Rounding down to its nearest decimal? BigDecimal vs Alternate Form

Question

I have the following formula and code:

I tried with Math.Floor, and others but I am not getting the result expected

double netPay = grossPay - totalDeduction;

System.out.printf("Net Pay: $%.2f", (netPay));

The idea is to get the number as follows:

46.225

Rounded as 46.22

and not as 46.23

Note that really, you should be using BigDecimal, not double. — Louis Wasserman, May 20 '19 at 05:44
@LouisWasserman: The given code does not perform any calculation which cannot be accomplished sufficiently accurate with `double`. — goodvibration, May 20 '19 at 05:47
@goodvibration `float` and `double` are approximate data types. When dealing with money, approximate is not good enough. — VGR, May 20 '19 at 05:53
How could I implement BigDecimal, and rounding that as I expected? @LouisWasserman — Hello World, May 20 '19 at 05:56

score 3 · Accepted Answer · edited May 20 '19 at 06:33

3

As stated by commenters: For financial purposes, use BigDecimal.

Aside from that:

double netPayCents = Math.floor(netPay * 100);
netPay = netPayCents / 100d;

Why use BigDecimal?

Floating point values(including double values) can be very precise, but they have issues when trying to be absolutely precise. They are always subject to rounding, after most mathematical operations.

See this thread for more detail: Double vs. BigDecimal?

edited May 20 '19 at 06:33

Anubhav Singh

8,321
4
25
43

answered May 20 '19 at 05:48

TreffnonX

2,924
15
23

The `100d` in the denominator should be optional, i.e. just `/ 100` should work. But other than that, +1 to what I was about to write. – Tim Biegeleisen May 20 '19 at 05:49
the 'd' is indeed optional, but I prefer to avoid implicit convertions, because then I am more sure that a calculation is not accidentally done in integer or long. It's just personal preference. :) – TreffnonX May 20 '19 at 05:50

score 1 · Answer 2 · answered May 20 '19 at 06:01

You can also use the below code:

import java.text.DecimalFormat;
import java.math.RoundingMode;
public class Main
{
    public static void main(String[] args) {
        double netPay = 46.225;
        DecimalFormat f = new DecimalFormat("##.00");
        f.setRoundingMode(RoundingMode.DOWN);
        System.out.println(f.format(netPay));
    }
}

Output:

46.22

score 1 · Answer 3 · 2019-05-24T09:25:25.360

Here is the method I use:

For Example netPay value is 46.225

By using Below code you will get the output like 46.22

double netPay =46.225;           // just assigning your decimal to a variable
netPay =netPay *100;              // this sets netPay to 4622.5
netPay =Math.floor(netPay);      // this sets netPay to 4622.0
netPay =netPay/100;              // this sets netPay to 46.22

Vnstead of above code you can just use below single line it will also work the same

netPay =Math.floor(netPay*100) / 100;

Happy Coding !!!!!!!!!!!!!!

Rounding down to its nearest decimal? BigDecimal vs Alternate Form

3 Answers3