Re: Data Scaling - C / C++ / MFC Discussion Boards

Re: Problem with mixed code of C and C++

C^hris L^osinger24-Jun-05 5:10

24-Jun-05 5:10

don't use C++ code in .C files, and don't #include headers that use C++ code in .C files.

the compiler compiles .C files as C and that means any C++ code it sees while compiling a .C file is just syntax errors.

Cleek | Image Toolkits | Thumbnail maker

Re: Problem with mixed code of C and C++

Jose Lamas Rios24-Jun-05 9:04

Jose Lamas Rios

24-Jun-05 9:04

prgr wrote:
Can anyone gimme an idea,how to deal with this?

You need to use extern to Specify Linkage[^]

Here's an example:

////////////////////////////////////
// cFunctions.h
 
#ifdef __cplusplus
extern "C" {
#endif
 
// Functions declared here can be used both
// in C and C++ code
 
int c_func_in_c_file(int someParm);
int c_func_in_c_file2();
 
#ifdef __cplusplus
}
#endif
 
 
////////////////////////////////////
// cFunctions.c
 
#include "cFunctions.h"
#include "cppFunctions.h"
 
int c_func_in_c_file(int someParm)
{
   return someParm + c_func_in_cpp_file();
}
 
int c_func_in_c_file2()
{
   return 42;
}
 
 
/////////////////////////////////////
// cppFunctions.h
 
#ifdef __cplusplus
extern "C"
{
#endif
 
// Functions declared here can be used both
// in C and C++ code
 
int c_func_in_cpp_file();
 
#ifdef __cplusplus
}
 
// Functions declared here can only be used
// in C++ code
 
int cpp_func_in_cpp_file();
 
#endif
 
 
/////////////////////////////////////
// cppFunctions.cpp
 
#include "cppFunctions.h"
#include "cFunctions.h"
 
extern "C" int c_func_in_cpp_file()
{
   return cpp_func_in_cpp_file() + c_func_in_c_file2();
}
 
int cpp_func_in_cpp_file()
{
   return c_func_in_c_file2();
}

--
jlr
http://jlamas.blogspot.com/[^]

Data Scaling

sweep12324-Jun-05 3:36

sweep123

24-Jun-05 3:36

I have to read and store data in a series of unsigned short variables in Shared Memory.

Now these variables are scaled and I need to convert them into their proper units. I display them as either float or integers.

For example for one variable:-
Max Val = 0.999969
Min Val = -1.000000

Least Sig Bit = 0.000030517578 = scaleFactor1.

So what I do is:-

// Get current data item for Shared Memory
unsigned short currentField = *(unsigned short *) fieldPtr; 
float scaleFactor1 = (float)GetScalingFactor(scaleFactorStr);
float rescaledDataEng =  currentField * scaleFactor1;

This works OK (rescaledDataEng ) for positive numbers, but for negative numbers it fails, e.g. I get for -0.44 the incorrect value displayed as (rescaledDataEng = 1.560028).

Are they a neat soultion to this problem other than adjusting for out of range data.

if(rescaledDataEng > maxValue) rescaledDataEng = -(1 -(rescaledDataEng - maxValue));

Also I dont think that I store the data (negative value corectly) when writing to the Sared Memory:-

*(unsigned short *) fieldPtr = (unsigned short)(fEngValue/m_fScalingFactor);

For example for one variable:-
Max Val = 179.999
Min Val = -180.000

Least Sig Bit = 0.005493 = m_fScalingFactor.

I enter -20 and 159.01 is displayed.

Any suggestions please as the more I look at the problem the more confused I get.

Re: Data Scaling

Budric B.24-Jun-05 5:35

Budric B.

24-Jun-05 5:35

How can you read in negative values into an unsigned variable? e.g.
unsigned short currentField = *(unsigned short *) fieldPtr;
You can never have currentField = -0.44 or -20 and have correct results.

Re: Data Scaling

Grahamfff24-Jun-05 9:11

Grahamfff

24-Jun-05 9:11

This may help to show the problem:-

For example for one variable:-
Max Val = 179.999
Min Val = -180.000

Least Sig Bit = 0.005493 = m_fScalingFactor.

I enter 20 and 20.00 is displayed; i.e. 20/0.005493 = 3640.9976 is stored
(0E38 Hex)

When I come to display the value I do the following:-

Get the value (0E38) and re-scale it = 0E38 * m_fScalingFactor = 20.00.

Works fine!

