteger, Text
+from app.db.database import Base
+
+class CleanResult(Base):
+    """清洗结果模型"""
+    
+    __tablename__ = "t_clean_result"
+    
+    instance_id = Column(String(64), primary_key=True, comment="实例ID")
+    src_file_id = Column(String(64), nullable=True, comment="源文件ID")
+    dest_file_id = Column(String(64), primary_key=True, comment="目标文件ID")
+    src_name = Column(String(256), nullable=True, comment="源文件名")
+    dest_name = Column(String(256), nullable=True, comment="目标文件名")
+    src_type = Column(String(256), nullable=True, comment="源文件类型")
+    dest_type = Column(String(256), nullable=True, comment="目标文件类型")
+    src_size = Column(BigInteger, nullable=True, comment="源文件大小")
+    dest_size = Column(BigInteger, nullable=True, comment="目标文件大小")
+    status = Column(String(256), nullable=True, comment="处理状态")
+    result = Column(Text, nullable=True, comment="处理结果")
+    
+    def __repr__(self):
+        return f"<CleanResult(instance_id={self.instance_id}, dest_file_id={self.dest_file_id}, status={self.status})>"

runtime/datamate-python/app/models/cleaning/clean_task.py

@@ -0,0 +1,27 @@
+from sqlalchemy import Column, String, BigInteger, Integer, TIMESTAMP
+from sqlalchemy.sql import func
+from app.db.database import Base
+
+class CleanTask(Base):
+    """清洗任务模型"""
+    
+    __tablename__ = "t_clean_task"
+    
+    id = Column(String(64), primary_key=True, comment="任务ID")
+    name = Column(String(64), nullable=True, comment="任务名称")
+    description = Column(String(256), nullable=True, comment="任务描述")
+    status = Column(String(256), nullable=True, comment="任务状态")
+    src_dataset_id = Column(String(64), nullable=True, comment="源数据集ID")
+    src_dataset_name = Column(String(64), nullable=True, comment="源数据集名称")
+    dest_dataset_id = Column(String(64), nullable=True, comment="目标数据集ID")
+    dest_dataset_name = Column(String(64), nullable=True, comment="目标数据集名称")
+    before_size = Column(BigInteger, nullable=True, comment="清洗前大小")
+    after_size = Column(BigInteger, nullable=True, comment="清洗后大小")
+    file_count = Column(Integer, nullable=True, comment="文件数量")
+    created_at = Column(TIMESTAMP, server_default=func.current_timestamp(), comment="创建时间")
+    started_at = Column(TIMESTAMP, nullable=True, comment="开始时间")
+    finished_at = Column(TIMESTAMP, nullable=True, comment="完成时间")
+    created_by = Column(String(256), nullable=True, comment="创建者")
+    
+    def __repr__(self):
+        return f"<CleanTask(id={self.id}, name={self.name}, status={self.status})>"

runtime/datamate-python/app/models/cleaning/clean_template.py

@@ -0,0 +1,18 @@
+from sqlalchemy import Column, String, Text, TIMESTAMP
+from sqlalchemy.sql import func
+from app.db.database import Base
+
+class CleanTemplate(Base):
+    """清洗模板模型"""
+    
+    __tablename__ = "t_clean_template"
+    
+    id = Column(String(64), primary_key=True, unique=True, comment="模板ID")
+    name = Column(String(64), nullable=True, comment="模板名称")
+    description = Column(String(256), nullable=True, comment="模板描述")
+    created_at = Column(TIMESTAMP, server_default=func.current_timestamp(), comment="创建时间")
+    updated_at = Column(TIMESTAMP, server_default=func.current_timestamp(), onupdate=func.current_timestamp(), comment="更新时间")
+    created_by = Column(String(256), nullable=True, comment="创建者")
+    
+    def __repr__(self):
+        return f"<CleanTemplate(id={self.id}, name={self.name})>"

runtime/datamate-python/app/models/cleaning/operator_instance.py

@@ -0,0 +1,15 @@
+from sqlalchemy import Column, String, Integer, Text
+from app.db.database import Base
+
+class OperatorInstance(Base):
+    """算子实例模型"""
+    
+    __tablename__ = "t_operator_instance"
+    
+    instance_id = Column(String(256), primary_key=True, comment="实例ID")
+    operator_id = Column(String(256), primary_key=True, comment="算子ID")
+    op_index = Column(Integer, primary_key=True, comment="算子索引")
+    settings_override = Column(Text, nullable=True, comment="配置覆盖")
+    
+    def __repr__(self):
+        return f"<OperatorInstance(instance_id={self.instance_id}, operator_id={self.operator_id}, index={self.op_index})>"

