A lint for potential name clashes with the Rust Prelude

[LikeLakers2]

What it does

This lint would check for potential name clashes in a crate with the Rust Prelude.

Name clashes with the Rust Standard Library are not too big of a problem - in those cases, you simply rename one import or the other. However, the Rust Prelude is special in that it is included everywhere - and its imports can be overridden.

This allows for name clashes to happen - and because Rust allows the rust prelude's imports to be overridden, it's not immediately obvious that it will cause problems.

As a real world example of this happening, see https://github.com/bevyengine/bevy/issues/14902, where the Bevy Game Engine accidentally introduced a name clash with std::ops::Drop - and nobody noticed until they tried to impl Drop for SomeStruct in the same file as importing the Bevy Prelude.

Advantage

• Avoid name clash issues similar to https://github.com/bevyengine/bevy/issues/14902
• Where a name clash is wanted or okay, the user can silence the lint - making it explicit to other users that this is intentional.

Drawbacks

• If a user is overriding a item from the Rust Prelude that they don't need, this may result in unwanted lint output.

Example

I do not imagine this lint could be automatically fixed, so this will be an example of when this lint will show, and how a user might fix it.

I imagine it should lint on code such as this:

pub struct Drop;

The lint would suggest renaming the offending struct. Afterwards, it might look like this, which should not trigger this lint:

pub struct DropEvent;

Alternatively, if a user is okay with the name clash (i.e. they don't use std::ops::Drop), they can edit their code to show that they're okay with that:

#[expect(clippy::rust_prelude_name_clash)]
pub struct Drop;

[BD103]

#[expect(clippy::rust_prelude_name_clash)]
pub struct Drop;

Is #[expect(...)] stable Rust? I didn't think it got out of nightly yet.

[LikeLakers2]

@BD103 I don't remember. If it's not, just pretend expect is replaced with allow. The point of that code block was to say "Here's what they can do to show that they acknowledge the clashing name."

[BD103]

Ok! I was hoping to use it in my own projects, which is why I was asking.