OWASP Internationalization Java Software


 * Autor: Daniel Muñiz

Why Internationalization
This document is intended to be a guide for OWASP tool projects developed or to be developed in Java. As part of its always growing effort OWASP Internationalization project pretends to provide the basic guidelines necessary to create software ready to be translated in other languages with the unique purpose of empowering its spread on thw world and reach as many people as possible.

Introduction to Java Internationalization
Usually, programs are written in only one language such as English or Spanish and the user does not have the possibility to change it. Most of the times, the initial purpose of the program changes and it needs to be adapted to other different languages. The process of designing or adapting software in order to be displayed in several languages and regions easily, cost-effectively, and in particular without engineering changes to the software, is called internationalization. Localization can be performed by simply adding locale-specific components such as translated text, data describing locale-specific behaviour, fonts, and input methods.

In the Java SE Platform, internationalization support is fully integrated into the stock classes and packages that provide language- or culture-dependent functionality. Therefore, it is important to use these tools to make the whole process easier.

Internationalizing a program is not too difficult, but it requires some planning and extra coding. On the other hand, the benefits are huge.

This document tries to determine the requirements to internationalize a program.

Objectives

 * Use the same executable file around the world.
 * Text and data displayed in the user's language.
 * Easy way to add new languages.
 * Quickly localized
 * Spanish translation

Check List
For this specific document, the application will be used as translation case. A checklist has been redacted with the specific needs, in order to internationalize the application.

Identify Culturally Dependent Data
Text messages may change depending on the language or region. First of all, it is important to determine which type of data varies and do a list to translate it. For instance:


 * Messages
 * Help online
 * Hours
 * Numbers

Isolating text in Resources Bundles.
The translation process is a hard work but it is possible to do it easier. Nowadays, status messages, error messages and file logs displayed to end users tend to be hardcoded into programs. At this point, it is necessary to isolate the text that must be translated into ResourcesBundle objects. This object is related to a Locale that describes the language and location of the end user. Locales are defined in the java.util package and they have numerous constructors and access methods such as getLanguage and getCountry which give us information about the user's culture.

By means of these objects, we can perform a controlled translation and reduce its costs.

Dealing with Compound messages
If the displayed message contains variable data, like a username or password, the difficulty to translate the text increases. Whenever possible, we should avoid construction compound messages or modify the sentences since they are difficult to translate as we can see in the examples below:


 * Black Box - Caja Negra (Different Order)
 * The program has one vulnerability - The program has five vulnerabilities (Countable)

Formatting date and time
Date and time formats differ with region and language, therefore, it is important to determine how to display this information. By using the date-formatting classes we can solve this problem.

Using Unicode character properties
Characters in Java programming language are encoded in Unicode and provide utils and methods to deal with Unicode characters properly. Methods like comparison or identifying a character may help us. A list of Character API methods is:


 * isDigit
 * isLetter
 * isLetterOrDigit
 * isSpaceChar
 * getType
 * getType

Treating exception messages
The text of the exceptions must be displayed in the user's language so we must treat and manipulate exceptions correctly.

Comparing strings properly
When sorting text we often compare strings. If the text is displayed, we should not use the comparison methods of the String class. Instead of the String class we should use the Collator class because it allows the application to perform comparisons for different languages. Besides, the collator class increases the efficiency of string comparisons.

Use always Unicode text
If the program uses no-Unicode text then we must translate it to Unicode in such a way that we can work easily.

Detecting Text Boundaries
Applications that manipulate texts need to locate boundaries within the text. For instance, consider some of the common functions of a word processor: highlighting a character, cutting a word, moving the cursor to the next sentence, and wrapping a word at a line ending. To perform each of these functions, the word processor must be able to detect the logical boundaries in the text. We can take advantage of the methods provided by the BreakIteratorclass to perform boundary analysis.

Bibliography and source.
If you need more information on how to implement internationalization

Java Internationalization (http://java.sun.com)

WebGoat ( http://code.google.com/p/webgoat/ )

Owasp ( www.owasp.org )