False negative Pylint Fails to output W0199 on single item tuple

[JasonF-BJSS]

Steps to reproduce

Given a file a.py:

# pylint: disable=missing-docstring
assert (1 == 2, "r")
assert (1 == 2,)

Current behaviour

Result of pylint a.py:

************* Module a
a.py:2:0: W0199: Assert called on a 2-item-tuple. Did you mean 'assert x,y'? (assert-on-tuple)

Expected behavior

************* Module a
a.py:2:0: W0199: Assert called on a 2-item-tuple. Did you mean 'assert x,y'? (assert-on-tuple)
a.py:3:0: W0199: Assert called on a 2-item-tuple. Did you mean 'assert x,y'? (assert-on-tuple)

pylint --version output

Result of pylint --version output:

pylint 2.6.2
astroid 2.4.2
Python 3.8.7 (v3.8.7:6503f05dd5, Dec 21 2020, 12:45:15) 
[Clang 6.0 (clang-600.0.57)]

[yushao2]

@cdce8p, this seems to be expected behavior from the already implemented functional test:

https://github.com/PyCQA/pylint/blob/main/tests/functional/a/assert_on_tuple.py#L8

[cdce8p]

@cdce8p, this seems to be expected behavior from the already implemented functional test:

Indeed, thanks for the lookup! I didn't look at it closely when adding the tag.
Just the question if it should change then otherwise this issue could be closed.

[Pierre-Sassoulas]

What is the legitimate use case of assert (1 == 2,) ?

[JasonF-BJSS]

There is no legitimate use case I could ever see for assert (1 == 2,), the linter should warn as per assert (1 == 2, "r").

The compiler warns on this in console:

assert (1 == 2,)
:1: SyntaxWarning: assertion is always true, perhaps remove parentheses?

[Pierre-Sassoulas]

It's minor as like @JasonF-BJSS said, python itself is raising a warning, so it will be caught when the code is run but pylint needs to warn beforehand. So the functional tests are wrong, we need to remove them and raise a warning.