Vít Kučera

OK, but what is the point of showing this message for a code that is perfectly fine as opposed to a code that seems to have the SC2028 error? Now...

Besides, ShellCheck exits with 1 which makes it look like there is something to worry about which, to my understanding, is not the case with `\\t`.

> You can specify severity level to avoid info messages. > > ``` > -S SEVERITY --severity=SEVERITY Minimum severity of errors to consider (error, warning, info, style) > ``` >...

Bisecting the issue located the origin to be between commits: ee112dfe1da04590a0fd8ed9656eb4a83a948596 (last good) and de19eb1ddb9d1b2edcb5d2187078b2d67bb7b784 (first bad), concluding that the bug was introduced with https://github.com/root-project/root/pull/11252 .

Commit 9dd78b6ea3 is the culprit.

Yes, the Python linters seem to run in the following order: 1. black 2. isort 3. ruff 4. pylint 5. mypy 6. pyright 7. flake8 8. bandit If black, isort...

Besides, another benefit of running the formatters last would be the possibility to speedup MegaLinter's running time by skipping the formatters if some of the linters have failed.

> But on the other hand, some easily fixed issues, like line length, types of quotes etc, that can be fixed by a formatter without failing, will make the linters...

I would agree with you if the non-formatting linters could fail because of a formatting issue (hence my last, yet unanswered, question). But I don't think that is the case....

OK, if you say that formatting by Black can fix something that would make Pylint fail, then I agree that it makes sense to run the formatters first.