Over at#3790 (comment) discusses gitignoring it and the outcome was not to do it.

Thanks for bringing it up! I see that the following options have been considered:

1. Tell git to ignore changes to .envrc or to never check out the file (after some research, doesn't seem possible)
2. Tell direnv to source an alternative file

There is another option, but it isn't straightforward. We could commit the file to the repository, and then update the.gitignore, so that anyone can overwrite it locally and adjust it to their own needs, but won't overwrite the remote. I don't know what's the adoption ofdirenv and how many devs have their own custom.envrc (as@mafredri).

If we don't agree on marking.envrc as ignored, I can update our CONTRIBUTING docs with instructions to add the file to global .gitignore. I'm afraid that this file will start leaking to community pull-requests as people tend togit add --all.

Over at#3790 (comment) discusses gitignoring it and the outcome was not to do it.

@ghuntley I was probably unclear on that point. Adding to.gitignore is fine (and good even IMO). It was simply the case that having.envrc committed to the repoand added to.gitignore produces no additional benefit for the issue.

I probably would've raised a PR like this myself but.envrc is part of my personal.gitignore so I didn't notice. 😅

Ok, glad to see that we've an agreement, so I will merge it :)