What's wrong with this division?

Question

Here's the code:

class testsum
{ 
    public static void main(String arg[]) 
    {
        double sum=0; 
        double fraction;
        fraction=-1/9;
        System.out.println("fraction: "+fraction); 
        fraction=-1; 
        fraction=fraction/9; 
        System.out.println("fraction: "+fraction); 
    } 
}

the outputs are 0 and then -0.11111111

why was the first output 0 and not -0.11111111111?

Dude... this question is 4 years old. – David May 18 '15 at 15:58 — David, May 18 '15 at 15:58

score 8 · Accepted Answer · answered Apr 04 '11 at 20:02

8

It's doing integer division in the first example as this is the default type for a numeric literal. Try changing it to -1.0/9 (or 1d/9d - the d suffix indicates a double) and you should get the same answer.

answered Apr 04 '11 at 20:02

Jeff Foster

43,770
11
86
103

so it does integer division even though the value is going to be stored in a variable of type double? – David Apr 04 '11 at 20:03
1

The division happens before it knows it's going to be a double. It does the division in integer, and then realizes it has to be converted to a double. This has been the source of many bugs since epoch. – corsiKa Apr 04 '11 at 20:04
Yes, it doesn't take that into account (rightly or wrongly!) – Jeff Foster Apr 04 '11 at 20:04

score 2 · Answer 2 · answered Apr 04 '11 at 20:02

2

1 and 9 are both integer. Try

1.0/9

This is why it works for fraction/9, since fraction is a double.

answered Apr 04 '11 at 20:02

Gustav Larsson

8,199
3
31
51

score 1 · Answer 3 · answered Apr 04 '11 at 20:03

1

When you do -1/9, it says "-1, that's an int. 9, that's an int. -1 / 9 in integer division is 0. Oh, now I need to cast to double."

Changing it to -1.0 / 9 should solve the problem.

answered Apr 04 '11 at 20:03

corsiKa

81,495
25
153
204

score 0 · Answer 4 · answered Apr 04 '11 at 20:03

0

Try wrapping the "-1/9" in brackets.

answered Apr 04 '11 at 20:03

Chris Bornhoft

4,195
4
37
55

score 0 · Answer 5 · answered Apr 04 '11 at 20:03

0

The first one is 0 because it is doing integer division. -1 and 9 and integers and when divided equal 0. The result is then converted into a double so it can be stored in fraction. The easiest solution is this:

fraction = -1.0/9;

answered Apr 04 '11 at 20:03

unholysampler

17,141
7
47
64

score 0 · Answer 6 · answered Apr 04 '11 at 20:04

Because -1 and 9 are integers, so -1/9 is an integer division (with the result 0, which when cast to double is 0.0).

To do a floating point division, you should convert one of the numbers to double, (double) 9, 9d or simply 9.0.

In the latter case, fraction is already double (-1.0) so fraction/9 is a floating point division.

What's wrong with this division?

6 Answers6

Linked

Related