Enhance drop forget ref hint

[Kcang-gna]

Replace let _ = ... with let _ = drop_arg
Changed some hint messages

Fixes: #155895

[rustbot]

r? @TaKO8Ki

rustbot has assigned @TaKO8Ki.
They will have a look at your PR within the next two weeks and either review your PR or reassign to another reviewer.

Use r? to explicitly pick a reviewer

Why was this reviewer chosen?

The reviewer was selected based on:

• Owners of files modified in this PR: compiler
• compiler expanded to 73 candidates
• Random selection from 18 candidates

[Urgau]

As I said in #155895 (comment), I don't think we should do that, being a bit vague was on purpose.

Further-more the suggestion is not right when there is an explicit borrow, like in https://github.com/rust-lang/rust/pull/156395/changes#diff-caa7b472da1b6aa2b42ea519f4bdd7eef6c5c031a32a342110116db9b83dde97R14.

[Kcang-gna]

Why is let _ = &x regarded as incorrect?

I don’t quite understand from what angle this way of writing is considered incorrect. From my perspective, this syntax either brings about side effects or is used to disable the "unused variable" lint, and its effect seems roughly equivalent to drop(&x).

However, developers may mistakenly assume that drop(&x)has some practical effect. It would be better to directly prompt developers to remove the drop call through code hints, along with prompt messages. This can better eliminate developers’ misunderstandings, make them realize this approach does not work, and that they need to adopt another method to achieve their goal.

Therefore in my opinion, if drop(&x); is highly discouraged, then let _ = &x; can only be classified as not recommended.What’s more, it is simply to disable the "unused variable" lint.

(Translated by LLM)

[Urgau]

Why is let _ = &x regarded as incorrect?

Same as drop(&x), it does nothing.

If we are at a point where we are suggesting something to the user, it would be better to suggest code that actually does something.

[Kcang-gna]

Ok, I got it

Maybe we can give the user links to primitive reference and std::mem::drop as a note, because users only try to drop references due to a lack of relevant knowledge. keep let _ = ... unchanged

[Urgau]

Sure, if you think linking to std::mem::drop improves things I'm fine with that.

I'm not sure we add a link to the reference type. If the goal was to make it clear that a reference is Copy, I think it would be better to update std::mem::drop documentation to make that clearer.

[rustbot]

This PR was rebased onto a different main commit. Here's a range-diff highlighting what actually changed.

Rebasing is a normal part of keeping PRs up to date, so no action is needed—this note is just to help reviewers.