Steps to reproduce :

Platforms: Solaris / windows
Build : b42
Testsuite: GS05/older version GS04.

1)Login with Hindi locale on solaris / set the language & regional settings on windows to
hindi .
2)Run the test suite using the following command:
 sh run_gs.sh -jdk:<JAVA-HOME-path> ../../testlist/GS/long-manual_hi_IN.txt
3) Consider for instance, the below set of test cases that covers all the strings where in
there are problems: Go to the corresponding rows listed below in each of these test case
and observe the problem. Please refer the attachments listed to identify the exact
location where there the chars are
repeated.
- DrawStringLTest1.java : Rownos= 7,8 .
   Refer Attachments: DrawStringLTest1_7thRowFail.jpeg and
DrawStringLTest1_8thRowFail.jpeg
- DrawStringLTest2.java : Rownos= 1.
   Refer Attachment : DrawStringLTest2_1stRowFail.jpeg
- DrawStringLTest3.java : Rownos= 9(last row)
   Refer attachment: DrawStringLTest3_9thRowFail.jpeg
-DrawStringLTest4.java : Rownos= 1
  Refer attachment : DrawStringLTest4_1stRowFail.jpeg

Other set of cases where in the problem can be observed for the same set of
rows/strings:
 JButtonLTest# , JCheckBoxLTest# , JComboboxLTest#, JEditorPaneLTest#, etc


Occuring in Older builds?:- NO. Tested for b41. No problem found.


NOTE: An Important Obsevation:
Use the following steps in any of the TextEditing components to notice this:
1)Input the hindi chars using unicode seq: \u091c\u093c\u0921
2)Observed that whatever char given after u093c will be repeated twice.

Hence in all the above jpeg files you may find that wherever this char(u093c) is used
the next char is repeated twice.

Tested using demo files under j2sdk1.5.0: <java>/demo/jfc/Notepad/java -jar Notepad.jar
The problem stated above can be generated here too.

Is regression? Yes. Not found in older builds. Occuring only in b42(latest been tested)





###@###.### 2004-03-16
###@###.### 2004-03-16