Stream: rustdoc

Topic: Notifications for issues to rustdoc team


view this post on Zulip apiraino (Jun 11 2021 at 08:06):

Hi, I'm still confused on what happens on the rust lang issue tracker when one of the *-rustdoc-* labels is applied. I know for sure that the I-nominated issues end up with a new topic in this stream (as [ can understand reading here).

view this post on Zulip apiraino (Jun 11 2021 at 08:06):

but what happens with any of the other labels?

view this post on Zulip apiraino (Jun 11 2021 at 08:08):

Example, were you notified of this issue, correct?
https://github.com/rust-lang/rust/issues/86208

view this post on Zulip GuillaumeGomez (Jun 11 2021 at 09:36):

So, unless I misunderstood, we kinda agreed on removing A-rustdoc and use T-rustdoc. When the team is needed, there is I-needs-decision and I-nominated.

view this post on Zulip GuillaumeGomez (Jun 11 2021 at 09:37):

You can see it in the last messages from https://rust-lang.zulipchat.com/#narrow/stream/266220-rustdoc/topic/A-rustdoc.20vs.20T-rustdoc

view this post on Zulip GuillaumeGomez (Jun 11 2021 at 09:37):

I think at this point, we simply need someone to remove the label

view this post on Zulip inquisitivecrystal (Jun 12 2021 at 01:07):

I think this is a different problem.

view this post on Zulip inquisitivecrystal (Jun 12 2021 at 01:09):

This isn't about A-rustdoc vs T-rustdoc. Rather, if I understand the situation correctly, wg-priortization is asking you how we should handle prioritizing rustdoc issues.

view this post on Zulip inquisitivecrystal (Jun 12 2021 at 01:12):

We get notified of all I-prioritize issues in rust-lang/rust. If the issue is in t-compiler's area, it's our responsibility to figure out what to do with it. But if it relates to rustdoc, we should be handling it in a way that works well for y'all.

There are a few ways we could handle I-priortize rustdoc issues. We could just ignore them, which would work well if y'all want to prioritize the issues yourselves. We could also assign the issue a priority tag, ping y'all on the issue, or both.

We may be in a slightly better position to do something than you are, given that we get notifications. Indeed, in the past, some members if wg-priortization had found that I-prioritize rustdoc issues often got lost, and we want to know if there's a way we can make sure that doesn't happen. That's why we're asking what we should do with these issues.

view this post on Zulip inquisitivecrystal (Jun 12 2021 at 01:14):

Previously someone had talked to y'all and gotten the impression that, if there was a rustdoc regression, you wanted wg-priortization to tag it as P-high and ping y'all. If that's what you want us to do, I think we're happy to do it — we're just trying to figure out how you want us to address rustdoc regressions.

view this post on Zulip inquisitivecrystal (Jun 12 2021 at 01:15):

Basically, when there's a rustdoc issue that gets tagged I-prioritize, what should wg-prioritization do with it?

view this post on Zulip inquisitivecrystal (Jun 12 2021 at 01:15):

apiraino can correct me if I'm explaining this incorrectly.

view this post on Zulip Joshua Nelson (Jun 12 2021 at 01:36):

inquisitivecrystal said:

Previously someone had talked to y'all and gotten the impression that, if there was a rustdoc regression, you wanted wg-priortization to tag it as P-high and ping y'all. If that's what you want us to do, I think we're happy to do it — we're just trying to figure out how you want us to address rustdoc regressions.

:wave: "someone" is me and @Manish Goregaokar

view this post on Zulip Joshua Nelson (Jun 12 2021 at 01:36):

at the time @apiraino was planning to completely ignore P-high/P-critical issues that weren't from T-compiler and I didn't want that to happen

view this post on Zulip Joshua Nelson (Jun 12 2021 at 01:37):

I'm not tied to "default to P-high and ping the team", but I think it's better than just ignoring regressions

view this post on Zulip inquisitivecrystal (Jun 12 2021 at 02:04):

In a sensible world, I think prioritization would either be a responsibility of each team or a responsibility of t-release (presumably through the triage WG). But we don't actually live in that word; wg-prioritization has a lot of specialized experience with prioritization decisons that the triage WG currently doesn't, and is in probably the best position to route prioritization for the entire project. (This is based purely on backreading and could be dead wrong; I've only been prioritizing issues for, like, a week, and shouldn't be treated as having any authority whatsoever on the matter.)

view this post on Zulip apiraino (Jun 12 2021 at 09:20):

As per GuillaumeGomez comment, I think we can go back to labelling (in case it's missing) with T-rustdocand that should be enough to get the attention from folks here. I-needs-decision and/or I-nominated to raise a flag that the issue needs a discussion.

view this post on Zulip apiraino (Jun 12 2021 at 09:20):

Agree with Joshua Nelson that we don't want to ignore issues :) and we'll keep an eye trying to avoid anything slipping the team's eye (the team is usually pretty reacting fast anyway)

view this post on Zulip apiraino (Jun 12 2021 at 09:20):

Details can then be judged on a per case basis. My main goal is to keep this lean and easy for rustdoc and avoid additional work/churn on them


Last updated: Oct 11 2021 at 22:34 UTC