fix [std_instead_of_core]: false positive for multi-imports

[bushrat011899]

Objective

• Fix false positive identified in this comment on fix [std_instead_of_core]: false positives for core::io/MSRV #16964.
• Fixes std_instead_of_core creates an invalid fix on stacked imports #12468
• Addresses this comment

Solution

The current version of this lint can improperly trigger a suggestion for certain complex use statements. This PR attempts to resolve this by tracking ItemKind::Use items, which allows determining when paths are used within the context of a multi-import (UseKind::ListStem). Lints can then be appropriately grouped before determining if a suggestion is possible.

This PR is the first commit from a re-write branch #17364, which acts on this suggestion and attempts to resolve many currently open issues for this lint.

One consequence of this PR is that CrateNum is used to determine what crate items are being imported from, instead of the symbol of the first segment in the path. This resolves an unreported issue where doing something like extern crate core as std; would cause std_instead_of_core to emit erroneously. It also caused cases like extern crate std as core; to not emit std_instead_of_core.

Unfortunately, this does lead to an issue for extern crate std as alloc; (which is used e.g. here), since the suggestion will be to replace alloc with alloc. This is fixed in #17364 since it also fixes #15836 & #16695 by checking what name alloc is available under, and demotes the suggestion to a help if it isn't.

---

Notes

• No AI tooling of any kind was used during the creation of this PR.
• Originally a part of fix [std_instead_of_core]: false positives for core::io/MSRV #16964, but has been split off for ease of review. Either PR can be merged first.

---

Please write a short comment explaining your change (or "none" for internal only changes)

changelog: [std_instead_of_core], [std_instead_of_alloc], [alloc_instead_of_core]: fix certain false positives for mulit-imports.
changelog: [std_instead_of_core], [std_instead_of_alloc], [alloc_instead_of_core]: lint cases where std or alloc are renamed.

[rustbot]

Thanks for the pull request, and welcome! The Rust team is excited to review your changes, and you should hear from @llogiq (or someone else) some time within the next two weeks.

Please see the contribution instructions for more information. Namely, in order to ensure the minimum review times lag, PR authors and assigned reviewers should ensure that the review label (S-waiting-on-review and S-waiting-on-author) stays updated, invoking these commands when appropriate:

• @rustbot author: the review is finished, PR author should check the comments and take action accordingly
• @rustbot review: the author is ready for a review, this PR will be queued again in the reviewer's queue

Why was this reviewer chosen?

The reviewer was selected based on:

• Owners of files modified in this PR: 8 candidates
• 8 candidates expanded to 8 candidates
• Random selection from Jarcho, dswij, llogiq, samueltardieu

[github-actions]

Lintcheck changes for 6adb667

Lint	Added	Removed	Changed
clippy::alloc_instead_of_core	3	1	1
clippy::std_instead_of_alloc	7	2	15
clippy::std_instead_of_core	11	40	26

This comment will be updated if you push new changes

[rustbot]

Reminder, once the PR becomes ready for a review, use @rustbot ready.

[Jarcho]

r? Jarcho

[bushrat011899]

A few comments that might be relevant for reviewers!

View changes since this review

[Jarcho]

This lint really needs to be re-written to work on check_item for use paths instead. Doing so will let you use UseKind::ListStem to know when braces were used and rebuild the use tree which you need to handle more complex cases.

You;ll want a setup like:

fn check_item(&mut self, cx: &LateContext<'tcx>, i: &'tcx Item<'_>) {
  if let ItemKind::Use(p, kind) = i.kind {
    if p.segments.len() > 2 {
      self.skip_paths += u8::from(p.res.value_ns.is_some()) + u8::from(p.res.type_ns.is_some()) + u8::from(p.res.mac_ns.is_some());
    }
    // Continue linting
  }
}

fn check_path(&mut self, cx: &LateContext<'tcx>, p: &'tcx Path<'_>) {
  if let [first, mid @ .., last] = path.segments {
    if let Some(count) = self.skip_paths.checked_sub(1) {
      self.skip_paths = count;
      return;
    }
    // Continue linting
  }
}

Note that you want to do the minimum amount of work possible for single segment paths since they are by far the most common and trivially not lintable.

---

If you want a fun case you can do something like:

use {
  some_crate::foo,
  ::{
    std::{fmt::{*, Formatter}, fs, sync::atomic},
    core::u32,
    some_crate::bar,
  },
  ::alloc::fmt::Result as FmtResult,
};

[bushrat011899]

That's a good idea, thanks! I'm going to play with it and see what I can come up with.

@rustbot author

[rustbot]

This PR was rebased onto a different master 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.

[bushrat011899]

@rustbot ready

Thanks for the suggestion Jarcho! I've done a larger re-write fixing multiple issues with this lint in #17364, but I've structured it so the multi-import fix can be pulled out into this PR for the sake of review. I've tried to structure my rewrite to match the original implementation as much as possible to make the diff easier to review, but it's probably easier to just read the new implementation as if it's new.