Stream: project leads (public)

Topic: Edition policy, rfc#2795


Florian Gilcher (Jul 17 2020 at 13:36, on Zulip):

That's a good point. It also isn't used much as far as I see, so there may actually be a good case for it.

nikomatsakis (Jul 17 2020 at 13:36, on Zulip):

@Florian Gilcher It's rather off topic, but I am pretty sure that, in practice, this does't invalidate existing code so much as add options.

nikomatsakis (Jul 17 2020 at 13:37, on Zulip):

(Off topic only in the sense that it's specific to that particular RFC and not so much to the topic of edition policy in general)

nikomatsakis (Jul 17 2020 at 13:38, on Zulip):

Basically there's an "edge case" where you can write panic!("foo {}") and that happens to panic with the string {}, but if you write panic!("foo {}", 22), that substitutes 22. I didn't know this until recently. That's the edge case that is affected by this RFC, essentially.

nikomatsakis (Jul 17 2020 at 13:38, on Zulip):

Also this behavior is different between core::panic and std::panic

nikomatsakis (Jul 17 2020 at 13:39, on Zulip):

(forked to a distinct topic, now it's on topic again :)

nikomatsakis (Jul 17 2020 at 13:41, on Zulip):

side note that I've been writing some TypeScript for fun on the side, and I do find that the JS "interpolation strings" are really just so much more convenient...

Florian Gilcher (Jul 17 2020 at 13:41, on Zulip):

Yeah, didn't want to turn that into a discussion about this - it's a good example for a small change that needs an edition, but maybe not motivate one per se.

Last update: Jun 20 2021 at 01:15UTC