Conversion of Some Greek and Turkish Characters

Description

The source used by Sybase and other companies to create the conversion tables between two ISO character sets (ISO 8859-7 and ISO 8859-8) and Unicode UTF-8 contained several errors. As a result, you may see incorrect characters in your data. 

However, this bug will show up only if your data contains these few, infrequently used characters and if your client is using either ISO 8859-7 or ISO 8859-8. The error may also appear if you migrated your data from either ISO 8859-7 or ISO 8859-8 to Unicode UTF-8 because the same conversion tables would have been used in the migration. 

The following are the affected characters:

• ISO 8859-7 (Greek)

• Left single quotation mark
• Horizontal bar (Quotation dash)
• Acute accent
  

• ISO 8859-9 (Turkish)

• Latin small letter e with circumflex
• Latin small letter i with grave
• Latin small letter i with diaeresis

Affected Products

All versions of ASE, Developer’s Kit for Unicode (UDK), DirectConnect, and Sybase Character Sets

Solution

To correct this problem, you will need to manually correct errors in your existing data and upgrade your software with the corrected conversion tables provided here.