I don’t think that’s a naive question at all, it’s something I’ve been thinking about a lot recently.

We have something like 70k-75k lines of ObjC and plain C in Hammerspoon, written at various times over the last 7ish years, with varying architectural styles as we’ve figured out better ways of doing things.

I would love to have it all ported to Swift with a single, consistent architecture, but it’s a heck of a lot of work!

I’ve been experimenting with some ideas for how the Swift might look, but to really justify all that work I think we’d need to revamp how our user-facing API works, so we’d also be forcing all our users to rewrite their Lua configs.

In summary… ¯\_(ツ)_/¯

Decision tree so you can stop pondering:

Does the current situation hinder progress?

- Yes: A lot?

  - Yes: rewrite

  - No: don't rewrite

- No: Don't rewrite

IDK i ported my objC code to swift and didn't see much of a benefit. If anything it was more work because I relied heavily on objC dynamism.