library: use core::ffi::c_* types

[xtqqczze]

No description provided.

[rustbot]

Some changes occurred in std_detect

cc @Amanieu, @folkertdev, @sayantn

compiler-builtins is developed in its own repository. If possible, consider making this change to rust-lang/compiler-builtins instead.

cc @tgross35

[rustbot]

r? @nia-e

rustbot has assigned @nia-e.
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: @ChrisDenton, libs
• @ChrisDenton, libs expanded to 13 candidates
• Random selection from 7 candidates

[xtqqczze]

r? @tgross35

[rustbot]

tgross35 is currently at their maximum review capacity.
They may take a while to respond.

[tgross35]

Most of this seems completely reasonable. However, I don't think we should change imports from std to core in examples. E.g.:

+ /// use core::ffi::c_char;
  /// use std::ffi::CString;
- /// use std::os::raw::c_char;

It's a bit cleaner to import only from std::ffi than to mix core and std.

View changes since this review

[rustbot]

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

[madsmtm]

I'm greatly in favor of using std::ffi::* instead of std::os::raw::*, but I'm not convinced that using core in documentation is the best idea? I'd prefer to guide users towards using the std re-exports.

Especially because core isn't available in the 2015 edition, so users copying the code into a sample file and running bare rustc on it will get an error. (Somewhat mitigated by rust-lang/compiler-team#1019, but still).

View changes since the review

[xtqqczze]

Especially because core isn't available in the 2015 edition, so users copying the code into a sample file and running bare rustc on it will get an error.

Is this really a significant issue? If someone compiles the example with the 2015 edition, the compiler already shows a suggested fix:

error[E0433]: cannot find `core` in the crate root
 --> <source>:1:5
  |
1 | use core::ffi::CStr;
  |     ^^^^ you might be missing crate `core`
  |
help: try using `std` instead of `core`
  |
1 - use core::ffi::CStr;
1 + use std::ffi::CStr;
  |

error: aborting due to 1 previous error

For more information about this error, try `rustc --explain E0433`.

[tgross35]

Ditto on that #159868 (review). It's also just consistency; if you have std available it's more common to import everything from std than to split it up.

[estebank]

I think it would be reasonable to customize the resolve error under 2015 to mention the difference in a note.

[rust-bors]

☔ The latest upstream changes (presumably #160517) made this pull request unmergeable. Please resolve the merge conflicts by rebasing.