C / C++ / MFC

tolw25-May-09 1:00

tolw

25-May-09 1:00

My guess is in the first example the sizeof operator can calculate the size of the array. When it comes to pointers - no such luck. Try passing the size of the allocated memory directly (preferably by using a #define):

#define ARRAY_SIZE 1024
erg=MultiByteToWideChar(CP_ACP, 0, text, -1, w, ARRAY_SIZE); // ANSI to UNICODE
erg=WideCharToMultiByte(CP_UTF8, 0, w, -1, text, ARRAY_SIZE, 0, 0); // UNICODE

Nibu babu thomas25-May-09 2:02

Nibu babu thomas

25-May-09 2:02

Souldrift wrote:
char* text = new char[1024];
WCHAR w[1024]={0};
int erg=0;

Try to avoid using numbers directly or hard coding instead store in a constant.

const int SIZE = 1024; // Bytes
char* text = new char[SIZE];
WCHAR w[SIZE]={0};

erg=MultiByteToWideChar(CP_ACP, 0, text, -1, w, SIZE); // ANSI to UNICODE
erg=WideCharToMultiByte(CP_UTF8, 0, w, -1, text, SIZE, 0, 0); // UNICODE to UTF-8

So when you change SIZE, this code still keeps working.

Nibu babu thomas
Microsoft MVP for VC++

Code must be written to be read, not by the compiler, but by another human being.

Programming Blog: http://nibuthomas.wordpress.com

Souldrift25-May-09 2:34

Souldrift

25-May-09 2:34

Thanks. That to both of you. That works.
Problem is that I wanted to avoid a const size and instead use the variable size of m_pData (the original text). Is that possible?

Souldrift

error LNK2019: unresolved external symbol "long __cdecl _com_dispatch_method

tolw25-May-09 2:56

tolw

25-May-09 2:56

Since m_pData is a string you can use string operations like strlen. Try this:

char *text = new char[strlen( m_pData ) + 1 /*The +1 is for the End-of-String \0*/];
WCHAR *w = new WCHAR[strlen( m_pData ) + 1];
int erg=0;

strcpy(text, m_pData);

erg=MultiByteToWideChar(CP_ACP, 0, text, -1, w, strlen( text )); // ANSI to UNICODE
erg=WideCharToMultiByte(CP_UTF8, 0, w, -1, text, wcslen( w ), 0, 0); // UNICODE to UTF-8

Just make sure that the strings are all NULL terminated!!

pandit8424-May-09 23:14

pandit84

24-May-09 23:14

Hi I am using a ADO in my project. But when i compile my code i am getting the follwoing error.

error LNK2019: unresolved external symbol "long __cdecl _com_dispatch_method(struct IDispatch *,long,unsigned short,unsigned short,void *,wchar_t const *,...)" (?_com_dispatch_method@@YAJPAUIDispatch@@JGGPAXPB_WZZ) referenced in function "public: long __thiscall ConnectionEvents::InfoMessage(struct Error *,enum EventStatusEnum *,struct _Connection *)" (?InfoMessage@ConnectionEvents@@QAEJPAUError@@PAW4EventStatusEnum@@PAU_Connection@@@Z)

I have added this in my stdafx.cpp

#import "..\..\..\..\..\Microsoft\Redistribute\MSADO15.dll" no_namespace \
                  rename("Parameters","ADOParams") \
                  rename("Connection","ADOCon") \
                  rename("Parameter","ADOParam") \
                  rename("EOF", "adoEOF") \                  implementation_only

and below code in my stdafx.h file

#import "..\..\..\..\..\Microsoft\Redistribute\MSADO15.dll" no_namespace \
                  rename("Parameters","ADOParams") \
                  rename("Connection","ADOCon") \
                  rename("Parameter","ADOParam") \
                  rename("EOF", "adoEOF") \                  no_implementation

Please provide any comment i am not able to find out what is to be included to avoid this linking error.

