C / C++ / MFC

Re: Exit when press "Enter" ?

18-Apr-06 9:11

See this post[^]. Seems like everybody has the same problem, today Smile | :)

.

~RaGE();

Hamid_RT18-Apr-06 18:50

Hamid_RT

18-Apr-06 18:50

Hi Surivevoli
maybe it is some helpful to you

http://bobmoore.mvps.org/Win32/w32tip41.htm

Binary inputs

CPP91118-Apr-06 9:00

CPP911

18-Apr-06 9:00

Hi

I found in some postings the data types "BYTE" and "DWORD". Can someone tell me in what library they can be found if any?

Another question, the posting says BYTE b=50; but if we have a stream of binary digits in a file, is there an easy command to input every 32 (or k bits in general) into a variable for further processing or will i have to use a pointer and keep track of the number of bits already inputed and convert it step by step to a decimal integer (by adding 2^m each time...)?

Thanks

Re: Binary inputs

Rage18-Apr-06 9:09

18-Apr-06 9:09

A BYTE is an unsigned 8bit value.
A DWORD is an unsigned 32bit value.

Should be defined in <windows.h>

CPP911 wrote:
Another question, the posting says BYTE b=50; but if we have a stream of binary digits in a file, is there an easy command to input every 32 (or k bits in general) into a variable for further processing or will i have to use a pointer and keep track of the number of bits already inputed and convert it step by step to a decimal integer (by adding 2^m each time...)?

fread ?
CFile::Read ?

~RaGE();

Re: Binary inputs

David Crow18-Apr-06 9:37

David Crow

18-Apr-06 9:37

CPP911 wrote:
I found in some postings the data types "BYTE" and "DWORD". Can someone tell me in what library they can be found if any?

You can find them in windef.h.

"Let us be thankful for the fools. But for them the rest of us could not succeed." - Mark Twain

"There is no death, only a change of worlds." - Native American Proverb

Re: Binary inputs

Stephen Hewitt18-Apr-06 15:34

Stephen Hewitt

18-Apr-06 15:34

They are defined in <windef.h> but you only need to include <windows.h> if you want to use them as it includes <windef.h>.

Steve

Converting numbers

Waldermort18-Apr-06 8:49

18-Apr-06 8:49

Hi all,

I have found myself totaly confused yet again. I need to convert a 32bit int to a double. Well the docs say "an IEEE 754 floating-point value (64bit double precision). It then goes on to demonstarte the before and after (without any code).

0x3FF00000 -> (Decoded value) 0x3FF0000000000000 -> (Result) 1.0 'decimal'

I have managed to find some examples on the net, but unfortunatly none of them are in C/C++. Here is a C# method:

double ToDouble(int n){
    byte[] doubleBytes = new byte[8];
    byte[] uintBytes = BitConverter.GetBytes(n);
    Array.Copy(uintBytes, 0, doubleBytes, doubleBytes.Length - uintBytes.Length, uintBytes.Length);
    return BitConverter.ToDouble(doubleBytes, 0);
}

I hope somebody can help me with the answer, I am totaly stuck. Thankyou.

Rage18-Apr-06 9:06

18-Apr-06 9:06

OK, so the naive dumb question first Blush | :O

: Why is this not working :

int i=3;
double d=(double)i;

~RaGE();

Waldermort18-Apr-06 9:10

18-Apr-06 9:10

When I apply that to the given example 0x3FF00000, instead of getting 1.0 I get back 1072693248.0000000. Which is well away from what I want.

Rage18-Apr-06 9:14

18-Apr-06 9:14

Sorry, I must miss something:

3FF00000(hex) = 1072693248(dec), so why should 3FF00000(hex) be equal to 1.0 ?

~RaGE();

Waldermort18-Apr-06 9:30

18-Apr-06 9:30

Thats exactly why I am so confused. The 32bit int value comes from a byte array, and it is supposed to represent a number. There is a flag which indicates whether the int is an int or a float. I can convert the int, but I am totally lost while trying to convert to a float.

