Skip to content

Conversation

@kaffarell
Copy link
Contributor

@kaffarell kaffarell commented Feb 19, 2024

The private_in_public rustc lint is now deprecated and throws a warning on each cargo check. According to the rfc 1 this lint is being substituted by three other ones: private_interfaces, private_bounds and unnameable_types.
unnameable_types is not yet stabilized 2, so it will throw a warning, but we will be ready when it is :)

Footnotes

  1. https://rust-lang.github.io/rfcs/2145-type-privacy.html

  2. https://github.com/rust-lang/rust/pull/120144

The `private_in_public` rustc lint is now deprecated and throws a
warning on each `cargo check`. According to the rfc [0] this lint is
being substituted by three other ones: `private_interfaces`,
`private_bounds` and `unnameable_types`.
`unnameable_types` is not yet stabilized [1], so it will throw a warning,
but we will be ready when it is :)

[0]: https://rust-lang.github.io/rfcs/2145-type-privacy.html
[1]: rust-lang/rust#120144
@mladedav
Copy link
Contributor

We could add #![allow(unknown_lints)] for the time being so that there are no warnings now and we can keep the lint that is yet to be stabilised.

Ideally with a TODO comment to get rid of it once the last lint is stabilized.

@kaffarell
Copy link
Contributor Author

We could... but I am not sure—this could lead to us forgetting about it :/
@hawkw what do you think?

@hawkw
Copy link
Member

hawkw commented Mar 1, 2024

This has already been fixed on master in #2814, so I've backported that commit to v0.1.x. Thanks!

Regarding adding #[allow(unknown_lints)], I think it would be a good idea to change our CI builds to pass --allow unknown_lints when running lint checks (e.g., here:

args: --all --examples --tests --benches -- -D warnings
). That way, contributors still get warnings from local runs of cargo check/cargo clippy when adding attributes referencing lists that don't exist as part of a change, but CI won't fail when a lint is removed/deprecated. That seems like a good compromise between not breaking CI on new Rust versions and warning contributors that a lint isn't valid. What do you think?

@hawkw hawkw closed this Mar 1, 2024
@kaffarell
Copy link
Contributor Author

Yes, I agree!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

3 participants