Hacker News Reader

I've always viewed unsafe Rust as a sort of last resort you use when you just can't do something safely. Porting something to "unsafe Rust" to me feels pointless, and of course this hasty rewrite was probably not the best idea from any perspective.

View on HN

If the end state were unsafe Rust, you'd be right, but that's explicitly not the intended end state. Unsafe Rust was the reachable first step. Now they have a Rust version and can iteratively remove `unsafe`, which they couldn't do before when it was in Zig.

It's the wrong abstraction though and I am kind of not surprised the Bun maintainers went way. The correct abstraction would be to translate into Rust while using `clone` and `copy` liberally and then iteratively converting to a borrowing model. Nowhere at all was an introduction of the `unsafe` keyword needed.

That's only the correct abstraction if you intended to hold off on releasing the rewrite until you were completely finished, because there's no universe where Bun of all projects releases a production version that uses clone and copy liberally.

Unsafe in theory in a transliteration is only as unsafe as the original Zig was, which makes it a perfectly reasonable place to land as a transition point in a way that introducing large numbers of allocations simply isn't.

Reply on news.ycombinator.com