Rain

This appears to be repository corruption inside your .hg/git store. I'd suggest recloning.

Sadly I learned that uname -o doesn't have a (portable?) programmatic API. Other uses of uname can probably be replaced though.

I did find https://docs.rs/platform-info/latest/platform_info/, which hardcodes the OS: https://docs.rs/platform-info/latest/src/platform_info/lib_impl.rs.html

Love the feature! The design seems pretty great overall. ~~A thing I would like to suggest we add is a way to retrieve the current Cargo target directory given the...

Ah you're right, my bad. There's plenty of other things that cargo should provide (eg target runner) but the target dir isn't one of them. Consider my comment withdrawn 😊

Apologies if these have already been discussed, finally catching up with this. One thing I'm not totally clear about is how are we going to make the target directory discoverable?...

> Both of those are already a problem with $CARGO_TARGET_DIR, if it is set during one build and not the other. The first is less of a problem because $CARGO_TARGET_DIR/debug/binary/......

Thanks for the issue, and also for maintaining ratatui which I love! This occurs because nextest doesn't allocate a pty for each test. It's something I've been wanting to do...

Thanks! I think the way to address that would be to write an abstraction over the different ways to run various commands, behind an enum or trait object. However, I...

Thanks @wez! The whoami_async approach makes sense to me. > I think the signals side of this might be a bit easier to wrangle Hmm the Tokio code to handle...