runtime/datamate-python/app/models/collection/__init__.py

@@ -0,0 +1,13 @@
+# app/models/collection/__init__.py
+
+from .task_execution import TaskExecution
+from .collection_task import CollectionTask
+from .task_log import TaskLog
+from .datax_template import DataxTemplate
+
+__all__ = [
+    "TaskExecution",
+    "CollectionTask",
+    "TaskLog",
+    "DataxTemplate"
+]

runtime/datamate-python/app/models/collection/collection_task.py

@@ -0,0 +1,28 @@
+from sqlalchemy import Column, String, Text, Integer, BigInteger, TIMESTAMP
+from sqlalchemy.sql import func
+from app.db.database import Base
+
+class CollectionTask(Base):
+    """数据归集任务模型"""
+    
+    __tablename__ = "t_dc_collection_tasks"
+    
+    id = Column(String(36), primary_key=True, comment="任务ID（UUID）")
+    name = Column(String(255), nullable=False, comment="任务名称")
+    description = Column(Text, nullable=True, comment="任务描述")
+    sync_mode = Column(String(20), default='ONCE', comment="同步模式：ONCE/SCHEDULED")
+    config = Column(Text, nullable=False, comment="归集配置（DataX配置），包含源端和目标端配置信息")
+    schedule_expression = Column(String(255), nullable=True, comment="Cron调度表达式")
+    status = Column(String(20), default='DRAFT', comment="任务状态：DRAFT/READY/RUNNING/SUCCESS/FAILED/STOPPED")
+    retry_count = Column(Integer, default=3, comment="重试次数")
+    timeout_seconds = Column(Integer, default=3600, comment="超时时间（秒）")
+    max_records = Column(BigInteger, nullable=True, comment="最大处理记录数")
+    sort_field = Column(String(100), nullable=True, comment="增量字段")
+    last_execution_id = Column(String(36), nullable=True, comment="最后执行ID（UUID）")
+    created_at = Column(TIMESTAMP, server_default=func.current_timestamp(), comment="创建时间")
+    updated_at = Column(TIMESTAMP, server_default=func.current_timestamp(), onupdate=func.current_timestamp(), comment="更新时间")
+    created_by = Column(String(255), nullable=True, comment="创建者")
+    updated_by = Column(String(255), nullable=True, comment="更新者")
+    
+    def __repr__(self):
+        return f"<CollectionTask(id={self.id}, name={self.name}, status={self.status})>"

runtime/datamate-python/app/models/collection/datax_template.py

@@ -0,0 +1,23 @@
+from sqlalchemy import Column, String, Text, Boolean, TIMESTAMP
+from sqlalchemy.sql import func
+from app.db.database import Base
+
+class DataxTemplate(Base):
+    """DataX模板配置模型"""
+    
+    __tablename__ = "t_dc_datax_templates"
+    
+    id = Column(String(36), primary_key=True, comment="模板ID（UUID）")
+    name = Column(String(255), nullable=False, unique=True, comment="模板名称")
+    source_type = Column(String(50), nullable=False, comment="源数据源类型")
+    target_type = Column(String(50), nullable=False, comment="目标数据源类型")
+    template_content = Column(Text, nullable=False, comment="模板内容")
+    description = Column(Text, nullable=True, comment="模板描述")
+    version = Column(String(20), default='1.0.0', comment="版本号")
+    is_system = Column(Boolean, default=False, comment="是否系统模板")
+    created_at = Column(TIMESTAMP, server_default=func.current_timestamp(), comment="创建时间")
+    updated_at = Column(TIMESTAMP, server_default=func.current_timestamp(), onupdate=func.current_timestamp(), comment="更新时间")
+    created_by = Column(String(255), nullable=True, comment="创建者")
+    
+    def __repr__(self):
+        return f"<DataxTemplate(id={self.id}, name={self.name}, source={self.source_type}, target={self.target_type})>"

