I think I found a solution to this a while ago, promptly forgot it and now can't remember where I found it. In any case, here's the issue:

>>> class A:
	def __init__(self):
		self.v1 = 1
		self.v2 = 2

		
>>> class B(A):
	def __init__(self):
		self.v2 = 4
		self.v3 = 3

		
>>> c = B()
>>> c.v3
3
>>> c.v2
4
>>> c.v1

Traceback (most recent call last):
  File "<pyshell#84>", line 1, in <module>
    c.v1
AttributeError: B instance has no attribute 'v1'
>>>

I can see why this is happening - B's init is overwriting A's completely. Is there a convenient way around this, so that I can leave v1 untouched in B while overwriting v2 and adding v3?

EDIT: Typo :(

I think you want something like this:

class A(object):
    def __init__(self):
        self.v1 = 1
        self.v2 = 2


class B(A):
    def __init__(self):
        super(B, self).__init__()
        self.v2 = 4
        self.v3 = 3
        
c = B()
print c.v3, c.v2, c.v1

>C:\Python25\pythonw -u "testClass.py"
3 4 1
>Exit code: 0

Note that you need to explicitly declare A as a child of objectfor super() to work correctly.

What about:

>>> class A(object):
…     def __init__(self):
…             self.v1 = 1
…             self.v2 = 2
…
>>> class B(A):
…     def __init__(self):
…             super(B, self).__init__()
…             self.v2 = 4
…             self.v3 = 3
…
>>> c = B()
>>> c.v3
3
>>> c.v2
4
>>> c.v1
1
>>>

Edit: Ninja'd!

Perfect! Thank you both.