Unfortunately, undoing a git clean is impossible, as C’s unlink() is called on each of the untracked files (and directories if -d is supplied).

I was working on a rails application earlier this week and wanted to implement a third party login authentication system. So I created a feature branch and began working on implementing the OAuth flow.

I created a new file to store all of my API keys and secrets, and quickly thereafter added this file to my ~/.gitignore so as not to track sensitive information.

Before I could complete the feature, I had to switch back to master to push a quick bug-fix. In the heat of the moment, I wanted to clean my working directory and ran git clean -fd to remove all untracked files and directories.

Unfortunately, this removed the file containing all the API keys and secrets, and I had to visit each of the API provider’s sites to recover them.

To prevent this sort of thing from happening in the future, I created a git alias named git clear.

If you add this to your ~/.gitconfig, you won’t permanently lose your untracked files.

[alias]
  clear = stash --keep-index --include-untracked

This will stash all and only your untracked files, and the --keep-index option will retain the state of the staging area.

If you accidentally “remove” untracked files with the new alias:

$ git clear

You can retrieve them with git stash pop. If you’ve stashed a few things since running git clear, you’ll have to specify the stash number, which you can find with git stash list.