Test case for issue #94 - Multiple occurences of year in license

[mdeweerd]

Test case fro issue #94 .

I triesd a simple fix, but this impacts the other test cases. So it's best that the codeowner checks this.

[Lucas-C]

Hi @mdeweerd

Thank you very much for contributing to this project and attempting to fix an existing issue!

I triesd a simple fix, but this impacts the other test cases. So it's best that the codeowner checks this.

I checked the pipeline execution, and the only tests that are failing are the ones related to your new tests cases.

You may have been mislead in thinking that other tests are failing by this output:

usage: pytest [-h] --whitespaces-count WHITESPACES_COUNT
              [filenames [filenames ...]]
pytest: error: argument --whitespaces-count: invalid int value: ''
usage: pytest [-h] --whitespaces-count WHITESPACES_COUNT
              [filenames [filenames ...]]
pytest: error: argument --whitespaces-count: expected one argument
usage: pytest [-h] --whitespaces-count WHITESPACES_COUNT
              [filenames [filenames ...]]
pytest: error: argument --whitespaces-count: invalid int value: 'a.b'
usage: pytest [-h] --whitespaces-count WHITESPACES_COUNT
              [filenames [filenames ...]]
pytest: error: argument --whitespaces-count: invalid int value: 'a/b'

But this is normal output, due to tests in tests/remove_tabs_test.py and the use of the -s Pytest option in .pre-commit-config.yaml.

I have just added a commit to get rid of this -s / --capture=no option so if you rebase this PR, you should not see such misleading debugging output anymore.

[mdeweerd]

Thank you for your quick feedback.

Yes, my intention is that these test cases validate the feature.

When I tried to amend the code, I ran into other test cases failing - the way I tried to fix it had too many side-effects and I had other tests that were failing.

I think that with the knowledge about how the tool is constructed, a better approach to fix it than mine is easier to determine.

[Lucas-C]

I think that with the knowledge about how the tool is constructed, a better approach to fix it than mine is easier to determine.

As you wish 🙂
Feel free to close this PR if you do not want to continue working on this.

On my side, I don't know when I'll take the time to try to solve this issue,
but I always try to review contributions quickly,
so other PRs to fix this are welcome!

[mdeweerd]

Ok, I do not have time to understand this codebase enough to fix this given the experience with the first attempt to fix it.