# Python 3
class Point(tuple):
def __init__(self, x, y):
super().__init__((x, y))
Point(2, 3)
would result in
TypeError: tuple() takes at most 1 argument (2 given)
Why? What should I do instead?
# Python 3
class Point(tuple):
def __init__(self, x, y):
super().__init__((x, y))
Point(2, 3)
would result in
TypeError: tuple() takes at most 1 argument (2 given)
Why? What should I do instead?
tuple
is an immutable type. It's already created and immutable before __init__
is even called. That is why this doesn't work.
If you really want to subclass a tuple, use __new__
.
>>> class MyTuple(tuple):
... def __new__(typ, itr):
... seq = [int(x) for x in itr]
... return tuple.__new__(typ, seq)
...
>>> t = MyTuple((1, 2, 3))
>>> t
(1, 2, 3)
Point(2,3)
looks for a __new__
method in Point
, fails to find it, invokes tuple.__new__(Point,2,3)
which fails since it doesn't simply pass arbitrary number of arguments to __init__
like a user-defined class would, but actually requires a single iterator to properly initialize tuple
, correct? The code does not even get to the Point.__init__(self, 2, 3)
call; that call, if it did occur, would be incorrect too since tuple
does not have __init__
, and so object.__init__
would just silently do nothing? –
Banner © 2022 - 2024 — McMap. All rights reserved.