AshesITR

Here are the total timing stats of a820b6e by linter on the 20 packages. `indentation_linter` took 1.3% of the total time spent, putting it at rank 59/73 of all linters...

> So yes, we should not lint by default. I wouldn't even make it an option for now. The difficulty with that is, that hanging indents are _also_ okay and...

IINM auto-indent of RStudio and PyCharm does make these hanging though... I also tend to use that style if overall readability benefits from the fewer lines used up by that...

Feel free to chip in as I don't have much time to contribute to lintr at the moment.

I've resolved all merge conflicts so work on this branch can continue. Note to self: Don't de-lint lintr for new linters until the PR is merged.

@MichaelChirico PTAL. The CI failure is because `indentation_linter()` applies to us as well. If the review is positive, I will see to fixing the lints. In an effort to prevent...

> I notice there's a lot of "false positives" where indentation is used to align across lines (e.g. tests/testthat/test-implicit_integer_linter.R)... it would be great if this linter could recognize that and...

Here is a table of the most recent lint counts ``` package N 1: sf 4705 -- indent = 1 2: rgl 1635 -- indent = 4 3: rgeos 1146...

After giving this some more thought: Handling mixtures of tabs and spaces for indenting is very ugly when trying to obtain correct lint locations, so I'd punt on the vctrs...

Hi, can you name the package that `lint_package()` fails on? Or, even better, try to provide a minimal [reprex](https://reprex.tidyverse.org/articles/learn-reprex.html)?