bjoernager/rust
bjoernager/rust
1
Fork 0
Code Activity

Rollup merge of #128735 - jieyouxu:pr-120176-revive, r=cjgillot

Browse source 
Add a special case for `CStr`/`CString` in the `improper_ctypes` lint

Revives #120176. Just needed to bless a test and fix an argument, but seemed reasonable to me otherwise.

Instead of saying to "consider adding a `#[repr(C)]` or `#[repr(transparent)]` attribute to this struct", we now tell users to "Use `*const ffi::c_char` instead, and pass the value from `CStr::as_ptr()`" when the type involved is a `CStr` or a `CString`.

The suggestion is not made for `&mut CString` or `*mut CString`.

r? ``````@cjgillot`````` (since you were the reviewer of the original PR #120176, but feel free to reroll)
This commit is contained in:
Trevor Gross 2024-08-24 21:03:31 -05:00 • committed by GitHub
commit 198a68df1c
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
7 changed files with 172 additions and 21 deletions
Download patch file Download diff file Expand all files Collapse all files

5
compiler/rustc_lint/messages.ftl
View file

@ -361,6 +361,11 @@ lint_improper_ctypes_box = box cannot be represented as a single pointer
lint_improper_ctypes_char_help = consider using `u32` or `libc::wchar_t` instead
lint_improper_ctypes_char_reason = the `char` type has no C equivalent
lint_improper_ctypes_cstr_help =
consider passing a `*const std::ffi::c_char` instead, and use `CStr::as_ptr()`
lint_improper_ctypes_cstr_reason = `CStr`/`CString` do not have a guaranteed layout
lint_improper_ctypes_dyn = trait objects have no C equivalent
lint_improper_ctypes_enum_repr_help =