Re: error LNK2019: unresolved external symbol "long __cdecl _com_dispatch_method

Stuart Dootson25-May-09 1:42

Stuart Dootson

25-May-09 1:42

Make sure you're linking against comsupp.lib (or some variant thereof - there's comsupp, comsuppd, comsuppw and comsuppwd) - that's where that symbol is defined. There is a #pragma in the #import process that specifies to link against that library, but it's possible, I guess, that doing the #import in stdafx negates the effect of the #pragma.

Java, Basic, who cares - it's all a bunch of tree-hugging hippy cr*p

HEAP CORRUPTION DETECTED!

susanne124-May-09 23:01

24-May-09 23:01

hallo, i have the following piece of code , when i try to delete the lpKeyPoint i get HEAP CORRUPTION DETECTED!

LPPOINT lpKeyPoint;
lpKeyPoint = (LPPOINT)new LPPOINT();
GetCaretPos((LPPOINT)lpKeyPoint);
if(lpKeyPoint)
{
delete lpKeyPoint;
lpKeyPoint = NULL;
}
How can i solve this problem?
Thanks.

CPallini24-May-09 23:19

24-May-09 23:19

susanne1 wrote:
lpKeyPoint = (LPPOINT)new LPPOINT()

That should be:

lpKeyPoint = new POINT();

If the Lord God Almighty had consulted me before embarking upon the Creation, I would have recommended something simpler.
-- Alfonso the Wise, 13th Century King of Castile.

This is going on my arrogant assumptions. You may have a superb reason why I'm completely wrong.
-- Iain Clarke

[My articles]

susanne124-May-09 23:50

24-May-09 23:50

thanks it worsk,but when i move the curosor up i get a worng value of y:

CPoint* ptrKeyPoint = new CPoint();
int nCursorLine = 0;
if((pMsg->message == WM_KEYDOWN)||(pMsg->message == WM_LBUTTONDOWN))
{
if((pMsg->wParam == VK_RETURN) || (pMsg->wParam == VK_UP)||(pMsg->wParam == VK_DOWN) || (pMsg->message == WM_LBUTTONUP))
{
GetCaretPos((LPPOINT)ptrKeyPoint);
nCursorLine = ptrKeyPoint->y;
}

if(ptrKeyPoint)
{
delete ptrKeyPoint;
ptrKeyPoint = NULL;
}
}

CPallini25-May-09 0:03

25-May-09 0:03

susanne1 wrote:
when i move the curosor up i get a worng value of y:

What do you mean (i.e. details please)?
Smile | :)

susanne125-May-09 0:13

25-May-09 0:13

when i press the arrow key up.

Garth J Lancaster25-May-09 0:18

Garth J Lancaster

25-May-09 0:18

susanne1 wrote:
when i press the arrow key up.

I think he can get that - what y value do you get ?

'g'

CPallini25-May-09 0:25

25-May-09 0:25

Garth J Lancaster wrote:
I think he can get that

Well, it's Monday............. Big Grin | :-D

CPallini25-May-09 0:27

25-May-09 0:27

I try again: "How do you know that y value is wrong (please provide us details)?"
Smile | :)

susanne125-May-09 0:32

25-May-09 0:32

the cursor stands on the second line and ptrKeyPoint->y = 32;
it sohould be 16 instead of 32 because it starts with ptrKeyPoint->y = 0; (for the first line).

Stuart Dootson25-May-09 1:49

Stuart Dootson

25-May-09 1:49

Ummmm - just don't use the heap, OK - this should work just as well:

CPoint keyPoint;
int nCursorLine = 0;
if((pMsg->message == WM_KEYDOWN)||(pMsg->message == WM_LBUTTONDOWN))
{
if((pMsg->wParam == VK_RETURN) || (pMsg->wParam == VK_UP)||(pMsg->wParam == VK_DOWN) || (pMsg->message == WM_LBUTTONUP))
{ 
GetCaretPos(&keyPoint);
nCursorLine = keyPoint.y; 
}

See what I've done there - I've declared and defined the CPoint item as a local, so it's on the stack. Then you don't need to worry about deallocation or anything. Much simpler.

Java, Basic, who cares - it's all a bunch of tree-hugging hippy cr*p