An RK value is an encoded integer or floating-point value. 
RK values have a size of 4 bytes and are used to decrease file size for floating-point values. 
If bit 1 is cleared, the encoded value represents the 30 most significant bits of an IEEE 754 
floating-pointvalue (64-bit double precision). The 34 least significant bits must be set 
to zero. If bit 1 is set, the encoded valuerepresents a signed 30-bit integervalue. To get 
the correct integer,the encoded value has to beshifted right arithmeticaly by 2 bits. If bit0 
is set, the decoded value(both integer and floating-point)must be divided by 100 to get thefinal 
result.

I am able to convert the value to an integer, but this text just doesn't explain enough for me.

-- modified at 15:32 Tuesday 18th April, 2006

John M. Drescher18-Apr-06 10:10

18-Apr-06 10:10

Here is a very non optimal way (the shift can be replaced with a copy):

Take the 32bit # and mask off the lowest two bits. Then store it in a 64 bit integer __int64. Then left shift the integer by 32 bits. Then declare a double* and give it the address of the integer. Then declare a double and set it to the value that the double* (above) points to. Now if bit 0 of the first # is 1 divide the double by 100...

int code; //Set this to your 32 bit code
__int64 temp=(code & 0xFFFFFFFD); //Mask the lowest two bits of the code
temp << 32; // Shift the code 32 bits to the left
double* pFloat = (double *)temp; 
double ans = *pFloat;
if (code & 1) ans /= 100.0;

John

-- modified at 16:20 Tuesday 18th April, 2006

Waldermort18-Apr-06 10:23

18-Apr-06 10:23

Thankyou John,

Your suggestion worked a treat.

rk <<= 2;
__int64 *buf = new __int64;
*buf = rk;
*buf <<= 32;
double* dub = (double *)buf;
val = *dub;

John M. Drescher18-Apr-06 10:35

18-Apr-06 10:35

Please check my code again as I have updated it because I forgot the last rule about dividing by 100.

John

Waldermort18-Apr-06 10:40

18-Apr-06 10:40

You wouldn't beleive it, but it has taken me nearly all day to get this one little function working correctly. I just hope it will all be worth the effort.

Thanks again.

John M. Drescher18-Apr-06 10:42

18-Apr-06 10:42

Your welcome, glad I could help...

John

Rage18-Apr-06 22:18

18-Apr-06 22:18

Geez, I ended up with exactely the same code BUT for the mask that was wrong (masking only one bit and not two...). Sigh | :sigh:

That was a very interesting question, thanks for sharing it. Cool | :cool:

~RaGE();

Stephen Hewitt18-Apr-06 14:34

Stephen Hewitt

18-Apr-06 14:34

The cast from int to double is implicit. i.e.

int i=3;
double d = i;

Steve

John M. Drescher18-Apr-06 9:53

18-Apr-06 9:53

http://steve.hollasch.net/cgindex/coding/ieeefloat.html[^]

John

Waldermort18-Apr-06 20:36

18-Apr-06 20:36

Michael Dunn wrote:
FYI this belongs in the C# forum[^]

Why on earth would I want to put a C++ question in the C# forum? Please re-read my thread, I stated that I could only find a C# example of what I wanted to do. I added the function with the offchance that one of the C++ gurus here would understand it and know a way to convert it to C++.

Also if you read the repies to my initial post, you should notice there is nothing C# about them. All the code and examples are pure C++.

Toolbar

Rage18-Apr-06 8:47

18-Apr-06 8:47

Hi,

In my (MFC) application, I would like to let the user configure his toolbars (add/remove buttons, associate custom bitmaps/commands with the buttons he adds). I have no idea as where to start. I know how to use a normal CToolBar, but with a fixed amount of buttons and an imagelist with the right size, and with the event handlers associated to each button. But how do we do that dynamically ?

~RaGE();

Re: Toolbar

Rage18-Apr-06 9:22