|
Advertisement |
Transliteration using JAVA
Posted On August 26, 2010 by GB Shah filed under Programming
HTML clipboard
This article focuses on the concept of Transliteration and how to design transliteration engine using JAVA.
Transliteration is a mapping from one system of writing in one language into another language. Generally computer users have the English type writing skill, this is enough for typing other languages. This is the main advantage of transliteration.
UNICODE:
Unicode provides a unique number for representing every character for all of its supporting languages and symbols. Before Unicode, no other systems had enough character supporting. The Unicode standard has been adopted by such industry leaders as Microsoft, Sun, IBM, Apple, and Oracle. So it is supported in many operating systems and browsers. For more details about Unicode visit, http://www.unicode.org
JAVA provides a rich set of Unicode support for the world’s most popular languages as well as Indian languages like Hindi, Bengali, Tamil, Telugu, Gujarati, Kannada, etc... Each of these languages have been provided a specific Unicode range. For example, Hindi language has the Unicode range from U+0900 to U+097F and Tamil language has U+0B80 to U+0BFF. Characters of those languages are within the specified range.
If we want to represent other languages, we must specify the Unicode of that language characters in program. Each Unicode number is accepted as a character by java. True type fonts are used to represent these characters.
FONTS IN WINDOWS:
Windows has the true type fonts for all popular languages that support Unicode characters. If we want to add another true type font, just copy the font and paste it in the C:\WINDOWS\Fonts directory.
Windows has the default Tamil font “Latha.ttf”.
FONTS IN LINUX:
If we want to work in Linux platform we must install our language supporting fonts. In Linux,( I am using Ubuntu8.10) to install fonts, go to “/home” directory, then go to “view” menu and enable “Show hidden files” option, and check “.fonts” directory. If it appears then paste the “.ttf” Unicode supporting fonts from windows also. If it does not appear, then create a directory by the same name “/home/user/.fonts” then paste the fonts to that directory. Next we must be refreshing the font cache, Go to,
Applications->Accessories->Terminal
Then enter,
~$ fc-cache –f –v
Now the font cache has been refreshed.
AWT AND SWING:
We should not use AWT (Abstract Window Toolkit) components. It may provide display problems. We must use swing components because AWT components first convert the Unicode characters to the host's native character set encoding. If the target character set does not have the needed Unicode character, a substitute character is often used to represent the original character. AWT components are not typically flexible enough to display wide ranges of multilingual text because of their dependence on a single, rather limited charset or codepage.
Swing components do not suffer from the limitations as AWT components. Because Swing components do not convert a Unicode character to the host's native charset or codepage, these components can typically display a wide range of multilingual text.
For example, we want to display “akaram” in Tamil; we must enter the specified character’s Unicode number in print function by the Unicode table specified.
a->”\u0B85”
ka->”\u0B95”
ra->”\u0BB0”
m->”\u0BAE\u0BCD”
It is represented in a program by the following way,
Font font;
JTextArea area;
--------------------------------
---------------------------------
font =new Font(“Latha”,Font.PLAIN,14);
area.setFont(font);
area.setText
(“\u0B85\u0B95\u0BB0\u0BAE\u0BCD”);
---------------------------------
---------------------------------
Result:
String eng[]={“a”,”aa”,”i”,”ii”,……….”k”,”ka”,”kaa”,………………}
String tam[]={“\0B85”,”\u0B86”,”\u0B87”,”\u0B88”,…………”\u0B95\u0BCD”,”\u0B95”,”\u0B95\u0BBE”,… }
This is the sample code for displaying Tamil text. But if we want to design a transliteration engine, first we will create the array of string for possible letters in English to the corresponding Tamil letters then create another array of string for representing Tamil Unicode characters. These two arrays will be mapped to transliterate.
These two arrays are compared for each character entry in English. It checks whether the entered character is in the array eng[], if it is in the array then the program prints the Tamil character specified in the corresponding tam[] array position.
The program must be written carefully for representing character combinations. The character “a” entered then the program will print the corresponding letter in Tamil. But if the next character is again “a” then the program checks the correct equivalent for “aa” and the previous character for “a” must be replaced. The transliteration engine should be designed in this way.
And one more thing is reading and writing Unicode characters from and to file. Input Streams do not support as they do byte reads as well as Output Streams do not support Unicode file writing. So we should use Reader and Writer classes as UTF-8 format by following.
