如何让SQLAlchemy多个子类对象共享同一个父类实例？

首先明确：你当前用的SQLAlchemyJoined Table多态继承模式，本身不支持一个父类（Employee）实例对应多个子类（Manager+Engineer）实例。因为多态继承依赖polymorphic_on字段（这里是type）来区分子类，每个Employee记录的type只能是单一值（比如"manager"或"engineer"），没法同时对应两个子类。

要实现一个Employee对应Manager和Engineer的需求，你需要放弃多态继承，改用关联关系来设计模型，具体如下：

修改后的模型代码 #

from sqlalchemy import ForeignKey
from sqlalchemy.orm import DeclarativeBase, Mapped, mapped_column, relationship

class Base(DeclarativeBase):
    pass

class Employee(Base):
    __tablename__ = "employee"
    id: Mapped[int] = mapped_column(primary_key=True)
    name: Mapped[str]
    # 建立与子类的双向关联，uselist=False表示一对一
    engineer: Mapped["Engineer"] = relationship(back_populates="employee", uselist=False)
    manager: Mapped["Manager"] = relationship(back_populates="employee", uselist=False)

    def __repr__(self):
        return f"Employee({self.name!r} {self.id!r})"

class Engineer(Base):
    __tablename__ = "engineer"
    id: Mapped[int] = mapped_column(ForeignKey("employee.id"), primary_key=True)
    engineer_name: Mapped[str]
    # 关联到Employee
    employee: Mapped[Employee] = relationship(back_populates="engineer")

class Manager(Base):
    __tablename__ = "manager"
    id: Mapped[int] = mapped_column(ForeignKey("employee.id"), primary_key=True)
    manager_name: Mapped[str]
    # 关联到Employee
    employee: Mapped[Employee] = relationship(back_populates="manager")

使用方式 #

先创建Employee实例，再基于同一个Employee创建Manager和Engineer：

# 创建Employee并提交
emp = Employee(name='name1')
session.add(emp)
session.commit()

# 关联同一个Employee创建Manager和Engineer
manager = Manager(id=emp.id, manager_name="manager1", employee=emp)
engineer = Engineer(id=emp.id, engineer_name="engineer1", employee=emp)
session.add_all([manager, engineer])
session.commit()

补充说明 #

如果你坚持想用继承语法，也可以让Engineer和Manager都继承自Employee，但去掉多态相关配置（__mapper_args__和polymorphic_on），本质还是关联关系，只是子类继承了Employee的字段，但查询时需要手动关联，不如上面的方案直接。

内容的提问来源于stack exchange，提问作者wsl