Wemake-python-styleguide: WPS122 should allow a no-op loop var

[webknjaz]

Bug report

What's wrong

Imagine I want to do something 3 times but I don't care about the attempt number. So I'd normally write something like this:

for _ in range(3):
    do_something()

But then the linter complains with WPS122 Found all unused variables definition: _ which is incorrect. While things like assignment allow to fix this with

-_something_unused = get_useless_result()
+get_useless_result()

it is impossible with for-loops not to define this loop variable at all.

How is that should be

This case shouldn't be a violation of the given rule.

System information

N/A

[sobolevn]

I agree, this is a valid use-case. Thanks a lot!

[sobolevn]

The same applies for comprehensions.

[nnuzaba]

Hi, I would like to work on this issue.

I’m a newcomer to this project (and open source in general), so I was hoping for some help and clarification :).

Is it enough to just get rid of the visit_any_for function .

Or should I keep it and try to do something more complex where I detect if the for loop is trying to iterate through a range of numbers, and only then not raise an error?

Is there anything extra that has to be done for comprehensions? I couldn’t really find anything specific to that.

Thank you!

[sobolevn]

@nn3un thanks a lot!

Is it enough to just get rid of the visit_any_for function

Looks like it!

Is there anything extra that has to be done for comprehensions?

Yes, we should allow for _ in ... there too.

I’m a newcomer to this project (and open source in general), so I was hoping for some help and clarification

Feel very welcome! 👍

[vvkh]

Was this actually fixed?
I'm still getting the same error (version 0.14.1).

$ cat test.py

def do_something():
    return

for _ in range(3):
    do_something()

$ flake8 test.py

test.py:1:1: D100 Missing docstring in public module
test.py:1:1: D103 Missing docstring in public function
test.py:2:5: WPS324 Found inconsistent `return` statement
test.py:5:1: WPS122 Found all unused variables definition:

Is this a regression or I'm doing something wrong?

[sobolevn]

It is not released. Will be released in 0.15

[Pejnikov]

The line "Fixes WPS122 not raising for for and async for definitions" in the Bugfixes of "0.14.0" confuses so much when trying to understand why it doesn't work for "0.14.1".

[sobolevn]

@Pejnikov can you please move it to the correct section? Thanks! 👍