Stream: t-compiler/major changes

Topic: Support const parameters in type dependent compiler-team#304


triagebot (Jun 08 2020 at 12:17, on Zulip):

A new proposal has been announced #304. It will be brought up at the next meeting.

triagebot (Jun 15 2020 at 15:36, on Zulip):

@T-compiler: Proposal #304 has been seconded, and will be approved in 10 days if no objections are raised.

oli (Jun 15 2020 at 15:43, on Zulip):

oh wow, I didn't know about that going on at all

oli (Jun 15 2020 at 15:44, on Zulip):

definitely a sign/anecdotal evidence that MCPs are a good thing

oli (Jun 15 2020 at 15:44, on Zulip):

just read up and very very superficially skimmed the PR. This is pretty cool

triagebot (Jul 01 2020 at 21:03, on Zulip):

This proposal has been accepted: #304.

triagebot (Jul 01 2020 at 21:03, on Zulip):

This proposal has been accepted: #304.

Santiago Pastorino (Jul 01 2020 at 21:55, on Zulip):

unsure why did we get the message twice cc @simulacrum

simulacrum (Jul 01 2020 at 21:55, on Zulip):

eh probably a bug, mind filing an issue?

Santiago Pastorino (Jul 01 2020 at 21:55, on Zulip):

:+1:

Santiago Pastorino (Jul 01 2020 at 21:56, on Zulip):

just in case, what I'm usually doing is removing the FCP label, adding the accepted label and closing the issue

Santiago Pastorino (Jul 01 2020 at 21:56, on Zulip):

just in case you were expecting something different to happen

simulacrum (Jul 01 2020 at 21:58, on Zulip):

yeah my guess is we're just not triggering on the right event, we should only check if the accepted label has been added

simulacrum (Jul 01 2020 at 21:59, on Zulip):

but we're probably just triggering on all issue events that touch issues with accepted proposals

Santiago Pastorino (Jul 02 2020 at 14:04, on Zulip):

triagebot#666

Santiago Pastorino (Jul 02 2020 at 14:04, on Zulip):

666 :O

lcnr (Jul 08 2020 at 10:57, on Zulip):

The implementation of this MCP changed slightly in https://github.com/rust-lang/rust/pull/74113.

While the general idea is still the same, most queries are now split into query_name and query_name_of_const_arg, due issues with incremental compilation. What's the correct procedure here? cc @nikomatsakis on how smallish design changes like this should be handled

nikomatsakis (Jul 09 2020 at 16:58, on Zulip):

@lcnr I think that's fine, the MCP is an initial proposal

nikomatsakis (Jul 09 2020 at 16:58, on Zulip):

it can evolve

Last update: May 07 2021 at 06:30UTC