-
Notifications
You must be signed in to change notification settings - Fork 12.9k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
attempt to clarify align_to docs #105245
attempt to clarify align_to docs #105245
Conversation
r? @thomcc (rustbot has picked a reviewer for you, use r? to override) |
Hey! It looks like you've submitted a new PR for the library teams! If this PR contains changes to any Examples of
|
Feedback from a random reader: It seems a bit confusing to be discussing "implementations". From my perspective, there is one implementation of the Standard Library, and I'm reading the docs for it. What other implementations are there? Why does this seem to be written as a generalized spec for the Standard Library API? That's not how the rest of the std docs are usually written. I'd be interested in what was found confusing in the original docs, since they seem somewhat straightforward to me. The mention of an example situation (sanitizers) where the behavior may be different seems useful, but I would just tack that on to the end of the previous version. ("For example, a sanitizer may change the behavior of this function to…") |
I agree "implementations" sounds... wonky. @ehuss The Rust community by and large has had a pattern of advising each other to avoid this function because of the weakness of the guarantees and the wording being at risk for obfuscating the reasoning why, even when it might be an appropriate function, and that's why we discussed potentially clarifying it. Also see #95851 |
The specific feedback was that "you can only rely on this guarantee for performance, not correctness" was hard to understand. Sadly I don't know why or how this is hard to understand so all I can do is throw pasta at the wall and see what sticks. 🤷 I figured it'd help to pretend as if there were multiple implementations here and one wants to write code that is correct against all of them. Maybe that's not the right take. This is not the only place where we do something like that though -- e.g. sort_unstable does not guarantee how equal elements are sorted. That's exactly the same kind of non-determinism. Somehow though that is less confusing? I'm honestly lost here, maybe @workingjubilee should make a PR since they seem to have a better idea for what even is the problem here... |
I gave this another shot, based on a proposal by Jubilee. |
I read caveats like that in (I apologize if my comment veers into "change the spec" territory. But I think it is helpful to clarify exactly what the current intended guarantees are.) |
@ehuss That's fair -- we don't really plan to change this in the future, the reason for the non-det is different here. What do you think about the new proposal? |
Sorry for the delay (real life, etc), this looks great to me. Even if I would like this to always behave correctly (I understand why this is difficult...), it's hard not to see this as a large improvement. Still, even though I am strongly in favor of this, it adds a new guarantee to this functions documentation's implementation so I can't r+ it -- it needs a libs-api sign-off. Hopefully it will be painless. @rustbot label +T-libs-api -T-libs r? @Amanieu |
LGTM! I think the new wording is much clearer than the old one. While it does technically add more guarantees, I think this is sufficiently minor to not require an FCP. @bors r+ rollup |
…iaskrgr Rollup of 10 pull requests Successful merges: - rust-lang#105216 (Remove unused GUI test) - rust-lang#105245 (attempt to clarify align_to docs) - rust-lang#105387 (Improve Rustdoc scrape-examples UI) - rust-lang#105389 (Enable profiler in dist-powerpc64le-linux) - rust-lang#105427 (Dont silently ignore rustdoc errors) - rust-lang#105442 (rustdoc: clean up docblock table CSS) - rust-lang#105443 (Move some queries and methods) - rust-lang#105455 (use the correct `Reveal` during validation) - rust-lang#105470 (Clippy: backport ICE fix before beta branch) - rust-lang#105474 (lib docs: fix typo) Failed merges: r? `@ghost` `@rustbot` modify labels: rollup
attempt to clarify align_to docs This is not intended the change the docs at all, but `@workingjubilee` said the current docs are incomprehensible to some people so this is an attempt to fix that. No idea if it helps, so -- feedback welcome. (Please let's not use this to discuss *changing* the spec. Whoever wants to change the spec should please make a separate PR for that.)
…iaskrgr Rollup of 10 pull requests Successful merges: - rust-lang#105216 (Remove unused GUI test) - rust-lang#105245 (attempt to clarify align_to docs) - rust-lang#105387 (Improve Rustdoc scrape-examples UI) - rust-lang#105389 (Enable profiler in dist-powerpc64le-linux) - rust-lang#105427 (Dont silently ignore rustdoc errors) - rust-lang#105442 (rustdoc: clean up docblock table CSS) - rust-lang#105443 (Move some queries and methods) - rust-lang#105455 (use the correct `Reveal` during validation) - rust-lang#105470 (Clippy: backport ICE fix before beta branch) - rust-lang#105474 (lib docs: fix typo) Failed merges: r? `@ghost` `@rustbot` modify labels: rollup
This is not intended the change the docs at all, but @workingjubilee said the current docs are incomprehensible to some people so this is an attempt to fix that. No idea if it helps, so -- feedback welcome.
(Please let's not use this to discuss changing the spec. Whoever wants to change the spec should please make a separate PR for that.)