--> s = []
--> s += 1
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
TypeError: 'int' object is not iterable

For comparison, when NotImplemented is appropriately returned the message looks like this:

--> s -= 1
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
TypeError: unsupported operand type(s) for -=: 'list' and 'int'

Which is certainly more helpful than "<blank> object is not iterable"

Please don't replicate existing issues.

Also, this behavior for lists is long-standing, tested, and designed by Guido.  More than a little code relies on this behavior.

My understanding was that if a different patch will be needed, it is not the same issue.  Is that not correct?

Here's the test:

    def test_iadd(self):
        ...
        self.assertRaises(TypeError, u.__iadd__, None)

It would still pass, as returning NotImplemented will still result in a TypeError, just with a friendlier message.

If any code is depending on this, it is depending on the error message text, which is explicitly not guaranteed

nosy: + rhettinger
messages: + msg230401

assignee: rhettinger
resolution: not a bug