python子类增加属性_Python：如何在基类的方法中获取子类的新属性名称？

I want to put all attribute names in SubClass to a list, I want to do that in Base class. How can I do that?

My method now is:

class Base():

def __init__(self):

# print SubClass' new attribues' names ('aa' and 'bb' for example)

for attr in dir(self):

if not hasattr(Base, attr):

print attr

class SubClass(Base):

aa = ''

bb = ''

is there a better way to do that ?

Thanks for any help.

解决方案

As the @JohnZwinck suggested in a comment, you're almost certainly making a design mistake if you need to do this. Without more info, however, we can't diagnose the problem.

This seems the simplest way to do what you want:

class Base(object):

cc = '' # this won't get printed

def __init__(self):

# print SubClass' new attribues' names ('aa' and 'bb' for example)

print set(dir(self.__class__)) - set(dir(Base))

class SubClass(Base):

aa = ''

bb = ''

foo = SubClass()

# set(['aa', 'bb'])

The reason you need self.__class__ instead of self to test for new attributes is this:

class Base(object):

cc = ''

def __init__(self):

print set(dir(self)) - set(dir(Base))

# print SubClass' new attribues' names ('aa' and 'bb' for example)

class SubClass(Base):

def __init__(self):

self.dd = ''

super(SubClass, self).__init__()

aa = ''

bb = ''

foo = SubClass()

# set(['aa', 'dd', 'bb'])

If you want the difference between the classes as defined, you need to use __class__. If you don't, you'll get different results depending on when you check for new attributes.