Unicode character reading from File
if(e.getSource()==b2)
{
JFileChooser dlg=new JFileChooser();
int a = dlg.showOpenDialog(frame);
if (a == JFileChooser.APPROVE_OPTION)
{
File file = dlg.getSelectedFile();
try
{
FileInputStream fis=new FileInputStream(file.getPath());
Reader reader = new InputStreamReader(fis, "UTF-8");
char[] x=new char[1024];
StringBuilder data = new StringBuilder();
while(true)
{
int size=reader.read(x);
if(size<0)
break;
data.append(x,0,size);
}
area.setText("");
area.append(""+data);
} catch(Exception ex) { }
}
}
Unicode character writing to the File:
if(e.getSource()==b3)
{
JFileChooser dlg=new JFileChooser();
int a = dlg.showSaveDialog(frame);
File file = dlg.getSelectedFile();
try{
String x="";
x=area.getText();
Writer w = new OutputStreamWriter(new FileOutputStream(file.getPath()),"UTF-8");
w.append(x);
w.close();
} catch(Exception ex) { }
}
CODE FOR ENGLISH TO TAMIL TRANSLITERATION:
The following full code is used for English to Tamil transliteration. Simply edit the Unicode string array and we can use this same code for other languages.
transliter.java
|
import java.io.*; import java.awt.*; import java.awt.event.*; import javax.swing.*; import javax.swing.event.*; import javax.swing.text.*; public class transliter implements ActionListener { String enk[]= {"a","aa","i","ii","u","uu","e","ee","ai","o","oo","au","k","ng","s","nj","d","N","th","n-", String tak[]={"\u0B85","\u0B86","\u0B87","\u0B88","\u0B89","\u0B8A","\u0B8E","\u0B8F","\u0B90","\u0B92","\u0B93","\u0B94", JTextField text1; b1.setMnemonic('a'); b1.setToolTipText("Press \"Alt+a\" for append text to TextArea"); b1.setBounds(2,62,145,25); b2=new JButton("open"); b2.setToolTipText("Press \"Alt+o\" to open a file"); b2.setMnemonic('o'); b2.setBounds(149,62,145,25); b3=new JButton("save"); b3.setToolTipText("Press \"Alt+s\" to save as file"); b3.setMnemonic('s'); b3.setBounds(296,62,145,25); b4=new JButton("clear"); b4.setToolTipText("Press \"Alt+c\" to clear the TextArea"); b4.setMnemonic('c'); b4.setBounds(443,62,136,25);
panel.setLayout(null); if(e.getSource()==b2) FileInputStream fis=new FileInputStream(file.getPath()); Reader reader = new InputStreamReader(fis, "UTF-8"); char[] x=new char[1024]; StringBuilder data = new StringBuilder(); while(true) { int size=reader.read(x); if(size<0) break; data.append(x,0,size); } area.setText(""); area.append(""+data);
} catch(Exception ex) { } } } if(e.getSource()==b3) { JFileChooser dlg=new JFileChooser(); int a = dlg.showSaveDialog(frame); File file = dlg.getSelectedFile(); try { String x=""; x=area.getText(); Writer w = new OutputStreamWriter(new FileOutputStream(file.getPath()),"UTF-8"); w.append(x); w.close(); } catch(Exception ex) { } } if(e.getSource()==b4) area.setText(""); } public void changedUpdate(DocumentEvent e) s=text1.getText(); } } if(c==0) { if((i+1)==(l-1)) continue ; } if(c==1) r=Character.toString(s.charAt(i-1))+Character.toString(s.charAt(i))+Character.toString(s.charAt(i+1)); if(cc==0&&c==1) if(cc==1) r=Character.toString(s.charAt(i-2))+Character.toString(s.charAt(i-1))+Character.toString(s.charAt(i)) else r=Character.toString(s.charAt(i))+Character.toString(s.charAt(i+1))+Character.toString(s.charAt(i+2))+ for(int n=47;n<338;n++) if(c1==0) if(c4==4||c4==0) }catch(Exception ex) { } } public static void main(String args[]) |
SCREEN SHOTS:
About Author
R.HEMANATHAN, M.C.A ( Final Year)
A.C.College of
Engg & Technology ( GOVT),
KARAIKKUDI-630 004