But for this example:-
Max Val = 179.999
Min Val = -180.000

Least Sig Bit = 0.005493 = m_fScalingFactor.

I enter -20 and 159.01 is displayed.

How to handle negative numbers?!?

grahamfff

Re: Data Scaling

liquid_24-Jun-05 9:12

liquid_

24-Jun-05 9:12

Additionally, you can use below formula to scale any value from one range to another:

a = (q - p)/(s - r)*(b - r) + p

if you'd like to convert value of b belonging to range (r,s) to value a belonging to range (p,q).

Re: Data Scaling

Grahamfff25-Jun-05 0:03

Grahamfff

25-Jun-05 0:03

will this equation cope with signed and unsigned data; e.g. could have a counter defined as unsigned with range 0 to 65534 with LSB = 64.

Also have angle, signed range -1 tp +1 with LSB = 0.0055

(Range -180 to +180 degrees)

Cheers,

Graham.

grahamfff

Re: Data Scaling

liquid_26-Jun-05 9:55

liquid_

26-Jun-05 9:55

The formula I gave is the most appropriate for floating point arithmetic. It does work in all cases where you have to scale one set of values from an interval to another interval. In the case of integer values there is a matter of precision loss but at some level of accuracy it works as well. LSB converts from t (old) to v (new) as follows:

v=((q-p)/(s-r))*t

PS. If I understood you well...

casync socket

_tasleem24-Jun-05 3:25

_tasleem

24-Jun-05 3:25

hi i want to create chatting applicatoin for LAN for that i m using the CAsyncsocket class. i had searched the MSDN for that purpose.
i made the client and server from the help of MSDN but i works on the single computer but not on lan.
The steps specified in the MSDN for making the server is
create (portno);
listen( );
accept();
but the server is not binded with any adress. what i should do use the bind ( ) function or give these paramter on create ( ).
there is no description what is the default values are.
can anybody help me and give the names of functions for making the Lan server and client with paramete bcos no decription what are defaul
and what not.

ddd

Re: casync socket

Ravi Bhavnani24-Jun-05 4:44

Ravi Bhavnani

24-Jun-05 4:44

Perhaps this[^] article will help?

/ravi

My new year's resolution: 2048 x 1536
Home | Articles | Freeware | Music
ravib@ravib.com

Re: casync socket

Alexander ^M.,24-Jun-05 10:33

Alexander ^M.,

24-Jun-05 10:33

Why don't you trace the code to find out what CAsyncSocket::Create does?
This is the easiest way of learning MFC Wink | ;)

.

Don't try it, just do it! Wink | ;-)

Unique number generation

Hmmkk24-Jun-05 3:13

Hmmkk

24-Jun-05 3:13

I need a good and smart way to pick a number and then make it "occupied". This is needed on 9 lines with 9 numbers from 1-9. I have a system allready with a class that makes a bool true when its occupied but there's some leak=/ because it's looped and must be applied on every line.
Does someone have a good general way to do this?

I'm trying to make a solver for this puzzle.
http://www.puzzle.jp/letsplay/applet/sd_sample_002-e.html[^]

Re: Unique number generation

David Crow24-Jun-05 3:51

David Crow

24-Jun-05 3:51

One way might be:

int nNumbers[9][9] = {0}; // initialize all to unoccupied
...
while (1)
{
    int nRow = rand() % 9;
    int nCol = rand() % 9;
    if (0 == nNumbers[nRow][nCol])
    {
        nNumbers[nRow][nCol] = 1; // mark as occupied
        break;
    }
}

There are several other approaches to this.

"Ideas are a dime a dozen. People who put them into action are priceless." - Unknown

Re: Unique number generation

Hmmkk24-Jun-05 4:04

Hmmkk

24-Jun-05 4:04

Thanks for answering but it's not really what I'm looking for.
You see, I have a class called Number.
I create a [9][9] array of it.
It is preset with 1-9 for every line:
for (int i = 0; i < 9; i++)
{
for (int j = 0; j < 9; j++)
{
myNumbers[i][j].setNumber(j + 1);
}
}
It is suppose to return a value:
//The puzzlecell is a class too.
//The random picked number for the line is put at the current position in the PuzzleCell which stores a value and will be presented in a Label when done.
myPuzzleCell[i][j].setTheValue(myNumbers[i][tempRand1].getNumber());

All this has to have the ability to preset and reset stuffs since I have all this in a huge loop which checks if the lines are good due to the Puzzle rules, which you can find at the page I linked previously. Please check the puzzle out so you get a hint.
Notice that I cannot only set random values and wish for it to be done in a century, I also have a system where I if a Line == 45 // (1+2+3+4+5+6+7+8+9) then save that line (by setting them to permanent true).

