|
Yes, this works! Why, however this doesn't:
struct s2 : POINT
{
int b;
};
struct s3
{
int c;
};
__asm mov eax, [eax]s3.c.b.x // 'x' : ambiguous member name in 'second operand'
'x' couldn't be a reserved name, could it??
Is there any way to clarify it to compiler except for defining an offset explicitly, like:
enum { s3s2x = offsetof(s3, s2.x) };
and then
__asm mov eax, [eax + s3s2x]
|
|
|
|
|
How are you expecting to get to 'b' or 'x' through struct s3 since struct s3 and struct s2 know nothing of each other?
__asm mov eax, [eax]s3.c.b.x
^ this dot makes no sense
"The pointy end goes in the other man." - Antonio Banderas (Zorro, 1998)
|
|
|
|
|
Sorry, of course:
struct s2 : POINT
{
int b;
};
struct s3
{
s2 c;
};
__asm mov [eax]s3.c.b.x // error
__asm mov [eax]s3.c.x // error also
And, unfortunately, if s2 has no data members,
that is
struct s2 : POINT
{
void foo();
};
then I wouldn't know how to access POINT::x at all as
__asm mov [eax]s3.c.x
results in error again...
|
|
|
|
|
See if this helps:
int xx = ss3.c.x;
__asm mov eax, [eax][xx]
"The pointy end goes in the other man." - Antonio Banderas (Zorro, 1998)
|
|
|
|
|
Well,
enum
{
s3cx = offsetof(s3, c.x)
};
__asm mov eax, [eax + s3cx]
is better as it requires no additional code, but the point was to access it from within an __asm block with no helpers...
Take a look at this:
struct POINT3
{
int x;
int y;
int z;
};
struct s2 : POINT3
{
int b;
};
struct s3
{
s2 c;
};
Now,
__asm mov eax, [eax]s3.c.b.z // works fine
but
__asm mov eax, [eax]s3.c.b.x // doesn't!!
That's why I asked if 'x' was reserved or smth...
However, if s2 is
struct s2 : POINT3
{
void foo();
};
then
__asm mov eax, [eax]s3.c.z // fails too...
Any suggestions will be greatly appreciated and thanks for your input!
|
|
|
|
|
jk61 wrote:
Now,
__asm mov eax, [eax]s3.c.b.z // works fine
Why this does not yield a compiler error is odd, as 'b' cannot be resolved into anything further as it is an int type. Does that make sense? When you have . and -> notation, it means that the object (i.e., structs and classes) can be broken down even further.
"The pointy end goes in the other man." - Antonio Banderas (Zorro, 1998)
|
|
|
|
|
Yes, it would be odd if it was C++ code, but for inline asm it's sort of normal, I suppose.
You suggested that syntax yourself, right? It didn't even cross my mind to try to access 'x' through 'b'. Do you have a ms vc compiler there to try it out?
|
|
|
|
|
jk61 wrote:
You suggested that syntax yourself, right?
Indeed I did. Shame on me!
"The pointy end goes in the other man." - Antonio Banderas (Zorro, 1998)
|
|
|
|
|
I give up... That strange access, while compiles ok, results, in fact, in wrong offsets... So, it doesn't work after all and I go with enums
|
|
|
|
|
Hi,
I'm getting the following error whenever I try to build a 'debug' version of any application.
LINK : fatal error LNK1209 program database "C:\test\Debug\test.pdb" differs from previous link; relink or rebuild
Error executing link.exe.
I have tried re-building but it doesn't solve the problem. I even re-installed VC++ but that also didn't help. This is not specific to an application since it occurs even when i build a simple AppWizard application.
Any idea how to solve this problem would be greatly appreciated.
Thanks,
Mel
|
|
|
|
|
Does it happen every time? Can you manually delete the .pdb file and re-build?
"The pointy end goes in the other man." - Antonio Banderas (Zorro, 1998)
|
|
|
|
|
Yes, it happens every time and for any application.
Yes, I can delete the .pdb files, but rebuilding doesn't solve the problem.
First time i've experienced this kind of problem in my 5+ years of using VC++
-Mel
|
|
|
|
|
You did reapply the service packs for VS60, I take it...
Steve S
|
|
|
|
|
Yes, I just applied Service Pack 5....but still can't build debug configurations
-Mel
|
|
|
|
|
Hello everyone,
I used the following code to implement the tooltip for a toolbar. But the tooltip won't be displayed when the toolbar window is not active. I know that I can set TTS_ALWAYSTIP to solve this problem. But I don't know how to set it. There is no CToolTipCtrl in my application.
Thanks a lot in advance.
BEGIN_MESSAGE_MAP(CMyView, CView)
...
