"""
知识库文档模型
"""
from sqlalchemy import Column, String, Text, Integer, Boolean, DateTime, BigInteger

from app.base_model import BaseModel


class KnowledgeDocument(BaseModel):
    """
    知识库文档
    
    记录上传到知识库的文档信息及处理状态
    """
    __tablename__ = "ai_knowledge_document"

    knowledge_base_id = Column(String(21), nullable=False, index=True, comment="所属知识库ID（逻辑外键关联ai_knowledge_base）")
    file_id = Column(String(21), nullable=True, comment="关联文件ID（逻辑外键关联core_file_manager）")
    name = Column(String(255), nullable=False, comment="文档名称")
    file_type = Column(String(20), nullable=True, comment="文件类型: pdf/docx/txt/md/xlsx/csv/html/pptx")
    file_size = Column(BigInteger, default=0, comment="文件大小（字节）")
    content_hash = Column(String(64), nullable=True, index=True, comment="内容MD5（用于去重）")

    # 处理结果
    segment_count = Column(Integer, default=0, comment="分段数量")
    token_count = Column(Integer, default=0, comment="Token 总数")
    char_count = Column(Integer, default=0, comment="字符总数")

    # 处理状态
    status = Column(String(20), default="pending", index=True, comment="状态: pending/indexing/completed/failed/disabled")
    error_message = Column(Text, nullable=True, comment="错误信息")
    indexing_started_at = Column(DateTime, nullable=True, comment="索引开始时间")
    indexing_completed_at = Column(DateTime, nullable=True, comment="索引完成时间")

    # 去重
    duplicate_warning = Column(Text, nullable=True, comment="内容重复警告（跨知识库检测）")

    # 是否启用
    enabled = Column(Boolean, default=True, comment="是否启用（禁用后不参与检索）")