在sqlalchemy中查询相关表

Question

So I have two tables Employee and Details like this.

class Employee(Base):
    __tablename__ = 'employees'
    id = Column(Integer, Sequence('employee_id_seq'), primary_key=True)
    name = Column(String(50), nullable=False)
    ............

class Detail(Base):
    __tablename__ = 'details'
    id = Column(Integer, Sequence('detail_id_seq'), primary_key=True)
    start_date = Column(String(50), nullable=False)
    email = Column(String(50))
    employee_id = Column(Integer, ForeignKey('employee.id'))
    employee = relationship("Employee", backref=backref('details', order_by=id))
    ............

Now what I want to do is get all the employees and their corresponding details, here is what I tried.

for e, d in session.query(Employee, Detail).filter(Employee.id = Detail.employee_id).all():
    print e.name, d.email

The problem with this is that it prints everything twice. I tried using .join() and also prints the results twice.

What I want to achieve is like

print Employee.name
print Employee.details.email

Answer 1

If you really care only about few columns, you can specify them in the query directly:

q = session.query(Employee.name, Detail.email).filter(Employee.id == Detail.employee_id).all()
for e, d in q:
    print e, d

If you do really want to load object instances, then I would do it differently:

# query all employees
q = (session.query(Employee)
        # load Details in the same query
        .outerjoin(Employee.details)
        # let SA know that the relationship "Employee.details" is already loaded in this query so that when we access it, SA will not do another query in the database
        .options(contains_eager(Employee.details))
        ).all()

# navigate the results simply as defined in the relationship configuration
for e in q:
    print(e)
    for d in e.details:
        print(" ->", d)

As to your duplicate result problem, I believe you have some "extra" in your real code which produces this error...