Wrong line number for invalid iterable that uses __getattr__

[JukkaL]

This generates an error false positive which is reported on the wrong line so there are likely two bugs:

from typing import Union, Tuple

class C:
    def __getattr__(self, name):
        pass

class D:
    def f(self) -> C: # "C" has no attribute "__iter__" (not iterable)
        return C()

    def g(self) -> None:
        a, b = self.f()  # This is where the error originates from

C should be treated as iterable, since it defines __getattr__.

[AlexWaygood]

C should be treated as iterable, since it defines __getattr__.

Is that correct? At runtime, dunders are looked up on the class object rather than the instance for many operations (including the iterator protocol), so __getattr__ doesn't actually work for dunders like __iter__:

>>> class Foo:
...     def __getattr__(self, name):
...         if name == "__iter__":
...             return lambda self: iter(range(10))
...
>>> Foo().__iter__
<range_iterator object at 0x00000194BCD3BB50>
>>> list(Foo().__iter__)
[0, 1, 2, 3, 4, 5, 6, 7, 8, 9]
>>> list(Foo())
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
TypeError: 'Foo' object is not iterable

iter(Foo()) at runtime results in a lookup equivalent to _tmp = Foo(); type(_tmp).__iter__(_tmp).

[AlexWaygood]

Using the original example:

>>> class C:
...     def __getattr__(self, name):
...         if name == "__iter__":
...             return lambda self: iter(range(2))
...
>>> class D:
...     def f(self):
...         return C()
...     def g(self):
...         a, b = self.f()
...
>>> D().g()
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "<stdin>", line 5, in g
TypeError: cannot unpack non-iterable C object

[JukkaL]

Yes, you are quite right. So the issue is only about the incorrect line number.