runtime/datamate-python/app/models/collection/task_execution.py

@@ -0,0 +1,34 @@
+from sqlalchemy import Column, String, Text, Integer, BigInteger, DECIMAL, JSON, TIMESTAMP
+from sqlalchemy.sql import func
+from app.db.database import Base
+
+class TaskExecution(Base):
+    """任务执行明细模型"""
+    
+    __tablename__ = "t_dc_task_executions"
+    
+    id = Column(String(36), primary_key=True, comment="执行记录ID（UUID）")
+    task_id = Column(String(36), nullable=False, comment="任务ID")
+    task_name = Column(String(255), nullable=False, comment="任务名称")
+    status = Column(String(20), default='RUNNING', comment="执行状态：RUNNING/SUCCESS/FAILED/STOPPED")
+    progress = Column(DECIMAL(5, 2), default=0.00, comment="进度百分比")
+    records_total = Column(BigInteger, default=0, comment="总记录数")
+    records_processed = Column(BigInteger, default=0, comment="已处理记录数")
+    records_success = Column(BigInteger, default=0, comment="成功记录数")
+    records_failed = Column(BigInteger, default=0, comment="失败记录数")
+    throughput = Column(DECIMAL(10, 2), default=0.00, comment="吞吐量（条/秒）")
+    data_size_bytes = Column(BigInteger, default=0, comment="数据量（字节）")
+    started_at = Column(TIMESTAMP, nullable=True, comment="开始时间")
+    completed_at = Column(TIMESTAMP, nullable=True, comment="完成时间")
+    duration_seconds = Column(Integer, default=0, comment="执行时长（秒）")
+    config = Column(JSON, nullable=True, comment="执行配置")
+    error_message = Column(Text, nullable=True, comment="错误信息")
+    datax_job_id = Column(Text, nullable=True, comment="datax任务ID")
+    result = Column(Text, nullable=True, comment="执行结果")
+    created_at = Column(TIMESTAMP, server_default=func.current_timestamp(), comment="创建时间")
+    updated_at = Column(TIMESTAMP, server_default=func.current_timestamp(), onupdate=func.current_timestamp(), comment="更新时间")
+    created_by = Column(String(255), nullable=True, comment="创建者")
+    updated_by = Column(String(255), nullable=True, comment="更新者")
+    
+    def __repr__(self):
+        return f"<TaskExecution(id={self.id}, task_id={self.task_id}, status={self.status})>"

runtime/datamate-python/app/models/collection/task_log.py

@@ -0,0 +1,26 @@
+from sqlalchemy import Column, String, Text, Integer, BigInteger, TIMESTAMP
+from sqlalchemy.sql import func
+from app.db.database import Base
+
+class TaskLog(Base):
+    """任务执行记录模型"""
+    
+    __tablename__ = "t_dc_task_log"
+    
+    id = Column(String(36), primary_key=True, comment="执行记录ID（UUID）")
+    task_id = Column(String(36), nullable=False, comment="任务ID")
+    task_name = Column(String(255), nullable=False, comment="任务名称")
+    sync_mode = Column(String(20), default='FULL', comment="同步模式：FULL/INCREMENTAL")
+    status = Column(String(20), default='RUNNING', comment="执行状态：RUNNING/SUCCESS/FAILED/STOPPED")
+    start_time = Column(TIMESTAMP, nullable=True, comment="开始时间")
+    end_time = Column(TIMESTAMP, nullable=True, comment="结束时间")
+    duration = Column(BigInteger, nullable=True, comment="执行时长(毫秒)")
+    process_id = Column(String(50), nullable=True, comment="进程ID")
+    log_path = Column(String(500), nullable=True, comment="日志文件路径")
+    error_msg = Column(Text, nullable=True, comment="错误信息")
+    result = Column(Text, nullable=True, comment="执行结果")
+    retry_times = Column(Integer, default=0, comment="重试次数")
+    create_time = Column(TIMESTAMP, server_default=func.current_timestamp(), comment="创建时间")
+    
+    def __repr__(self):
+        return f"<TaskLog(id={self.id}, task_id={self.task_id}, status={self.status})>"

runtime/datamate-python/app/models/common/__init__.py

@@ -0,0 +1,7 @@
+# app/models/common/__init__.py
+
+from .chunk_upload_request import ChunkUploadRequest
+
+__all__ = [
+    "ChunkUploadRequest"
+]

