Derive Locale from formatted string

Question

I'm facing the following problem: From an external service I'm getting formatted number data which can be formatted using different locales:

1.234,00 (de_DE)
1,234.12 (en_US)

What's the best way to programmatically derive the Locale that has been used to format the String so I can parse the value to its primitive type?

Answer 1

One possibility to avoid throwing NumberFormatExceptions is to use regular expressions to detect the locale. From your example:

String sDE = "1.234,00"; // (de_DE)
String sEN = "1,234.12"; // (en_US)

Pattern patternEN = Pattern.compile("[0-9],[0-9].[0-9]");
Pattern patternDE = Pattern.compile("[0-9].[0-9],[0-9]");

System.out.println(" DE - EN : " + patternEN.matcher(sDE).find());
System.out.println(" EN - EN : " + patternEN.matcher(sEN).find());
System.out.println(" DE - DE : " + patternDE.matcher(sDE).find());
System.out.println(" EN - DE : " + patternDE.matcher(sEN).find());

Gives the following result:

 DE - EN : false
 EN - EN : true
 DE - DE : true
 EN - DE : false

You may need to refine the regex for a fully working environment.