Stream: t-compiler

Topic: impl trait stabilization


nikomatsakis (May 02 2019 at 14:25, on Zulip):

@centril regarding the prospect of stabilizing impl Trait and so forth --

one of the first steps I think we need is probably trying to breakdown the current existential type feature into layers than we can consider stabilizing separately. I wanted to deprioritize https://github.com/rust-lang/rust/issues/60407 (P-medium) but add it to a tracking issue, but I couldn't even quite figure out where to do that. I guess https://github.com/rust-lang/rust/issues/34511 could be extended? It doesn't seem to cover the new uses. Perhaps there are other tracking issues?

centril (May 02 2019 at 14:27, on Zulip):

@nikomatsakis I think one of the first orders of business is to break up the impl Trait tracking issue into several pieces and finally close the tracking issue; and then we should do a survey of the outstanding issues; perhaps a WG

centril (May 02 2019 at 14:28, on Zulip):

I believe there's just one tracking issue (and that's part of the problem... it's sort of unactionable)

nikomatsakis (May 02 2019 at 14:40, on Zulip):

I too feel a WG is a good option here

nikomatsakis (May 02 2019 at 14:40, on Zulip):

There feels like a lot of ground to cover

nikomatsakis (May 02 2019 at 14:40, on Zulip):

Among other things

nikomatsakis (May 02 2019 at 14:40, on Zulip):

impl Trait is not well documented from what I understand

nikomatsakis (May 02 2019 at 14:41, on Zulip):

(many people have brought this up to me over the last few days, for some reason)

centril (May 02 2019 at 14:47, on Zulip):

@nikomatsakis do you mean docs in the rustc guide, reference, the book, or somewhere else?

nikomatsakis (May 02 2019 at 14:47, on Zulip):

I think that is a question to be answered :)

centril (May 02 2019 at 14:47, on Zulip):

I'm not up to speed on either :D

nikomatsakis (May 02 2019 at 14:47, on Zulip):

what I think people were referring to was largely the book

nikomatsakis (May 02 2019 at 14:47, on Zulip):

I don't think we necessarily want to edit the book

nikomatsakis (May 02 2019 at 14:47, on Zulip):

but I do think we want "explainers" or something

centril (May 02 2019 at 14:48, on Zulip):

@nikomatsakis there are some good source materials in the combined RFCs to build upon

nikomatsakis (May 02 2019 at 14:48, on Zulip):

yes

nikomatsakis (May 02 2019 at 14:48, on Zulip):

I think going forward, writing a user-friendly explainer that thoroughly covers the feature should be part of the process

nikomatsakis (May 02 2019 at 14:48, on Zulip):

and I see no reason not to start here

nikomatsakis (May 02 2019 at 14:48, on Zulip):

(particularly as we have spent a lot of time thinking about this)

centril (May 02 2019 at 14:49, on Zulip):

@nikomatsakis happy to iterate on this, perhaps let's build a template or something

nikomatsakis (May 02 2019 at 14:49, on Zulip):

yes. It's also a place where I think we can do outreach --

nikomatsakis (May 02 2019 at 14:49, on Zulip):

i.e., I'd like each WG to have a set of "explainer people"

nikomatsakis (May 02 2019 at 14:49, on Zulip):

anyway, this is straying onto #t-lang/wg-meta -- maybe we can take it there even :)

centril (May 02 2019 at 14:49, on Zulip):

:slight_smile:

Last update: Nov 22 2019 at 04:30UTC