Chapter 12 ■ Localization

Introduction

Localization is all about making the software relevant and usable for the users from different cultures—in other words, customizing software for people from different countries or languages. How do you localize a software application? Two important guidelines should be heeded when you localize a software application:

•Do not hardcode text (such as messages to the users, textual elements in GUIs, etc.) and separate them into external files or dedicated classes. With this accomplished there is usually minimal effort to add support for a new locale in your software.

•Handle cultural-specific aspects such as date, time, currency, and formatting numbers with localization in mind. Instead of assuming a default locale, design in such a way that the current locale is fetched and customized.

 Text may not be the only thing that needs to be localized in an application. For example, if your application­ uses audio sounds to give instructions, they will need to be changed for localization. Similarly, if the software displays some glyphs or pictures for a specific locale, they also need to be transformed.

Locales

A locale is “a place representing a country, language, or culture.” Consider the Canada-French locale. French is spoken in many parts of Canada, and this could be a locale. In other words, if you want to sell software that is customized

for Canadians who speak French, then you need to facilitate your software for this locale. In Java, this locale is represented by the code fr_CA where fr is short for French and CA is short for Canada; we’ll discuss the naming scheme for locales in more detail later in this section.

The Locale Class

In Java, the Locale class provides programming support for locales. Table 12-1 lists important methods in this class.

362

The code in Listing 12-1 detects the default locale and checks the available locales in the JVM.

Listing 12-1.  AvailableLocales.java

import java.util.Locale;

class AvailableLocales {

public static void main(String []args) {

System.out.println("The default locale is: " + Locale.getDefault()); Locale [] locales = Locale.getAvailableLocales();

System.out.printf("No. of other available locales is: %d, and they are: %n", locales.length);

for(Locale locale : locales) {

System.out.printf("Locale code: %s and it stands for %s %n", locale, locale.getDisplayName());

}

}

}

It prints the following:

The default locale is: en_US

No. of other available locales is: 156, and they are: Locale code: ms_MY and it stands for Malay (Malaysia) Locale code: ar_QA and it stands for Arabic (Qatar)

Locale code: is_IS and it stands for Icelandic (Iceland)

Locale code: sr_RS_#Latn and it stands for Serbian (Latin,Serbia) Locale code: no_NO_NY and it stands for Norwegian (Norway,Nynorsk) Locale code: th_TH_TH_#u-nu-thai and it stands for Thai (Thailand,TH) Locale code: fr_FR and it stands for French (France)

Locale code: tr and it stands for Turkish

Locale code: es_CO and it stands for Spanish (Colombia) Locale code: en_PH and it stands for English (Philippines) Locale code: et_EE and it stands for Estonian (Estonia) Locale code: el_CY and it stands for Greek (Cyprus)

Locale code: hu and it stands for Hungarian [...rest of the output elided...]

Let’s look at the methods in the program before analyzing the output. You use the method getDefault() in Locale to get the code of the default locale. After that you use getAvailableLocales() in the Locale class to get the list of available locales in your JVM. Now, for each locale you print the code for the locale by implicitly calling the toString() method of locale and also print the descriptive name using the getDisplayName() method of Locale.

363

Chapter 12 ■ Localization

The program prints the default locale as en_US for this JVM, which means the default is the English language spoken in US. Then it prints a very long list of available locales; to save space, we’ve shown only small part of the output. From this program, you know that there are many locales available and supported, and there is a default locale associated with every JVM.

There are four different kinds of locale codes in this output:

•Just one code, as in the last entry shown above: hu for Hungarian.

•Two codes separated by underscore, as in the first locale shown, ms_MY, where ms stands for Malaysia and MY stands for Malay.

•Three codes separated by underscores, as in no_NO_NY where no stands for Norway, NO for Norwegian, and NY for Nynorsk.

•Two or three initial codes separated by underscores and the final one by # or _#, as in th_TH_TH_#u-nu-thai, which we’ll discuss now. Here is how these locale names are encoded:

language + "_" + country + "_" + (variant + "_#" | "#") + script + "-" + extensions

For the locale code of th_TH_TH_#u-nu-thai,

•The language code is th (Thai) and it is always written in lowercase.

•The country code is TH (Thailand) and it is always written in uppercase.

•The variant name is TH; here it repeats the country code, but it could be any string.

•The script name is an empty string here; if given, it will be a four-letter string with the first letter in uppercase and the rest in lowercase (e.g., Latn).

•The extension follows the # or _# character; it is u-nu-thai in this example.

This coding scheme is to allow programming variations even within the same language. For example, English is spoken in many countries, and there are variations in the language based on the country in which English is spoken. We all know that American English is different from British English, but there are many such versions. Let’s change the loop in Listing 12-1 to list only the locales that are related to English, like so:

for(Locale locale : locales) {

// filter and display only English locales if(locale.getLanguage().equals("en")) {

System.out.printf("Locale code: %s and it stands for %s %n", locale, locale.getDisplayName());

}

}

It prints the following:

Locale code: en_MT and it stands for English (Malta)

Locale code: en_GB and it stands for English (United Kingdom) Locale code: en_CA and it stands for English (Canada)

Locale code: en_US and it stands for English (United States) Locale code: en_ZA and it stands for English (South Africa) Locale code: en and it stands for English

Locale code: en_SG and it stands for English (Singapore) Locale code: en_IE and it stands for English (Ireland)

364

Chapter 12 ■ Localization

Locale code: en_IN and it stands for English (India) Locale code: en_AU and it stands for English (Australia) Locale code: en_NZ and it stands for English (New Zealand) Locale code: en_PH and it stands for English (Philippines)

The output refers to different locales in English. You use the getLanguage() method in Locale, which returns the locale code. What are other such methods? You’ll explore the methods available in the Locale class now.

Getting Locale Details

The getter methods in the Locale class such as getLanguage(), getCountry(), and getVariant() return codes, whereas the similar methods getDisplayCountry(), getDisplayLanguage(), and getDisplayVariant() return names. Listing 12-2 illustrates how to use these methods for the locale Locale.CANADA_FRENCH.

Listing 12-2.  LocaleDetails.java

import java.util.Locale;

public class LocaleDetails {

public static void main(String args[]) { Locale.setDefault(Locale.CANADA_FRENCH); Locale defaultLocale = Locale.getDefault();

System.out.printf("The default locale is %s %n", defaultLocale); System.out.printf("The default language code is %s and the name is %s %n", defaultLocale.getLanguage(), defaultLocale.getDisplayLanguage()); System.out.printf("The default country code is %s and the name is %s %n", defaultLocale.getCountry(), defaultLocale.getDisplayCountry()); System.out.printf("The default variant code is %s and the name is %s %n", defaultLocale.getVariant(), defaultLocale.getDisplayVariant());

}

}

It prints the following:

The default locale is fr_CA

The default language code is fr and the name is français The default country code is CA and the name is Canada The default variant code is and the name is Canada

Let’s understand the program. The setDefault() method takes a Locale object as argument. In this program, you set the default locale as Locale.CANADA_FRENCH with this statement:

Locale.setDefault(Locale.CANADA_FRENCH);

The Locale class has many static Locale objects representing common locales so that you don’t have to instantiate them and use them directly in your programs. In this case, Locale.CANADA_FRENCH is a static Locale object. Instead of using this static Locale object, you can choose to instantiate a Locale object. Here is an alternative way to set the default locale by creating a new Canada (French) locale object:

Locale newLocale = new Locale("fr", "CA", ""); Locale.setDefault(newLocale);

365