Replacing dot with comma

[lovelace]

Hey.

I have been wondering about this for some time now, and when I couldnt get to a solution myself, I tried with google.

I know there is a function in string to check for the first character you want it to find, etc. And find how many of that character theres in the string. But how do I replace something in a long float value?

like

double 3.234

I just want a few tips on how I could get started. That would be really awesome.

[KCfromNC]

You change float (and int) values by doing math on them. The decimal place is inherent in a floating point value, though, so you're not going to replace it as such.

If you want to print out 3.234 as 3,234 there are a few approaches. First, you can look into localization - e.g. The GNU C Library - Locales and Internationalization. I think this is POSIX rather than standard C, so your compiler may or may not support it.

The other approach, if you just need to do this for floating point values is to print them into a string using sprintf() and then use whatever approach is easiest to find and replace '.' with ',' (strchr() seems as good a suspect as any).

I'd prefer the former approach if you're doing this for real code. If you're doing it for a school assignment your instructor probably expects the latter. Whether they'd be impressed with or consider it cheating if you use the C localization functions is up for grabs.

[MK27]

But how do I replace something in a long float value?

like

double 3.234

Math or bit manipulation. You can't put a comma in there because it isn't a character type. Also, you cannot deal with a "single element" (such as '3' or '.') because they do not exist; it is not an array, it's just one value.

"3.234" is a convenient decimal representation which your compiler recognizes. Consider:

char a[] = "123456789";
int b = 123456789;
int c = 3;

In memory, the first one is 10 bytes long, 1 for each character/digit plus a null terminator. The second one is only four bytes long (or whatever the fixed size of an int is on your system). The third one is the exact same size as the second one because they are the same type, altho the value of b is > c.

Numbers are actually stored using power of two binary*. So actually are characters, each one is a single 8-bit byte with 256 possible combinations. "Power of two" means the possible combinations increase with each bit (binary digit), so 32-bits (a normal int) have 4294967295 possible combinations. For signed integers, this is the range -2147483648 to 2147483647. Try this:

	unsigned int x = ~0;
	printf("%u\n", x, n);

~ inverts the bits. 0 is no bits set, so inverted, it is all bits sets; therefore this will print "4294967295".

* signed types use a slight variation called two's complement.

[lovelace]

Ah cheers, ill have a look at it later, thanks for taking your time for answering I didnt know what to expect, but now I got someplace to start

[CommonTater]

Ah cheers, ill have a look at it later, thanks for taking your time for answering I didnt know what to expect, but now I got someplace to start

Are you trying to display the european money format such as 123,06?

That's actually pretty easy to do...
Use sprintf() to write the floating point value to a string.
Use strchr() to find the dot
Then change the value at the pointer to a coma...

char EuroStd[16];
char * dot;

sprintf(EuroStd,"%.2f", value);

dot = strchr(EuroStd,'.');

if (dot)
  *dot = ',';

printf("%s",EuroStd);