runtime/datamate-python/app/models/common/chunk_upload_request.py

@@ -0,0 +1,19 @@
+from sqlalchemy import Column, String, Integer, Text, TIMESTAMP
+from sqlalchemy.sql import func
+from app.db.database import Base
+
+class ChunkUploadRequest(Base):
+    """文件切片上传请求模型"""
+    
+    __tablename__ = "t_chunk_upload_request"
+    
+    id = Column(String(36), primary_key=True, comment="UUID")
+    total_file_num = Column(Integer, nullable=True, comment="总文件数")
+    uploaded_file_num = Column(Integer, nullable=True, comment="已上传文件数")
+    upload_path = Column(String(256), nullable=True, comment="文件路径")
+    timeout = Column(TIMESTAMP, server_default=func.current_timestamp(), comment="上传请求超时时间")
+    service_id = Column(String(64), nullable=True, comment="上传请求所属服务：DATA-MANAGEMENT(数据管理)")
+    check_info = Column(Text, nullable=True, comment="业务信息")
+    
+    def __repr__(self):
+        return f"<ChunkUploadRequest(id={self.id}, service_id={self.service_id}, progress={self.uploaded_file_num}/{self.total_file_num})>"

runtime/datamate-python/app/models/dm/__init__.py

@@ -0,0 +1,21 @@
+# app/models/dm/__init__.py
+
+from .annotation_template import AnnotationTemplate
+from .labeling_project import LabelingProject
+from .dataset import Dataset
+from .dataset_files import DatasetFiles
+from .dataset_statistics import DatasetStatistics
+from .dataset_tag import DatasetTag
+from .tag import Tag
+from .user import User
+
+__all__ = [
+    "AnnotationTemplate",
+    "LabelingProject",
+    "Dataset",
+    "DatasetFiles",
+    "DatasetStatistics",
+    "DatasetTag",
+    "Tag",
+    "User"
+]

runtime/datamate-python/app/models/dm/annotation_template.py

@@ -0,0 +1,24 @@
+from sqlalchemy import Column, String, JSON, TIMESTAMP
+from sqlalchemy.sql import func
+from app.db.database import Base
+import uuid
+
+class AnnotationTemplate(Base):
+    """标注模板模型"""
+    
+    __tablename__ = "t_dm_annotation_templates"
+    
+    id = Column(String(36), primary_key=True, default=lambda: str(uuid.uuid4()), comment="UUID主键ID")
+    name = Column(String(32), nullable=False, comment="模板名称")
+    description = Column(String(255), nullable=True, comment="模板描述")
+    configuration = Column(JSON, nullable=True, comment="配置信息（JSON格式）")
+    created_at = Column(TIMESTAMP, server_default=func.current_timestamp(), comment="创建时间")
+    deleted_at = Column(TIMESTAMP, nullable=True, comment="删除时间（软删除）")
+    
+    def __repr__(self):
+        return f"<AnnotationTemplate(id={self.id}, name={self.name})>"
+    
+    @property
+    def is_deleted(self) -> bool:
+        """检查是否已被软删除"""
+        return self.deleted_at is not None

