Question

我有以下情况：

class A(Base):
    a_string = Column(String)

class B(Base):
    id = Column(Integer, primary_key=True)

class C(Base):
    b_id = Column(Integer, ForeignKey(B.id))
    value = Column(String)

我需要执行以下查询：

SELECT c2.*
  FROM a
  JOIN c c1 on c1.value = a.a_string
  JOIN c c2 on c2.b_id = c1.b_id

问题是我需要在模型A内部的关系中进行上述查询。例如：

class A(Base):
    a_string = Column(String)
    c_list = relationship('C', secondary=...)

Answer 1

为了自我加入，您需要C的别名：

In [3]: c_to_c = aliased(C.__table__)

然后，您可以定义主要和次要连接以满足您的需要。这种关系应该只能查看：

In [4]: A.c_list = relationship(
   ...:     C, secondary=c_to_c,
   ...:     primaryjoin=A.a_string == c_to_c.c.value,
   ...:     secondaryjoin=C.b_id == c_to_c.c.b_id,
   ...:     viewonly=True
   ...: )

然后，您可以验证您是否可以使用加入的预先加载：

In [5]: print(session.query(A).options(joinedload(A.c_list)))
SELECT a.id AS a_id, a.a_string AS a_a_string, c_1.id AS c_1_id, c_1.b_id AS c_1_b_id, c_1.value AS c_1_value 
FROM a LEFT OUTER JOIN (c AS c_2 JOIN c AS c_1 ON c_1.b_id = c_2.b_id) ON a.a_string = c_2.value

SQLAlchemy与自引用二级关系

1 个答案: