Develop Reference

Java NumberFormat seams to be stuck on / permanently overriden to english format

I am trying to parse a german number from a string. Here is a minimal example:
String stringValue = "1,00";
double value = NumberFormat.getInstance(Locale.GERMANY).parse(stringValue).doubleValue();
// value = 100.0 (locale machine from cmd, server jdk)
// value = 1.0 (locale machine from cmd, locale jdk)
Somehow the format is permanently stuck at an englisch formater. I debugged the line and had a look at the symbols.decimalSeperator and it is still a dot.
I must admit, that I can only reproduce this behaviour if I run the application with a JDK from a server in our company. Running it with my locale JDK everything works fine and as expected. So maybe the locale Locale on the server is englisch, but how can this override my hardcode Locale.GERMANY?
We are using Java 8.

Java - Error message showing question marks instead of characters

$ javac Increment.java
Output:
Increment.java:6: ??: ?????
System.out.println(++a++);
^
??: ??
??: ?
1 ???
here is the code
class Increment{
public static void main(String[] args) {
int a = 5;
System.out.println(++a++);
}
}
Does any one know what may be happening and how to fix it?
Increment is just a class for testing so that a error will appear.
I am running it in git-bash terminal, but I have tried it in cygwin terminal and windows's terminal as well.
character-set is UTF-8.

Most probably you have a national locale set up (e.g. Russian, Chinese or anything else) that makes the Java compiler to return nationalized error messages, but your terminal (cygwin) does not support the UTF-8 output or your system does not support UTF-8 locale.
As the quickest work-around you could switch java compiler to provide error messages in English:
$ javac -J-Duser.language=en Increment.java

Sublime Text cannot recognize NumberFormat

I have the following simple code in a (Windows 10) sublime text 3 file NumberFormat.java, to format a double to US currency:
import java.util.Scanner;
import java.text.NumberFormat;
import java.util.Locale;
public class NumberFormat {
public static void main(String[] args) {
double num = 1000.322;
NumberFormat defaultFormat = NumberFormat.getCurrencyInstance();
System.out.println("US: " + defaultFormat.format(num));
return;
}
}
When I build, however, the follow errors show up exclusively for NumberFormat.
NumberFormat.java:2: error: NumberFormat is already defined in this compilation unit
import java.text.NumberFormat;
^
NumberFormat.java:10: error: cannot find symbol
NumberFormat defaultFormat = NumberFormat.getCurrencyInstance();
^
symbol: method getCurrencyInstance()
location: class NumberFormat
NumberFormat.java:11: error: cannot find symbol
System.out.println("US: " + defaultFormat.format(num));
^
symbol: method format(double)
location: variable defaultFormat of type NumberFormat
3 errors
As far as I can see, there are no errors with the code. In addition, this code snippet outputs US: $1000.32 as it should on all other IDEs and text editors. Can someone explain what these errors mean? And why they exclusively occur with sublime text?

The problem is that you have imported a class called NumberFormat into a class with the same name. This is resulting in confusion over which class NumberFormat actually refers to.
(The confusion is in your mind actually. The Java language spec is clear about it. It says that NumberFormat will refer exclusively to your class. The import has not effect. Any compliant Java compiler will make the same interpretation.)
There are two solutions:
Don't import the class. Instead refer to it using its fully qualified name.
Change the name of your class so that it doesn't collide with the class you are importing.
To my mind, the second solution is better. Especially since NumberFormat is not a good name for your class anyway.
And why they exclusively occur with sublime text?
They don't. The code you have written won't compile with any conformant Java compiler.

I think you should modify your class name to a class that is different from the imported class name,or you can change you source to
java.text.NumberFormat defaultFormat = java.text.NumberFormat.getCurrencyInstance();

This is not an issue of Sublime. Sublime simply invokes the javac command from your path. You get the same result if you compile using javac outside the sublime
changing class name or be part of an explicit package name solves the problem

The
import java.text.NumberFormat
conflicts with a type defined in the same file. Changing your class name would resolve the problem.

Getting Specific symbols to appear using getCurrencyInstance() in Java

I'm writing a very simple and small example class to try and show how the .getCurrencyInstance() of the NumberFormat class in Java works, but my output is looking a little funky. It seems the class is working, but when it tries to print out the actual symbols for currency (like the Japanese yen symbol or the British pound symbol) I get weird symbols instead. Is there any way to fix this? Is there something I need to import to get these symbols on my computer?
Here's my code:
import java.util.*;
import java.text.*;
public class CurrencyFormatExample
{
public static void main(String[] args)
{
double convert = 9398.9398;
NumberFormat currencyFormat = NumberFormat.getCurrencyInstance();
System.out.println("American: " + currencyFormat.format(convert));
Locale swedish = new Locale("sv", "SE");
NumberFormat swedishFormat = NumberFormat.getCurrencyInstance(swedish);
System.out.println("Swedish: " + swedishFormat.format(convert));
Locale japanese = new Locale("ja", "JP");
NumberFormat japaneseFormat = NumberFormat.getCurrencyInstance(japanese);
System.out.println("Japanese: " + japaneseFormat.format(convert));
Locale russian = new Locale("ru", "RU");
NumberFormat russianFormat = NumberFormat.getCurrencyInstance(russian);
System.out.println("Russian: " + russianFormat.format(convert));
Locale british = new Locale("en", "GB");
NumberFormat britishFormat = NumberFormat.getCurrencyInstance(british);
System.out.println("British: " + britishFormat.format(convert));
}
}
Here is the output I get with this:
American: $9,398.94
Swedish: 9á398,94 kr
Japanese: ?9,399
Russian: 9á398,94 ???.
British: ú9,398.94
As you can see that's clearly not right. Any way I can remedy this?

In jGRASP: "Settings" > "Font", "Charset" tab. Change "I/O Charset" to UTF-8. In the "Compiler" tab, "Flags / Args" sub-tab, for "FLAGS2" of the "Run" command, add "-Dfile.encoding=UTF-8".
The "file.encoding" change will also change the default encoding files that your program reads and writes. Unfortunately, Java does not provide a way to specify a different charset for normal file I/O and I/O from stdin, stdout, and stderr when they are connected to pipes rather than a console (most IDEs will use pipes). So if you want native encoding to be used by default within your program and still want UTF-8 for I/O to the IDE, there is no way to do it. Thus, it would not be wise for the IDE to automatically add the "file.encoding" flag, unless there is a "Just use UTF-8 for everything" setting.

If you're executing this code in Eclipse, adjust your Window > Preferences > General > Workspace text encoding to UTF-8

I can't get a pound sign (£) to print to the screen

If I write
System.out.println("\u00A3");
System.out.println("\u0024");
it outputs:
ú
$
am I doing something wrong? Is it something to do with command prompt? the dollar sign is displaying correctly...but not the pound sign
any and all help appreciated, thanks :)

Find the encoding being used in the command prompt terminal using the chcp command. If this produces for example
Active code page: 850
then the encoding corresponds to the charset IBM850. Then you can use
java -Dfile.encoding=IBM850 MyApplication
which will produce the correct output

Your current Locale may change the symbols you get. Try using getSymbol() from the Currency class.