loss of precision

[Tyler Jordan]

I wanted to know if there is a way to get around this error by the compiler. I have two lines of code where this error is happening.
Variables (theoctet) and (remainder) are integers:

theoctet = 256 - Math.pow(2, remainder);
intOctet[whole] = theoctet;

Based on the possible values I could use, the value of theoctet will never exceed the integer threshold. The problem is if I change 'theoctet' to a double I'm also going to have to change every other array or variable I have later on in code that uses this method. I really don't want to do that especially if I don't need it. Is there anyway around this error? Maybe a try can be used to throw and exception if the result exceeds the value?

[Ernest Friedman-Hill]

Hi Tyler,

The way to tell the compiler "I know, but it's OK" is to use a cast. This will give you a "posible loss of precision" type of error:

int x = 0;
double d = 1;
x = d;

Now, you say to yourself "That's silly, I know d is 1, and so this is going to be just fine". To tell the compiler that it's going to be fine, you "cast d to an int:"

x = (int) d;

d is then converted to an int before the assignment is made. Any actual loss of precision that occurs (i.e., overflow) is ignored.

Casts are useful in many other contexts as well -- maybe you've already used them with objects, but not with primitives in this way.

[Tyler Jordan]

Thanks for the reply.
Is there any way to get rid of the .0 at the end of the value now. The problem is that I put it in a string and unfortunately the cast carries over the .0 into the integer.

[Tyler Jordan]

Figured it out, I didn't change the array back to integer, so it was maintaining the double format. That was my bust. Thanks for the information about the cast. Works appropriately.