MFC and Internet ??

Rookieee24-Jun-05 3:09

Rookieee

24-Jun-05 3:09

can any one tell me how to display a web page using MFC? just like we do in VB making use of the web controls...

regards,
Rookie

Re: MFC and Internet ??

David Crow24-Jun-05 3:25

David Crow

24-Jun-05 3:25

One way is to create an SDI application using a view derived from CHtmlView. Then you can use the Navigate() or Navigate2() methods to display the page.

"Ideas are a dime a dozen. People who put them into action are priceless." - Unknown

Re: MFC and Internet ??

Ravi Bhavnani24-Jun-05 4:49

Ravi Bhavnani

24-Jun-05 4:49

Another way is to directly use a web browser control. See this[^] link.

/ravi

My new year's resolution: 2048 x 1536
Home | Articles | Freeware | Music
ravib@ravib.com

Re: MFC and Internet ??

Eytukan24-Jun-05 5:29

Eytukan

24-Jun-05 5:29

Nice link Ravi.. thanks.. i cant believe how fast i'm getting help with codeproject..!

vivek

Re: MFC and Internet ??

Smith#24-Jun-05 5:38

Smith#

24-Jun-05 5:38

thanks david n ravi. the link is really resourceful.

learning,
Rookie

Re: MFC and Internet ??

Eytukan24-Jun-05 5:40

Eytukan

24-Jun-05 5:40

i just had similar doubt.. thankz rook

vivek

RSA implementation

T1TAN24-Jun-05 2:52

T1TAN

24-Jun-05 2:52

hi ppl,

I've been looking around for a while now, and I still can't find a simple, minimum & easy to use RSA implementation (and free of course). If know of such implementation, I'd appreciate some info Smile | :)

I need to be able to use the implementation in a shareware app, but only for my protection scheme, i.e. internal use. Further on, MFC or any other framework should not be used (so, plain Win32 API or STL are good Hmmm | :|

)

I know this is a bit too specific, but if anyone knows anything, it would be a great help Smile | :)

---
YOU KNOW WHAT YOU ARE BLONDIE!?!? YOU'RE JUST A SON OF A BA A A A AAAAAAAAAA!!!!!
http://sprdsoft.cmar-net.org
http://t1tan.cjb.net

Re: RSA implementation

C^hris L^osinger24-Jun-05 5:12

C^hris L^osinger

24-Jun-05 5:12

crypto++ is what i use. it's not simple, it's not small, and it's not easy to use. but it works. Smile | :)

Cleek | Image Toolkits | Thumbnail maker

Re: RSA implementation

Alexander ^M.,24-Jun-05 10:34

Alexander ^M.,

24-Jun-05 10:34

There is also a version from RSA laboratories... It is much smaller and easier to use... but I've forgotten the download link Frown | :(

Don't try it, just do it! Wink | ;-)

Re: RSA implementation

T1TAN27-Jun-05 22:03

T1TAN

27-Jun-05 22:03

I've been digging through RSA labs site for a loooong time, and all I could find is a bunch of docs Confused | :confused:

Is there another download location for RSA stuff??

thx Cool | :cool:

---
YOU KNOW WHAT YOU ARE BLONDIE!?!? YOU'RE JUST A SON OF A BA A A A AAAAAAAAAA!!!!!
http://sprdsoft.cmar-net.org
http://t1tan.cjb.net

Re: RSA implementation

T1TAN27-Jun-05 21:43

T1TAN

27-Jun-05 21:43

yeeeees

I've tried that, I couldn't get the thing working.. D'Oh! | :doh:

Anyways, since RSA is going to be a part of anti-crack-hack-slash Smile | :)

protection, I think it would be unwise to use crypto++ since it's a separate module..could be easily avoided & my proggy would eventually get hacked (like that's not gonna happen anyway Roll eyes | :rolleyes:

)

But I might take a closer look whatsoever, thanx Cool | :cool:

---
YOU KNOW WHAT YOU ARE BLONDIE!?!? YOU'RE JUST A SON OF A BA A A A AAAAAAAAAA!!!!!
http://sprdsoft.cmar-net.org
http://t1tan.cjb.net

Use Ctrl+Left/Right to switch messages, Ctrl+Up/Down to switch threads, Ctrl+Shift+Left/Right to switch pages.