Java unsigned - remove sign bit

Question

I have this function:

public static final int UNSIGN(short b){

    return (b & 0xFFFF);
}

The intent here is to use the value in 'b' as if it were positive, by this I mean as if it were an unsigned int, capable of holding 65536 values. However, it doesn't work. For example:

If 'b == -122' the function returns '65414'.

I also tried this way:

public static final int UNSIGN(short b){

    return (b & 0xFF);
}

But that would return me '48' if 'b == 304'.

I'm really lost here, hope someone can help out.

-122 is not positive so what value did you expect the function to return? — Peter Lawrey
– Peter Lawrey, Commented Jan 27, 2013 at 21:22

Jon Skeet · Accepted Answer · 2013-01-27 21:21:22Z

3

The intent here is to use the value in 'b' as if it were positive.

Do you mean you want to find the absolute value of b? That would be Math.abs.

But your code is treating b as if it were an unsigned short. When b is -122, the bits will be

1111111110000110

... which is 65414 when viewed as an unsigned value.

(122 is 0000000001111010, so to negate it you reduce the value by 1 and flip all the bits.)

answered Jan 27, 2013 at 21:21

Jon Skeet

1.5m893 gold badges9.3k silver badges9.3k bronze badges

Sign up to request clarification or add additional context in comments.

9 Comments

sazr Over a year ago

ok I guess I did not explain myself very well. I do not want the absolute value, I want the positive value for the 16bits. basically I want the value as if it were unsigned short.

Jon Skeet Over a year ago

@sazr: And that's exactly what it's giving you. What value did you want it to give you for b == -122?

Cephalopod Over a year ago

why do you expect that value?

Cephalopod Over a year ago

Exactly. Furthermore, in your question you write that you expect some result for b=304. So apparently, you don't want bytes either. Maybe you should take a step back and think about what you actually want, because if Jon Skeet's answer isn't right, the question is wrong.

Cephalopod Over a year ago

If the parameter is always in [0;255], you have already given the correct answer in your question: return (b & 0xFF);. Just make sure you never call it with something else. (Also, if your question is hard to explain, what kind of answers do you expect when you give almost no explanation at all?)

|

Collectives™ on Stack Overflow

Java unsigned - remove sign bit

1 Answer 1

9 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

9 Comments

Your Answer

Sign up or log in

Post as a guest

Related