Thread: school assignment; decimal rounding

Hi, I've been trying to get this problem to work for two days for my school. I'm just starting computer science class in Java. I really really really need help on my code. I've sent the teacher an email but I'm scared I won't get a response in time. I think I'm good at this class but I've only been doing it for a week or so. I have very little time to finish. Please help.

By the way, I know how to round to one digit. Just not a variable number of digits.

Here is the problem:

"In the following segment of code, a and b are non-negative doubles and d is an int between 0 and 5 (inclusive). Write code that assigns to b the result of rounding a to d decimal places."

 
Here is my code so far:
 
b = a * Math.pow(10, d);
b += 5; 
b = (int)b;
b *= 10;
b = (int)b;
b /= Math.pow(10, d);
b = (int)(b * Math.pow(10, d + 1));
b /= Math.pow(10, d + 2);

Help appreciated. Thanks.

Just looking at the API, I'm not even sure how that Math.pow() is working at all. There are apparently three classes in the Math package. Both BigInteger and BigDecimal have pow() methods, so how is it even differentiating which one you're calling?

Either way, it looks like you're using a two-argument version in any case. It's Math.BigDecimal.pow(int n, MathContext mc) that takes two args, and as you can see, the second one is a MathContext object.

So if I'm reading that right, you would need to feed a MathContext object to it.

Looks like you need a MathContext with a "precision," whatever that is (I'm learning this too lol), appropriate for rounding to the nearest tenth, then feed that to your code.

You have to write a method that creates a MathContext object USING the value of d as its precision, and set b equal to Math.round(a, mc).

At least, so it seems.

-summit45

@summit45 Thanks for the help. I would really appreciate more experienced advice though. But you can still help.

I don't think he wants me to use Math.round() because he's never taught me that before. Are you sure there's no way to do it with casting? I know it works for one decimal place.

Originally Posted by Junky

What kind of "rounding" do you need?

Is 4.637 rounded to 2 decimal places 4.63 or 4.64 rounded up?

4.637 rounded would be 4.64.

--- Update ---

I'll see what he says. I sent him an email. If he says "no" which I believe he won't, what would you say to do?

Here's my code:

 
b = a + (5 * Math.pow(10, -d));
b *= Math.pow(10, d);
b = Math.round(b * d);
b /= Math.pow(10, d);

I think I'm doing SOMETHING wrong still. What you guys have said has helped a lot.