Fix #12438 std_instead_of_core regression

[MarcusGrass]

Fixes #12438.

Boy-scouting removed two paths that checks for duplication since I thought they were unused. However, that's just because I didn't spot it in the diff.

I installed difftastic and ran it on the old one:

And the new one (fixed):

New one (stderr):

Good teachings for the future when inspecting diffs with a lot of line changes, should've thought of that before, sorry for the trouble!

changelog: [std_instead_of_core] Fix false positive for crates that are in std but not core

[rustbot]

r? @y21

rustbot has assigned @y21.
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

[y21]

Is the order in which check_path is called for different resolutions in a use guaranteed? Otherwise it seems a bit fragile, since I think this would break if check_path is called with the macro resolution first, because it hasn't seen the mod res at that point.
Though fixing that likely requires reworking the lint a bit to be able to buffer the lint emissions until all parts of a use statement are checked.

Either way, that's how it was before (for a long time) and it seems this hasn't been an issue, so this LGTM. Thank you for fixing this!

@bors r+

[bors]

📌 Commit b44ab66 has been approved by y21

It is now in the queue for this repository.

[bors]

⌛ Testing commit b44ab66 with merge 7ee75f8...

[bors]

☀️ Test successful - checks-action_dev_test, checks-action_remark_test, checks-action_test
Approved by: y21
Pushing 7ee75f8 to master...

[MarcusGrass]

Is the order in which check_path is called for different resolutions in a use guaranteed? Otherwise it seems a bit fragile, since I think this would break if check_path is called with the macro resolution first, because it hasn't seen the mod res at that point. Though fixing that likely requires reworking the lint a bit to be able to buffer the lint emissions until all parts of a use statement are checked.

Either way, that's how it was before (for a long time) and it seems this hasn't been an issue, so this LGTM. Thank you for fixing this!

@bors r+

Yeah it is a bit problematic, added an issue for it here #12468