Commit d926ddc
doc: add missing float32/float64 FFI type names
`ToFFIType()` accepts `float32` and `float64` as aliases for `float`
and `double`, and both names already appear in the `ffi.types`
constants list further down the same page, but they were missing from
the list of supported type names.
Also split `char` onto its own line. Unlike `u8`, `uint8` and `bool`,
which always map to `ffi_type_uint8`, `char` maps to either
`ffi_type_sint8` or `ffi_type_uint8` depending on the platform C ABI,
as the paragraph below the list already explains.
Signed-off-by: Soul Lee <alus20x@gmail.com>
PR-URL: #64874
Refs: #62892
Refs: #64848
Reviewed-By: René <contact.9a5d6388@renegade334.me.uk>
Reviewed-By: Paolo Insogna <paolo@cowtech.it>
Reviewed-By: Ulises Gascón <ulisesgascongonzalez@gmail.com>1 parent 5e433fd commit d926ddc
1 file changed
Lines changed: 4 additions & 3 deletions
| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
@@ -62,16 +62,17 @@ FFI signatures use string type names. | |||
| 62 | 62 | Supported type names: | |
| 63 | 63 | ||
| 64 | 64 | * `void` | |
| 65 | + * `char` | ||
| 65 | 66 | * `i8`, `int8` | |
| 66 | - * `u8`, `uint8`, `bool`, `char` | ||
| 67 | + * `u8`, `uint8`, `bool` | ||
| 67 | 68 | * `i16`, `int16` | |
| 68 | 69 | * `u16`, `uint16` | |
| 69 | 70 | * `i32`, `int32` | |
| 70 | 71 | * `u32`, `uint32` | |
| 71 | 72 | * `i64`, `int64` | |
| 72 | 73 | * `u64`, `uint64` | |
| 73 | - * `f32`, `float` | ||
| 74 | - * `f64`, `double` | ||
| 74 | + * `f32`, `float`, `float32` | ||
| 75 | + * `f64`, `double`, `float64` | ||
| 75 | 76 | * `pointer`, `ptr` | |
| 76 | 77 | * `string`, `str` | |
| 77 | 78 | * `buffer` | |
0 commit comments