C / C++ / MFC

Cedric Moonen6-Jan-09 20:14

Cedric Moonen

6-Jan-09 20:14

tasumisra wrote:
i want to know how to call a DLL

You don't call a dll. You first load the dll and when this is succesfull you can retrieve the addresses of functions in the dll and call those functions.

Cédric Moonen
Software developer

Charting control [v1.5]
OpenGL game tutorial in C++

Hamid_RT6-Jan-09 20:19

Hamid_RT

6-Jan-09 20:19

Call no load yes.
How to load a dynamic link library (DLL) into a Microsoft Visual C++ 6.0 project[^]
And see
Step by Step: Calling C++ DLLs from VC++ and VB - Part 4[^]( Hans explains for you four parts for dlls.) Wink | ;)

Of one Essence is the human race
thus has Creation put the base
One Limb impacted is sufficient
For all Others to feel the Mace
(Saadi )

CPallini6-Jan-09 21:57

6-Jan-09 21:57

tasumisra wrote:
1.cocreate

Do you mean CoCreateInstance [^]?

BTW: What about DLL's documentation [^]?

Smile | :)

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]

nyeboy7-Jan-09 21:16

nyeboy

7-Jan-09 21:16

There are two methods generally.
One:Dynamic load DLL
HINSTANCE hInst;
//DLL2.dll must in current directory.
//Implement on a button clicked event.
hInst=LoadLibrary("DLL2.dll");
typedef int (*ADDPROC) (int a,int b);
ADDPROC add=(ADDPROC)GetProcAddress(hInst,"Add");
if(!add)
{
MessageBox("Failed to get address！");
return;
}

int ret=add(5,3);
CString str;
str.Format("%d",ret);
MessageBox(str);
Two:Static load DLL
First place DLL2.dll and DLL2.lib into current directory,
then set project/settings/Link to load DLL2.lib
_declspec(dllimport) int add(int a,int b);
void CDLL2TestDlg::OnButton1()
{
int a=add(5,5);//use this method from dll
}

studing is processing in this life.

struct definition with :

KASR16-Jan-09 18:20

KASR1

6-Jan-09 18:20

what does the declaration (:8, :3, :1, etc.,) mean in the folowing struct?

struct CELL {   // Declare CELL bit field
   unsigned short character  : 8;  // 00000000 ????????
   unsigned short foreground : 3;  // 00000??? 00000000
   unsigned short intensity  : 1;  // 0000?000 00000000
   unsigned short background : 3;  // 0???0000 00000000
   unsigned short blink      : 1;  // ?0000000 00000000
} screen[25][80];       // Array of bit fields <pre>

«_Superman_»6-Jan-09 18:48

«_Superman_»

6-Jan-09 18:48

character is an unsigned short value but will occupy only 8 bits and not 2 bytes.
foreground is an unsigned short value but will occupy only 3 bits and not 2 bytes.

Similarly other fields of the structure.

For example if valid values for intensity are only 0 and 1, you only need 1 bit to represent that.
That is the idea here.

«_Superman_»

Re: struct definition with : [modified]

KASR16-Jan-09 19:09

KASR1

6-Jan-09 19:09

Thanks for your reply.
In case the size exceeds the predefined size what will happen?

«_Superman_»6-Jan-09 19:15

«_Superman_»

6-Jan-09 19:15

If value cannot fit into the number of bits allocated, it will then overflow to the next field.
It is up to the application to limit that.

«_Superman_»

modified on Wednesday, January 7, 2009 3:11 AM

Stuart Dootson7-Jan-09 1:42

Stuart Dootson

7-Jan-09 1:42

Superman's answer is incorrect - the behaviour on assignment of a value that cannot be held in the bit-field is implementation (i.e. compiler) dependent - that's what the C++ Standard says (section 4.7p3), anyway, and I suspect the C standard will be much the same.

Certainly, for Microsoft C/C++, values are truncated to what will fit in the bit-field. Consider this code:

#include <stdio.h>

struct X
{
   unsigned int a : 3;
   unsigned int b : 3;
};

int main(int argc, char** argv)
{
   {
      struct X x = {0};
      x.a = 15;
      printf("x.a = %d, x.b = %d\n", (int)x.a, (int)x.b);
   }
   {
      struct X x = {0};
      x.b = 15;
      printf("x.a = %d, x.b = %d\n", (int)x.a, (int)x.b);
   }
   return 0;
}

This produces the following output when compiled as C or C++:

x.a = 7, x.b = 0
x.a = 0, x.b = 7

Randor 6-Jan-09 19:26

Randor

6-Jan-09 19:26

Those are Bit Fields[^]

Best Wishes,
-David Delaune

How to read unicode text files

John R. Shaw7-Jan-09 6:32

John R. Shaw

7-Jan-09 6:32

KASR1 wrote:
what does the declaration (:8, :3, :1, etc.,) mean in the folowing struct?

