Update to ClangSharp 16.0 #1601
Assignees
Labels
enhancement
New feature or request
Comments
|
|
@tannergooding I'm working on updating to ClangSharp 16.0. Above is the diff before I try to add the new bitfield parameter. Could you take a look? It looks like |
|
@tannergooding these enums all seem to have a FORCE_DWORD or FORCE_UINT member which means they should be scraped as UInt32 as with earlier versions right? |
That's how ClangSharp set itself up initially, but it was actually incorrect. Despite the name, MSVC treats these enums as You can see this difference in godbolt here: typedef enum D2D1_ALPHA_MODE {
D2D1_ALPHA_MODE_UNKNOWN = 0,
D2D1_ALPHA_MODE_PREMULTIPLIED = 1,
D2D1_ALPHA_MODE_STRAIGHT = 2,
D2D1_ALPHA_MODE_IGNORE = 3,
D2D1_ALPHA_MODE_FORCE_DWORD = 0xffffffff
};
bool isNegative() { return D2D1_ALPHA_MODE_FORCE_DWORD < 0; }
typedef enum D2D1_ALPHA_MODE_RETYPED : unsigned {
D2D1_ALPHA_MODE_RETYPED_UNKNOWN = 0,
D2D1_ALPHA_MODE_RETYPED_PREMULTIPLIED = 1,
D2D1_ALPHA_MODE_RETYPED_STRAIGHT = 2,
D2D1_ALPHA_MODE_RETYPED_IGNORE = 3,
D2D1_ALPHA_MODE_RETYPED_FORCE_DWORD = 0xffffffff
};
bool isNegativeRetyped() { return D2D1_ALPHA_MODE_RETYPED_FORCE_DWORD < 0; }Which is to say, MSVC never retypes the enum implicitly. It requires you to explicitly opt into a type change via the dedicated language syntax. |
|
Ok. So all is expected. Thanks! |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Unblocks #1392.
The text was updated successfully, but these errors were encountered: