file operation for floating point data for window OS

The following program is not giving the desired output. No warning is being shown. i am using an ide named "dev c++" . My main concern is to perform this operation on floating point no.s . But i am presently struck at integer itself. So please help me out
1)To get an output out of this
2)Suggest something to workout this on float no.s too..

thank you...

Code:

#include<stdio.h>
#include<conio.h>

main()

{

FILE *fp,*f2;

int i,n,x[100],num;

int add,mul,y[10];

int h[10]={5,2,3,6,1};

printf("contents of DATA file");

fp=fopen("DATA.c","w");


/*writing integers as an array into "data" file*/

for(i=0;i<100;i++)

{

n=i;

x[n]=2*i;

putw(x[n],fp);

}


/*printing the written data*/

while ((num=getw(fp))!=EOF)

printf("%d",num);

fclose(fp

);

fp=fopen("DATA.c","r");

f2=fopen("result.c","w");


/*using the file "data" for further processing*/

while((num==getw(fp))!=EOF)

{


/*logic for convolution*/

for(n=0;n<=104;n++)

{

add=0;

for(i=0;i<=n;i++)

{

mul=num*h[n-i];// hope 'num' holds the values of "data" file

add=mul+add;

}

y[n]=add;

putw (y[n],f2);

}

}

fclose(fp);

fclose(f2);


/*printing the contents of file "result"*/
f2=fopen("result.c","r");
while((num=getw(f2))!=EOF)
printf("%d",num);
fclose(f2);
}

If you get no warnings compiling that then you should turn the warning level up. It should at least be complaining that :

Code:

main()

should be:

Code:

int main(void)

Secondly, if you want to make it use floats, then make it use floats. Try this and post your attempt, instead of expecting someone else to do it for you (which wont happen).

Last (but not least), study up on how to do indentation and apply that to your code.

actually i was taliking about the putw() and getw() functions which doesnt work for float.... so any other function to read and write float??? and does the read() and write() work on window OS, coz i heard that they are used in unix only....

sorry that the indentation came up like this but its not same in my original code... may b somthng went wrong with my copy-paste

and coming to the point... to be specific the above code prints the 1st printf() statement and then prints few rows of "0000000"(zeros).

guidance needed....

Well I'm going to guess you're fresh from being a TurboC user, where integers are 16-bit, and this putw/getw stuff happily worked by coincidence.

But now your int's are 32-bit, so putw/getw are losing information.

> and does the read() and write() work on window OS, coz i heard that they are used in unix only....
Anything you find in stdio.h is portable across all operating systems that support stdio.

> sorry that the indentation came up like this but its not same in my original code... may b somthng went wrong with my copy-paste
You could try "go advanced" and preview your post to make sure everything is OK, and try to fix your process before pressing submit.

> for(n=0;n<=104;n++)
Given that y and h have only 10 elements, you have some big overrun issues to fix first.

Yes, read and write a *nix specific. Look into fread and fwrite.

Code:

#include<stdio.h>
#include<conio.h>
main()
{
	FILE *fp,*f2;
	int i,n,x[100],num;
	int add,mul,y[110];
	int h[10]={5,2,3,6,1};
	printf("contents of DATA file");
	fp=fopen("DATA.c","w");
	for(i=0;i<100;i++)
	{
		n=i;
		x[n]=2*i;
		fwrite(x[n],fp);
	}
	while ((num=fread(fp))!=EOF)
		printf("%d",num);
	fclose(fp);
	fp=fopen("DATA.c","r");
	f2=fopen("result.c","w");
	while((num==fread(fp))!=EOF)
	{
		for(n=0;n<=104;n++)
		{
			add=0;
			for(i=0;i<=n;i++)
			{
				mul=num*h[n-i];
				add=mul+add;
			}
			y[n]=add;
			fwrite (y[n],f2);
		}
	}
	fclose(fp);
	fclose(f2);
	f2=fopen("result.c","r");
	while((num=fread(f2))!=EOF)
	printf("%d",num);
	fclose(f2);
}

yeah its true... infact i stopped working in turboc just today and switched over to new ide.
the code above is what i modified following the suggestions.
and this is what i get after compiling.
I am sorry but couldn't really understand what they meant.

In function `main':
15 [Warning] passing arg 1 of `fwrite' makes pointer from integer without a cast
15 [Warning] passing arg 2 of `fwrite' makes integer from pointer without a cast
15 too few arguments to function `fwrite'
17 too few arguments to function `fread'
22 too few arguments to function `fread'
33 [Warning] passing arg 1 of `fwrite' makes pointer from integer without a cast
33 [Warning] passing arg 2 of `fwrite' makes integer from pointer without a cast
33 too few arguments to function `fwrite'
39 too few arguments to function `fread'

Did you even read the manual page before trying to use those functions?
fread(3): binary stream input/output - Linux man page

Originally Posted by saya407

Code:

#
	fp=fopen("DATA.c","w");
	
	fp=fopen("DATA.c","r");
	f2=fopen("result.c","w");
	
	f2=fopen("result.c","r");
	w
	fclose(f2);
}

Are you sure you want to use result.c and DATA.c ? one can end up overwriting source files using the .c "extension"