Thread: floor() problems

I am doing an exercise and I have to make a program that uses the funtion floor() to round a decimal down to the nearest integer.

I have never used floor() before but I am aware it is in the <cmath> header. After doing a bit of online searching on it I discovered it takes and returns and double value.

This is what I have done but I am getting an identical output each time and the function is not working. If I attempt to change anything to an int I get a warning about "possible loss of data".

Can anyone offer my any advice?

Code:

#include <iostream>
#include <iomanip>
#include <cmath>

// function prototype
double roundToInteger ( double );

// main function - driver //////////////////////////////////////////////////////
//
int main ( void )
{
    std::cout << std::setprecision( 2 ) << std::fixed;

   for ( double i = 1.0; i <= 10.0; i++ )
   {
      std::cout << i << " rounded is: " << roundToInteger ( i )
	     << std::endl;
   }

    std::cin.get(); // freeze console output window
   
    return 0; // return value from int main
}

// function that rounds and returns the passed
// value to the nearest integer
double roundToInteger ( double num )
{
	double x = 0;

	x = floor ( num );

	return x;
}

I am getting an identical output each time

That's because you are providing input for which floor() has no net effect. The idea is to provide input that are in between integral values, and then see the effect that floor() would have.

What do you mean "not working"?

Mine lists:

Code:

1.00 rounded is: 1.00
2.00 rounded is: 2.00
3.00 rounded is: 3.00
4.00 rounded is: 4.00
5.00 rounded is: 5.00
6.00 rounded is: 6.00
7.00 rounded is: 7.00
8.00 rounded is: 8.00
9.00 rounded is: 9.00
10.00 rounded is: 10.00

which is what I expect the function to give.

Floor will return the integral value (with the fraction chopped off -> chopping towards zero). Since your double values are already integers, you can't see the difference.

If you where to add for example 0.7 on each iteration, the two columns would show different values.

--
Mats

I believe swgh is confused that i++ adds 0.1 to the value or something each iteration, which it doesn't. It adds 1, so the loop is 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, which have no decimals to chop.

Originally Posted by tabstop

Eh -- for positive numbers, that's true, but not for negatives. Floor always goes down, so floor(-2.5) is -3. To chop toward zero, use trunc().

Ok, thanks for the correction. I tend to use floor/trunc very rarely (I think about once or twice in my life), so I must confess I'm a bit rusty on how they work.

--
Mats

Thanks guys for the tips. I appreciate it.