runtime/datamate-python/app/models/dm/dataset.py

@@ -0,0 +1,35 @@
+from sqlalchemy import Column, String, Text, BigInteger, Integer, Boolean, JSON, TIMESTAMP
+from sqlalchemy.sql import func
+from app.db.database import Base
+import uuid
+
+class Dataset(Base):
+    """数据集模型（支持医学影像、文本、问答等多种类型）"""
+    
+    __tablename__ = "t_dm_datasets"
+    
+    id = Column(String(36), primary_key=True, default=lambda: str(uuid.uuid4()), comment="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")
+    category = Column(String(100), nullable=True, comment="数据集分类：医学影像/问答/文献等")
+    path = Column(String(500), nullable=True, comment="数据存储路径")
+    format = Column(String(50), nullable=True, comment="数据格式：DCM/JPG/JSON/CSV等")
+    schema_info = Column(JSON, nullable=True, comment="数据结构信息")
+    size_bytes = Column(BigInteger, default=0, comment="数据大小(字节)")
+    file_count = Column(BigInteger, default=0, comment="文件数量")
+    record_count = Column(BigInteger, default=0, comment="记录数量")
+    retention_days = Column(Integer, default=0, comment="数据保留天数（0表示长期保留）")
+    tags = Column(JSON, nullable=True, comment="标签列表")
+    metadata = Column(JSON, nullable=True, comment="元数据信息")
+    status = Column(String(50), default='DRAFT', comment="状态：DRAFT/ACTIVE/ARCHIVED")
+    is_public = Column(Boolean, default=False, comment="是否公开")
+    is_featured = Column(Boolean, default=False, comment="是否推荐")
+    version = Column(BigInteger, nullable=False, default=0, comment="版本号")
+    created_at = Column(TIMESTAMP, server_default=func.current_timestamp(), comment="创建时间")
+    updated_at = Column(TIMESTAMP, server_default=func.current_timestamp(), onupdate=func.current_timestamp(), comment="更新时间")
+    created_by = Column(String(255), nullable=True, comment="创建者")
+    updated_by = Column(String(255), nullable=True, comment="更新者")
+    
+    def __repr__(self):
+        return f"<Dataset(id={self.id}, name={self.name}, type={self.dataset_type})>"

runtime/datamate-python/app/models/dm/dataset_files.py

@@ -0,0 +1,27 @@
+from sqlalchemy import Column, String, JSON, BigInteger, TIMESTAMP
+from sqlalchemy.sql import func
+from app.db.database import Base
+import uuid
+
+class DatasetFiles(Base):
+    """DM数据集文件模型"""
+    
+    __tablename__ = "t_dm_dataset_files"
+    
+    id = Column(String(36), primary_key=True, default=lambda: str(uuid.uuid4()), comment="UUID")
+    dataset_id = Column(String(36), nullable=False, comment="所属数据集ID（UUID）")
+    file_name = Column(String(255), nullable=False, comment="文件名")
+    file_path = Column(String(1000), nullable=False, comment="文件路径")
+    file_type = Column(String(50), nullable=True, comment="文件格式：JPG/PNG/DCM/TXT等")
+    file_size = Column(BigInteger, default=0, comment="文件大小(字节)")
+    check_sum = Column(String(64), nullable=True, comment="文件校验和")
+    tags = Column(JSON, nullable=True, comment="文件标签信息")
+    metadata = Column(JSON, nullable=True, comment="文件元数据")
+    status = Column(String(50), default='ACTIVE', comment="文件状态：ACTIVE/DELETED/PROCESSING")
+    upload_time = Column(TIMESTAMP, server_default=func.current_timestamp(), comment="上传时间")
+    last_access_time = Column(TIMESTAMP, nullable=True, comment="最后访问时间")
+    created_at = Column(TIMESTAMP, server_default=func.current_timestamp(), comment="创建时间")
+    updated_at = Column(TIMESTAMP, server_default=func.current_timestamp(), onupdate=func.current_timestamp(), comment="更新时间")
+    
+    def __repr__(self):
+        return f"<DatasetFiles(id={self.id}, dataset_id={self.dataset_id}, file_name={self.file_name})>"

runtime/datamate-python/app/models/dm/dataset_statistics.py

@@ -0,0 +1,25 @@
+from sqlalchemy import Column, String, Date, BigInteger, JSON, TIMESTAMP
+from sqlalchemy.sql import func
+from app.db.database import Base
+import uuid
+
+class DatasetStatistics(Base):
+    """数据集统计信息模型"""
+    
+    __tablename__ = "t_dm_dataset_statistics"
+    
+    id = Column(String(36), primary_key=True, default=lambda: str(uuid.uuid4()), comment="UUID")
+    dataset_id = Column(String(36), nullable=False, comment="数据集ID（UUID）")
+    stat_date = Column(Date, nullable=False, comment="统计日期")
+    total_files = Column(BigInteger, default=0, comment="总文件数")
+    total_size = Column(BigInteger, default=0, comment="总大小(字节)")
+    processed_files = Column(BigInteger, default=0, comment="已处理文件数")
+    error_files = Column(BigInteger, default=0, comment="错误文件数")
+    download_count = Column(BigInteger, default=0, comment="下载次数")
+    view_count = Column(BigInteger, default=0, comment="查看次数")
+    quality_metrics = Column(JSON, nullable=True, comment="质量指标")
+    created_at = Column(TIMESTAMP, server_default=func.current_timestamp(), comment="创建时间")
+    updated_at = Column(TIMESTAMP, server_default=func.current_timestamp(), onupdate=func.current_timestamp(), comment="更新时间")
+    
+    def __repr__(self):
+        return f"<DatasetStatistics(id={self.id}, dataset_id={self.dataset_id}, date={self.stat_date})>"