ON_NOTIFY_EX_RANGE(TTN_NEEDTEXTW, 0, 0xFFFF, OnToolTipNotify)
ON_NOTIFY_EX_RANGE(TTN_NEEDTEXTA, 0, 0xFFFF, OnToolTipNotify)
END_MESSAGE_MAP()
void CMyView::OnInitialUpdate()
{
CView::OnInitialUpdate();
CEdit* pEdit = new CEdit;
pEdit->Create(ES_MULTILINE | WS_CHILD | WS_VISIBLE | WS_TABSTOP | WS_BORDER,
CRect(10, 10, 100, 100), this, 111);
EnableToolTips(TRUE); // enable tool tips for view
}
//Notification handler
BOOL CMyView::OnToolTipNotify(UINT id, NMHDR *pNMHDR,
LRESULT *pResult)
{
// need to handle both ANSI and UNICODE versions of the message
TOOLTIPTEXTA* pTTTA = (TOOLTIPTEXTA*)pNMHDR;
TOOLTIPTEXTW* pTTTW = (TOOLTIPTEXTW*)pNMHDR;
CString strTipText;
UINT nID = pNMHDR->idFrom;
if (pNMHDR->code == TTN_NEEDTEXTA && (pTTTA->uFlags & TTF_IDISHWND) ||
pNMHDR->code == TTN_NEEDTEXTW && (pTTTW->uFlags & TTF_IDISHWND))
{
// idFrom is actually the HWND of the tool
nID = ::GetDlgCtrlID((HWND)nID);
}
if (nID != 0) // will be zero on a separator
strTipText.Format("Control ID = %d", nID);
if (pNMHDR->code == TTN_NEEDTEXTA)
lstrcpyn(pTTTA->szText, strTipText, sizeof(pTTTA->szText));
else
_mbstowcsz(pTTTW->szText, strTipText, sizeof(pTTTW->szText));
*pResult = 0;
return TRUE; // message was handled
}
Thanks
Bin
|
|
|
|
|
I might be misunderstanding the workings of a VARIANT , but I am trying to pack a SAFEARRAY and assign it to my VARIANT . When I place a break point I can see I am getting data into the array but when I try to debug the component in VB I can only get 1 element back which is 0. Does anyone see something wrong with this, this is the first time I have tried using a SAFEARRAY so bare with me. Thanks in advance.
STDMETHODIMP CMoses::GetNumbers(VARIANT *pVal)
{
AFX_MANAGE_STATE(AfxGetStaticModuleState())
HRESULT h;
long lDimension[2];
int startyear, runyear;
startyear = 3;
runyear = 68;
typedef vector<double> collection;
collection numbers;
for(int i = startyear * 12; i < runyear * 12; i += 12)
{
IColumn col(root.Column("nsv_e_0"));
if(col != NULL)
{
CString num;
num.Format("%d", i);
numbers.push_back(col.value(num));
}
}
SAFEARRAYBOUND bounds[1];
SAFEARRAY *psa = NULL;
bounds[0].lLbound = 1;
bounds[0].cElements = numbers.size();
psa = SafeArrayCreate(VT_R8, 1, bounds);
if (psa == NULL) throw ERROR_NOT_ENOUGH_MEMORY;
int j = 0;
vector<double>::iterator it;
for(it = numbers.begin(); it != numbers.end(); it++, j++)
{
lDimension[1] = j;
lDimension[0] = 0;
h = ::SafeArrayPutElement(psa, lDimension, &it);
}
pVal->parray = psa;
return S_OK;
}
- Nick Parker My Blog
|
|
|
|
|
Try
pVal->vt = VT_ARRAY|VT_R8;
just before your 'return S_OK'.
You're setting the VARIANT value, but not it's type.
Steve S
|
|
|
|
|
Steve S wrote:
You're setting the VARIANT value, but not it's type.
Gotcha, that makes sense. Now VB is returning a list of 0's. Not sure why I am not getting the numbers. I threw in a ::AfxMessageBox(...) to check that I am getting a value from the iterator when it is dereferenced, which it is. Am I following the correct syntax by assignment the address of the iterator to the SAFEARRAY ? Thanks again Steve.
[EDIT] I solved the problem, I had the lDimension flip-flopped, thanks again for your help. [/EDIT]
- Nick Parker My Blog
|
|
|
|
|
I am trying to write an application in C++ to implement privacy control. For that I need to know how to read a cookie. If someone knows, please let me know
Thanks,
subbu
|
|
|
|
|
Refer to MSDN articles Q241492 and Q240191.
"The pointy end goes in the other man." - Antonio Banderas (Zorro, 1998)
|
|
|
|
|
Yes, but I must use ATL COM...and without?
I should make an application .exe.
|
|
|
|
|
The code shown in article Q241492 (it's just COM) can be used with or without ATL.
"The pointy end goes in the other man." - Antonio Banderas (Zorro, 1998)
|
|
|
|
|
there are 2 buffers (size= 1 MB) in my app. one buffer is in video memory (get by DirectX), another one is in system memory (by new).
if copy from system buffer -> video buffer, time=15 ms
if copy from video buffer -> system buffer, time=180 ms
my question is:
why is it so slow from video to system memory? how to solve the problem?
thanks
includeh10
|
|
|
|
|
It's very common for reads from the video buffer to be slow. The details of why depends on your specific hardware and directx driver. In general terms the video buffer is not just another block of RAM; after you are done writing to it, it may not contain exactly what you put in yet (or ever). DirectX tries to hide these details, but when you read from video memory it has to make everything synch up so it behaves like you expected it to.
The most common workaround is to rework your application so it does not have to read from the video buffer. It's seldom absolutely necessary.
|
|
|
|
|
Hi antlers,
what i need is just "read from video memory". i don't need 'write to video memory'.
it can be assumed as a screen capture.
my process is
1. use surface->Lock to get video buffer
2. copy the buffer to system memory
u say 'it does not have to read from the video buffer', could u give me some details for re-working my app?
(the app is in realtime, speed is important, so i use DirectX)
thanks.
includeh10
|
|
|
|
|