Name: rmT116609 Date: 01/26/2003


DESCRIPTION OF THE PROBLEM :
Background:
ISCII91 is a common standard for all Brahmi based Indian scripts. The information about the particular script is encoded using a class of "ATTRIBUTE CODES" called "Font Attribute.

Problem:
If we take a proper ISCII91 Telugu language file and want to open it in a Text editor (say Jext) with encoding properly specified as ISCII91 we do not get the desired result. Similarly we can also take a proper Telugu Unicode file and try to save it in ISCII91 it again does not work. (For displaying Telugu Unicode file Pothana2000 open type fonts can be used.)

STEPS TO FOLLOW TO REPRODUCE THE PROBLEM :
1.Install Some open type Telugu fonts in jre/lib/fonts directory. Pothana2000.ttf can be obtained free of charge for personal use from
http://www.kavya-nandanam.com/dload.htm
2. Using some Java text editor (say Jext) display a Telugu Unicode file.
3.Change the Encoding option to ISCII91 and try to save the file under some different name. You will not get an ISCII91 file! You can also start with a ISCII91 Telugu file with proper "Font Attribute" and try to open it in the Editor. You will get a Devnagari file instead of a Telugu Unicode file! The
"Font Attribute" will also be visible as some junk characters.

EXPECTED VERSUS ACTUAL BEHAVIOR :
The ISCII91 Encoding should be able to handle all Indian scripts using the "Font Attribute" information. Right now it can handle only Devanagari script. Apparently the conversion algorithm is not using the "ATTRIBUTE CODE"
information as per ISCII91 standard.


CUSTOMER WORKAROUND :
 User must manually convert the Unicode files to ISCII91 files and vice versa using a separately written conversion program.
(Review ID: 158826)
======================================================================