runtime/datamate-python/app/models/dm/dataset_tag.py

@@ -0,0 +1,15 @@
+from sqlalchemy import Column, String, TIMESTAMP
+from sqlalchemy.sql import func
+from app.db.database import Base
+
+class DatasetTag(Base):
+    """数据集标签关联模型"""
+    
+    __tablename__ = "t_dm_dataset_tags"
+    
+    dataset_id = Column(String(36), primary_key=True, comment="数据集ID（UUID）")
+    tag_id = Column(String(36), primary_key=True, comment="标签ID（UUID）")
+    created_at = Column(TIMESTAMP, server_default=func.current_timestamp(), comment="创建时间")
+    
+    def __repr__(self):
+        return f"<DatasetTag(dataset_id={self.dataset_id}, tag_id={self.tag_id})>"

runtime/datamate-python/app/models/dm/labeling_project.py

@@ -0,0 +1,26 @@
+from sqlalchemy import Column, String, Integer, JSON, TIMESTAMP
+from sqlalchemy.sql import func
+from app.db.database import Base
+import uuid
+
+class LabelingProject(Base):
+    """DM标注项目模型（原 DatasetMapping）"""
+    
+    __tablename__ = "t_dm_labeling_projects"
+    
+    id = Column(String(36), primary_key=True, default=lambda: str(uuid.uuid4()), comment="UUID主键ID")
+    dataset_id = Column(String(36), nullable=False, comment="数据集ID")
+    name = Column(String(32), nullable=False, comment="项目名称")
+    labeling_project_id = Column(Integer, nullable=False, comment="Label Studio项目ID")
+    configuration = Column(JSON, nullable=True, comment="标签配置")
+    progress = Column(JSON, nullable=True, comment="标注进度统计")
+    created_at = Column(TIMESTAMP, server_default=func.current_timestamp(), comment="创建时间")
+    deleted_at = Column(TIMESTAMP, nullable=True, comment="删除时间（软删除）")
+    
+    def __repr__(self):
+        return f"<LabelingProject(id={self.id}, dataset_id={self.dataset_id}, name={self.name})>"
+    
+    @property
+    def is_deleted(self) -> bool:
+        """检查是否已被软删除"""
+        return self.deleted_at is not None

runtime/datamate-python/app/models/dm/tag.py

@@ -0,0 +1,21 @@
+from sqlalchemy import Column, String, Text, BigInteger, TIMESTAMP
+from sqlalchemy.sql import func
+from app.db.database import Base
+import uuid
+
+class Tag(Base):
+    """标签模型"""
+    
+    __tablename__ = "t_dm_tags"
+    
+    id = Column(String(36), primary_key=True, default=lambda: str(uuid.uuid4()), comment="UUID")
+    name = Column(String(100), nullable=False, unique=True, comment="标签名称")
+    description = Column(Text, nullable=True, comment="标签描述")
+    category = Column(String(50), nullable=True, comment="标签分类")
+    color = Column(String(7), nullable=True, comment="标签颜色(十六进制)")
+    usage_count = Column(BigInteger, default=0, comment="使用次数")
+    created_at = Column(TIMESTAMP, server_default=func.current_timestamp(), comment="创建时间")
+    updated_at = Column(TIMESTAMP, server_default=func.current_timestamp(), onupdate=func.current_timestamp(), comment="更新时间")
+    
+    def __repr__(self):
+        return f"<Tag(id={self.id}, name={self.name}, category={self.category})>"

