Re: shell32: Fix cast in RecycleBinMenu_InvokeCommand

2011-07-14 Thread Jay Yang 
On 07/14/2011 09:50 AM, Alexandre Julliard wrote:
> Jay Yang  writes:
>
>> @@ -240,7 +240,7 @@ static HRESULT WINAPI 
>> RecycleBinMenu_InvokeCommand(IContextMenu2 *iface,
>>  LPCSTR verb = pici->lpVerb;
>>  if(!HIWORD(verb))
>>  {
>> -switch((UINT)verb)
>> +switch((UINT_PTR)verb)
> Actually the HIWORD() is more problematic.
>
What's the best way to do this then. Since the msdn documentation on this 
assumes that pointers are
32 bit, which is obviously wrong. Should I do something like 
"if((verb&0x)==verb)"?

--
Jay Yang

Re: shell32: Make ISFHelper_fnCopyItems actually copy the items (try 2)

2011-07-14 Thread Alexandre Julliard 
Jay Yang  writes:

> Removed an unused variable.

Still failing ("ret may be used uninitialized").

-- 
Alexandre Julliard
julli...@winehq.org

Re: shell32: Fix cast in RecycleBinMenu_InvokeCommand

2011-07-14 Thread Alexandre Julliard 
Jay Yang  writes:

> @@ -240,7 +240,7 @@ static HRESULT WINAPI 
> RecycleBinMenu_InvokeCommand(IContextMenu2 *iface,
>  LPCSTR verb = pici->lpVerb;
>  if(!HIWORD(verb))
>  {
> -switch((UINT)verb)
> +switch((UINT_PTR)verb)

Actually the HIWORD() is more problematic.

-- 
Alexandre Julliard
julli...@winehq.org

Re: [PATCH 1/2] d3dx9_36: Added support for structs and arrays to ID3DXConstantTable (Try 2)

2011-07-14 Thread Matteo Bruni 
2011/7/14 Travis Athougies :
> Needed for Assassin's Creed Brotherhood and Dead Space 2.
>
> Changes from last patch:
>    - New way of distinguishing handles from strings
>    - Support for parent constants in ID3DXConstantTable::GetConstantByName 
> and ID3DXConstantTable::GetConstant
>    - Implemented ID3DXContantTable::GetConstantElement
>

You should split the reworking of D3DXHANDLEs into a separate patch.
GetConstantElement could be on its own patch too, probably.

Re: Bad flag check

2011-07-14 Thread Andrew Eikum 

On 07/14/2011 05:13 AM, Francois Gouget wrote:


One of my scripts noticed that the following code does not make sense
in dlls/winmm/waveform.c:

static DWORD WINMM_NotifyClient(WINMM_CBInfo *info, WORD msg,
DWORD_PTR param1, DWORD_PTR param2)
{
if(info->flags&  DCB_NULL)
 return MMSYSERR_NOERROR;

That's because DCB_NULL is 0 so this condition is always false. What was
the intent?




The intent is not to perform a callback if the callback type is 
DCB_NULL. But that check is wrong. It should be like "(info->flags & 
DCB_TYPEMASK) == DCB_NULL". I'll put together a patch to send off after 
today's updates.


Thanks,
Andrew

Re: GSoC-2011: Implement Missing Mesh Functions in Wine’s D3DX9

2011-07-14 Thread Michael Mc Donnell 
Hi Alexandre

Could I get you to commit my ConvertAdjacencyToPointReps patches? They
have ID 76054 and 76055 on http://source.winehq.org/patches/

Dylan Smith thinks they look ok.

Thanks,
Michael Mc Donnell

Bad flag check

2011-07-14 Thread Francois Gouget 

One of my scripts noticed that the following code does not make sense 
in dlls/winmm/waveform.c:

static DWORD WINMM_NotifyClient(WINMM_CBInfo *info, WORD msg,
   DWORD_PTR param1, DWORD_PTR param2)
{
   if(info->flags & DCB_NULL)
return MMSYSERR_NOERROR;

That's because DCB_NULL is 0 so this condition is always false. What was 
the intent?


-- 
Francois Gouget   http://fgouget.free.fr/
   RFC 2549: ftp://ftp.isi.edu/in-notes/rfc2549.txt
IP over Avian Carriers with Quality of Service

Re: Visual Studio project to compile the d3d code

2011-07-14 Thread Michael Mc Donnell 
On Wed, Jul 13, 2011 at 5:07 PM, Stefan Dösinger  wrote:
> On Wednesday 13 July 2011 11:15:47 Michael Mc Donnell wrote:
>> Now I'm getting error messages about "Wine should not include
>> windows.h internally" and a lot about "redefinition; different
>> linkage". I tried rebuilding but that didn't help. Any guesses?
> The Wine include directories should have a higher priority than the Platform
> SDk and DirectX SDK dirs.

It seems like the Wine include directories all come before the
Platform SDK and DirectX SDK dirs. Did you spot anything?