Problems using DecimalFormat

Question

Hi!

I am having problems using DecimalFormat when I am going to print out coefficients after a regression.

Here is the part of the code that is facing problems;

DecimalFormat twoDForm = new DecimalFormat("0.00");   
private double s(double d){  
    return Double.valueOf(twoDForm.format(d));  
}  

and here is the error message in eclipse;

Exception in thread "main" java.lang.NumberFormatException: For input string: "0,16"  
 at sun.misc.FloatingDecimal.readJavaFormatString(Unknown Source)  
 at java.lang.Double.valueOf(Unknown Source)  
 at model.ARF2.s(ARF2.java:126)  
 at model.ARF2.printBestModel(ARF2.java:114)  
 at testing.testclass3.bestForecastingModel(testclass3.java:69)  
 at testing.testclass3.main(testclass3.java:36)  

Please let me know if anyone has any surgestions on how to fix the code. I want two decimals on my coefficients.

Thank you

Lars

Answer 1

http://download.oracle.com/javase/1.4.2/docs/api/java/text/DecimalFormat.html

The following excerpt appears to be part of your problem:

To obtain a NumberFormat for a specific locale, including the default locale, call one of NumberFormat's factory methods, such as getInstance(). In general, do not call the DecimalFormat constructors directly, since the NumberFormat factory methods may return subclasses other than DecimalFormat. If you need to customize the format object, do something like this:

 NumberFormat f = NumberFormat.getInstance(loc);
 if (f instanceof DecimalFormat) {
     ((DecimalFormat) f).setDecimalSeparatorAlwaysShown(true);
 }

You may want to use the applyPattern method:

applyPattern

public void applyPattern(String pattern) Apply the given pattern to this Format object. A pattern is a short-hand specification for the various formatting properties. These properties can also be changed individually through the various setter methods. There is no limit to integer digits are set by this routine, since that is the typical end-user desire; use setMaximumInteger if you want to set a real value. For negative numbers, use a second pattern, separated by a semicolon

Example "#,#00.0#" -> 1,234.56

This means a minimum of 2 integer digits, 1 fraction digit, and a maximum of 2 fraction digits.

Example: "#,#00.0#;(#,#00.0#)" for negatives in parentheses.

In negative patterns, the minimum and maximum counts are ignored; these are presumed to be set in the positive pattern.

Throws: NullPointerException - if pattern is null IllegalArgumentException - if the given pattern is invalid.

Answer 2

If you want to round the double to 2 decimal places, see this question

And when using DecimalFormat, you can convert back and forth, so no need to use Double.valueOf(..). Use twoDForm.parse(..)

Answer 3

Are you trying to format the number? Or round it? If you're formatting it, shouldn't your "s" method (bad name IMO, btw, but it's private, so it's your call) return a java.lang.String instead of a double?

Answer 4

I think what you intended to do is:

private static String s(double d) {
   return twoDForm.format(d);
}

Answer 5

You are encountering an i18n issue. DecimalFormat is using your default locale which specifies the decimal separator as ,. However, the Double.valueOf does not use the locale. It always expects that the decimal separator is ..

If you want to parse a string formatted with DecimalFormat then you need to use DecimalFormat.parse