how to print 2 decimal places of a double?

nadroj · 07-11-2005, 09:18 PM

straightforward question..
just wondering if their is a standard c++ header that allows you to return a double with 2 decimal places.
Thanks
edit: without using printf()

nadroj · 07-11-2005, 09:50 PM

nevermind, i made my own

Code:

string round(string i)
{
       return i.substr(0, strcspn(i.c_str(),".") + 3);
}

jim mcnamara · 07-12-2005, 09:36 AM

Is there some reason you don't want to use a standard library call -- fprintf printf sprintf?

All this means is that two years from now nobody is going to know why your "round" is there.

You do know round is part of the standard library - you should pick another name for yours.

nadroj · 07-12-2005, 10:15 AM

its a c++ project so im using cout. i wouldnt want to use cout for the entire program then printf for that one line, it shouldnt be necessary.
is there a link to this round header file? either way, it would probably act like ceil or floor and wouldnt do what i am looking for here, i would think.
i just picked the name round because i couldnt think of anything at the moment.. and this project is my first c++ program, and i wouldnt be releasing it at all, so.

deiussum · 07-12-2005, 10:33 AM

You can use setprecision to set the number of significant digits displayed for a floating point number. That includes the number of digits in front of the decimal point as well, though. For example:

Code:

#include <iostream>
#include <iomanip>

using namespace std;

int main()
{
   float f = 100.123456;

   cout << setprecision(4) << f << endl;
}

Output:
100.1

I'm looking more to see if there is a way to set the precision similar to the C way, which I have usually reverted to. You could always use sprintf to put the result in a C-style string, then use that C-Style string in a C++ string object, or cout or whatever.

For instance:

Code:

char buffer[256];
float f = 100.1234567;
std::string s;

sprintf(buffer, "%.2f", f); // results in 100.12

s = buffer;

deiussum · 07-12-2005, 10:48 AM

Ok, I figured out a way you can do it using only C++ streams.

Code:

float f = 100.123456789;

cout << setprecision(2) << fixed << f << endl;

Output:
100.12

The good thing about this method is that if you have, say 1.2. It'll output 1.20. It doesn't appear that your round method takes that into account, but maybe you don't have any numbers with fewer than 2 decimal places so it doesn't matter.

nadroj · 07-12-2005, 11:03 AM

perfect deli!
Thanks alot, thats just what i need! too bad i already wrote mine, and, as you mentioned, no it doesnt take care of the XXX.X (1 decimal place) scenario, but again as you said, i havent seen a number like that in the program so far.. but ill probably go with the method you showed as its most likely faster and works better!