Troy Hinckley

Your point about character indexing is a good one. There is the famous post by Manish Goregaokar titled [stop ascribing meaning to codepoints](https://manishearth.github.io/blog/2017/01/14/stop-ascribing-meaning-to-unicode-code-points/) which essentially makes that same argument. However...

I think you have convinced me to use graphemes as an abstraction instead of codepoints. Emacs uses codepoints as it's abstraction, and when testing it with multi-codepoint graphemes the incompatibility...

> It'd be fun (and pretty easy) to add a gap buffer to some of the benchmarks we have kicking around. I'd use my own from jumprope, but it doesn't...

same issue. Interestingly, I don't always get this, but I can't seem to figure out why it happens.

That is actually a great idea. You have to be careful with this pattern though that you [don't trigger UB](https://github.com/rust-lang/unsafe-code-guidelines/issues/256). For example if this is my code: ```rust struct ObjectHeader...

> currently cons are 24 bytes. Would we benefit from shrinking to 16? We would benefit greatly. Cons is one of the most common objects (my current running instance has...

> Have you looked at how emacs / cl does this? They use a mark bitmap. https://github.com/emacs-mirror/emacs/blob/eb3a90619fed86298c96951af527a8483bdd1a3c/src/alloc.c#L2811-L2817

Thanks for the issues! I would be very interested in patches to improve this package. Especially anything that improves reliability or user experience.

> question regarding your other package native-completion. Which one do you consider the best long term solution? They serve two different purposes and can't be substituted for each other. `native-complete`...

Thanks for submitting this. It indeed does need a type, but it is not a string. it is form that is evaluated by the modeline. Using `:type 'sexp` should cover...