James Hawkins wrote:
> On Sat, Aug 2, 2008 at 4:09 PM, Andrew Talbot
> <[EMAIL PROTECTED]> wrote:
>> Changelog:
>> fusion: Use proper function pointer.
>>
>> diff --git a/dlls/fusion/fusion.c b/dlls/fusion/fusion.c
>> index ac01cf4..637346c 100644
>> --- a/dlls/fusion/fusion.c
>> +++ b/dlls/fusion/fusion.c
>> @@ -32,6 +32,9 @@
>>
>> WINE_DEFAULT_DEBUG_CHANNEL(fusion);
>>
>> +typedef HRESULT (WINAPI *PFNGETCORVERSION)(LPWSTR pbuffer, DWORD
>> cchBuffer,
>> + DWORD *dwLength);
>> +
>> /******************************************************************
>> * ClearDownloadCache (FUSION.@)
>> */
>> @@ -89,20 +92,18 @@ HRESULT WINAPI GetAssemblyIdentityFromFile(LPCWSTR
>> pwzFilePath, REFIID riid,
>> return E_NOTIMPL;
>> }
>>
>> -static HRESULT (WINAPI *pGetCORVersion)(LPWSTR pbuffer, DWORD cchBuffer,
>> - DWORD *dwLength);
>> -
>
> There's nothing wrong with this. We use this type of function pointer
> all over the code base.
>
Hi James,
I think the part of my patch you have copied excludes the crucial bit (please
excuse the line-wrapping).
- pGetCORVersion = (void *)GetProcAddress(hmscoree, "GetCORVersion");
+ pGetCORVersion = (PFNGETCORVERSION)GetProcAddress(hmscoree,
"GetCORVersion");
The thing that I would argue is wrong is that we are using an object pointer
(void*)
in place of a function pointer
<return type> (*)(<args>)
Granted there are probably around two thousand other instances like this in the
dlls, and some may argue that it works, so leave it. On the other hand, there
are also quite a lot of type definitions for *PFN...s and several
MAKE_FUNCPTR() macros around, too.
I seem to remember that someone tried to compile Wine with Visual Studio or
suchlike, recently, and had to suppress warnings about using object pointers
instead of function pointers (amongst other things).
Anyway, I just fly the odd flag in good faith, occasionally, and see whether it
gets saluted or shot down. :)
Thanks and kind regards,
--
Andy.
