What's the difference between the mro method and the __mro__ attribute of a class?

Question

I stumbled across this extra, no-underscores mro method when I was using __metaclass__ = abc.ABCMeta. It seems to be the same as __mro__ except that it returns a list instead of a tuple. Here's a random example (ideone snippet):

import abc
import copy

class Life(object):
    __metaclass__ = abc.ABCMeta

    @abc.abstractmethod
    def reproduce(self):
        pass

class Bacterium(Life):
    def reproduce(self):
        return copy.deepcopy(self)

wiggly = Bacterium()

print wiggly.__class__.__mro__
# (<class '__main__.Bacterium'>, <class '__main__.Life'>, <type 'object'>)

print wiggly.__class__.mro()
# [<class '__main__.Bacterium'>, <class '__main__.Life'>, <type 'object'>]

I found later that this isn't unique to ABCMeta but is available in all new-style classes.

So... why? What is this doing that __mro__ isn't?

Answer 1

Directly from the documentation:

This method can be overridden by a metaclass to customize the method resolution order for its instances. It is called at class instantiation, and its result is stored in _mro_.

So, mro() is called on instantiation and caches its result in __mro__. They have the same purpose.