How to provide Multi-language support
Article Metadata
Managing the different text translations of a game can easily be done by supplying a simple text file for each language in the resource folder. For instance, resource strings for English go in en.txt:
Add|Choose|Remove|View|etc...
resource strings for Spanish es.txt:
Añadir| Elegir| Eliminar| Ver | etc...
In this case we choose "|" as a delimiter. It's important to use a character that doesn't commonly appear in text.
Then you can parse the text into an array using the following code:
import java.io.*;
import java.util.Vector;
public class Translator
{
private static Vector text;
public static final short add = 0;
public static final short choose = 1;
public static final short remove = 2;
public static final short view = 3;
private static final int maxBuffer = 0x800;
public Translator()
{}
public static void readText()
{
text = new Vector();
try
{
// get the text resource as a stream
InputStream is = Runtime.getRuntime().getClass()
.getResourceAsStream("/en.txt");
InputStreamReader isr = new InputStreamReader(is);
char[] buffer = new char[maxBuffer];
char[] currItem = new char[400];
int n = 0;
int read = 0;
int currPos = 0;
char c;
String strText;
do
{
read = isr.read(buffer, 0, maxBuffer);
if (read == -1)
break;
n = 0;
do
{
// now we have a part of the file stored in buffer, so then
// we load it into curr string
while(n < read && buffer[n] != '|')
{
currItem[currPos] = buffer[n];
n++;
currPos++;
if (currPos >= currItem.length-1)
{
DebugStuff.print("buffer overflow at "+text.size()+" item");
return;
}
} // end while
if (n == read) break;
// we have recieved a token before finishing the buffer
if (n < read)
{
strText = new String(currItem, 0, currPos);
text.addElement(strText);
currPos = 0;
do
{
n++;
if (n == read) break;
c = buffer[n];
} while (c == '\n' || c == '\r' || c == ' ');
// Note: there might be a problem if white space after token occurs exactly after reading a block
}
} while(true);
} while (true);
// note: any trailing characters without a delimiter will be ignored
isr.close();
}
catch (IOException e)
{
DebugStuff.print(e.toString());
return;
}
}
static String get(int id)
{
if (id >= text.size()) return "missing string";
return (String)text.elementAt(id);
}
}
(In fact this is quite handy code to use for parsing any text delimited file)
- Note: this code does not support Unicode.
Finally you access the localised strings as follows:
String str = Translator.get(Translator.add)
21 Sep
2009
This article presents code snippets on how one can provide multi-language support in Game.
To accomplish this task the writer used different text file for every language and then calling those files using input string object when user chooses appropriate language at the start of the game.
Good Coding conventions have been used and code compiles without any bugs or run time errors. Developers now can provide same functionality to their users using the code given below even in normal application. A simple way to achieve Internationalization is given in the best way. ---
24 Sep
2009
This article describes one of several possible approaches for supporting multiple languages in a Java ME midlet. The approach described in this article relies on the use of text files, with different terms in an application stored in a text file in the resource folder, delimited using a suitable character. Code is also provided to show how to process the files to read the terms into a Vector data structure. A code snippet is also provided to show how to retrieve the appropriate language terms at runtime.
This approach allows midlets to support multiple languages, without commands having to be hard-coded into the source code. The approach described in this article is also easily extensible, as adding support for a new language would merely require a different text file to be included. A similar approach is described in the following article: How to create localized text resources