They are the number of bits assigned to each field. The author wanted to store the data in 2 bytes (unsigned short) and specify names for each set of bits within those bytes. This is normally considered non-portable (but legal) C-Code. There are other more portatable ways of doing the same thing, but they require a little more work by the programmer.

Note that they defined exactly 16-bits; if they had specified just one more bit, it would changed the memory block size to 32-bits (unsigned int = 4 bytes on 32-bit compilers) unless some compiler specific flag was set to compact it to single by boundries.

INTP
"Program testing can be used to show the presence of bugs, but never to show their absence."Edsger Dijkstra

PS@Codeproj6-Jan-09 17:39

PS@Codeproj

6-Jan-09 17:39

How to read unicode text files (japanese chars) in C prog?

Re: How to read unicode text files

«_Superman_»6-Jan-09 17:51

«_Superman_»

6-Jan-09 17:51

There is no difference in reading ascii files and unicode files.
The difference is in how you interpret it.

The windows APIs and the C runtime functions provide a void pointer into which the read contents will be stored. So if you give the correct number of bytes to read all is fine.

How you interpret the contents of the buffer after the read is where the difference is.

«_Superman_»

Re: How to read unicode text files

John R. Shaw7-Jan-09 6:59

John R. Shaw

7-Jan-09 6:59

I have not done it in a while, but it is basically the same as reading ASII files. The difference being that you use the functions that have a 'w' as part of their name. The 'w' implies that it is expecting the 'wchar_t' type (16-bits: unsigned short) instead of the 'char' type (8-bits: signed char) when reading. If you are using a Microsoft library then you can use the '_ts...' macros which will use the 'wchar_t' functions if UNICODE is define; otherwise it will use the ASCII functions.

I hope that helps.

INTP
"Program testing can be used to show the presence of bugs, but never to show their absence."Edsger Dijkstra

CAtlRegEx Unhandled Exception

dehseth6-Jan-09 4:45

dehseth

6-Jan-09 4:45

Hey there,

I have a string which includes Turkish chars. And a regex pattern.. my mfc app is compiled with MBSC (I am pretty sure)

I got an error at the last letter says unhandled exception..

my string is (i am not sure if you can see it properly) "paylaş"
so last letter makes rgx throws an error.. I kinda deep search about the error I would like to write it if needed..

So what am I missing why it fails on turkish chars? I do not define any type of CharTraits I mean I create my variable in this way: CAtlRgx<> myregx; did not specify char type between "<" and ">"

thank you all..

Re: CAtlRegEx Unhandled Exception

JaydeepB6-Jan-09 8:47

JaydeepB

6-Jan-09 8:47

CAtlRECharTraitsA contains the traits for ANSI characters, For multibyte characters, see CAtlRECharTraitsMB.

By the way Microsoft removed ATL server support from VS 2008 including
CATLRegExp class.

Try using basic_regex<> from VC++ 2008 Feature Pack MFC Enhancements.

MFC Dialog application

bhanu_85096-Jan-09 3:55

bhanu_8509

6-Jan-09 3:55

Dear All,

Is there any way to create global objects to all the classes in the MFC dialog based application ? I want to use the object of one class from the other class, but I want to create the object only once.

The object created for each class should be a singleton object.

Please help me.

Thanks in advance.

CPallini6-Jan-09 4:07

6-Jan-09 4:07

And what is your trouble about?
You may use static members or (as you already are aware) the singleton design pattern.
BTW: the (in)famous theApp object is a (wizard created) sample of a global MFC object.
Smile | :)

bhanu_85096-Jan-09 23:14

bhanu_8509

6-Jan-09 23:14

My exact problem is that, I am using four dialog in my MFC application. I want to open the a dialog from another dialog.

I am using OnCancel() for closing the current dialog and use DoModal() to open the other dialog. If I do like this for more no. of times then I got assertion and I found that this happens due to memory overflow due to the recursive call of DoModal().

I want to create global (single) objects for all classes and using those objects I need to hide the window and show the window instead of using OnCancel and DoModal.

But I am not so experienced in defining global objects and static objects. Please help me with some samples.

P.S : I dont want to use Property sheet or Tabbed dialogs or MDI.

Thanks in advance

CPallini6-Jan-09 23:51

6-Jan-09 23:51

bhanu_8509 wrote:
I want to open the a dialog from another dialog.

This may be easily done using modal dialog boxes (i. e. DoModal()), all that you need is the proper communication of each dialog with the common parent window (see this nice toxcct's article about [^]).
Smile | :)

Rajesh R Subramanian6-Jan-09 4:40

Rajesh R Subramanian

6-Jan-09 4:40

Hi Bhanu,

It's not clear from your question as to what exactly do you want.

May be you tell what exactly is your problem and I'm sure people over here will be suggest you the best way to go about solving it.

It is a crappy thing, but it's life -^ Carlo Pallini

CPallini6-Jan-09 5:19