Michael Greenberg

I think we shouldn't call anything `try` does 'hardened', because that seems to make a security claim we're not really going to be able to backup. But I think there...

All of this has been done except for #127.

If they're in `/tmp` we shouldn't need to care about it. I can imagine having some kind of `trap 'cleanup' EXIT` that would remove files if things borked _before_ we...

A good first issue: - Identify the temporary files we've created (the two shell scripts, tmp directory) and store them in a `CLEANUP` variable - Set a trap to remove...

I agree that `ps` doesn't quite cut it. These detection approaches seem good; we should also have a flag that lets you pick the shell that runs. Also: these APIs...

When might this happen? For something like #18 we should be deleting the original file... but I can imagine a race where the host system and the `try` process both...

Feels closely related to #23---we should do them both as an overhaul of the commit logic.

Yes, I think that's fine.

Bumping this issue to see if it's stale or still a bug. @adamryczkowski?

I'd be happy to accept a PR! But I doubt I'll have much implementation energy for this any time soon. Others have reported that ffs works fine on WSL, which...