runtime/datamate-python/app/models/dm/user.py

@@ -0,0 +1,24 @@
+from sqlalchemy import Column, String, BigInteger, Boolean, TIMESTAMP
+from sqlalchemy.sql import func
+from app.db.database import Base
+
+class User(Base):
+    """用户模型"""
+    
+    __tablename__ = "users"
+    
+    id = Column(BigInteger, primary_key=True, autoincrement=True, comment="用户ID")
+    username = Column(String(255), nullable=False, unique=True, comment="用户名")
+    email = Column(String(255), nullable=False, unique=True, comment="邮箱")
+    password_hash = Column(String(255), nullable=False, comment="密码哈希")
+    full_name = Column(String(255), nullable=True, comment="真实姓名")
+    avatar_url = Column(String(500), nullable=True, comment="头像URL")
+    role = Column(String(50), nullable=False, default='USER', comment="角色：ADMIN/USER")
+    organization = Column(String(255), nullable=True, comment="所属机构")
+    enabled = Column(Boolean, nullable=False, default=True, comment="是否启用")
+    last_login_at = Column(TIMESTAMP, nullable=True, comment="最后登录时间")
+    created_at = Column(TIMESTAMP, server_default=func.current_timestamp(), comment="创建时间")
+    updated_at = Column(TIMESTAMP, server_default=func.current_timestamp(), onupdate=func.current_timestamp(), comment="更新时间")
+    
+    def __repr__(self):
+        return f"<User(id={self.id}, username={self.username}, role={self.role})>"

runtime/datamate-python/app/models/operator/__init__.py

@@ -0,0 +1,11 @@
+# app/models/operator/__init__.py
+
+from .operator import Operator
+from .operator_category import OperatorCategory
+from .operator_category_relation import OperatorCategoryRelation
+
+__all__ = [
+    "Operator",
+    "OperatorCategory",
+    "OperatorCategoryRelation"
+]

runtime/datamate-python/app/models/operator/operator.py

@@ -0,0 +1,24 @@
+from sqlalchemy import Column, String, Text, Boolean, TIMESTAMP
+from sqlalchemy.sql import func
+from app.db.database import Base
+
+class Operator(Base):
+    """算子模型"""
+    
+    __tablename__ = "t_operator"
+    
+    id = Column(String(64), primary_key=True, comment="算子ID")
+    name = Column(String(64), nullable=True, comment="算子名称")
+    description = Column(String(256), nullable=True, comment="算子描述")
+    version = Column(String(256), nullable=True, comment="版本")
+    inputs = Column(String(256), nullable=True, comment="输入类型")
+    outputs = Column(String(256), nullable=True, comment="输出类型")
+    runtime = Column(Text, nullable=True, comment="运行时信息")
+    settings = Column(Text, nullable=True, comment="配置信息")
+    file_name = Column(Text, nullable=True, comment="文件名")
+    is_star = Column(Boolean, nullable=True, comment="是否收藏")
+    created_at = Column(TIMESTAMP, server_default=func.current_timestamp(), comment="创建时间")
+    updated_at = Column(TIMESTAMP, server_default=func.current_timestamp(), onupdate=func.current_timestamp(), comment="更新时间")
+    
+    def __repr__(self):
+        return f"<Operator(id={self.id}, name={self.name}, version={self.version})>"

runtime/datamate-python/app/models/operator/operator_category.py

@@ -0,0 +1,15 @@
+from sqlalchemy import Column, String, Integer
+from app.db.database import Base
+
+class OperatorCategory(Base):
+    """算子分类模型"""
+    
+    __tablename__ = "t_operator_category"
+    
+    id = Column(Integer, primary_key=True, autoincrement=True, comment="分类ID")
+    name = Column(String(64), nullable=True, comment="分类名称")
+    type = Column(String(64), nullable=True, comment="分类类型")
+    parent_id = Column(Integer, nullable=True, comment="父分类ID")
+    
+    def __repr__(self):
+        return f"<OperatorCategory(id={self.id}, name={self.name}, type={self.type})>"

runtime/datamate-python/app/models/operator/operator_category_relation.py

@@ -0,0 +1,13 @@
+from sqlalchemy import Column, String, Integer
+from app.db.database import Base
+
+class OperatorCategoryRelation(Base):
+    """算子分类关联模型"""
+    
+    __tablename__ = "t_operator_category_relation"
+    
+    category_id = Column(Integer, primary_key=True, comment="分类ID")
+    operator_id = Column(String(64), primary_key=True, comment="算子ID")
+    
+    def __repr__(self):
+        return f"<OperatorCategoryRelation(category_id={self.category_id}, operator_id={self.operator_id})>"