Some existing Unicode applications on Windows have been using the UTF-8 encoding of
Unicode to be able to pass Unicode data through to the Database using ODBC ANSI function
calls. These applications utilize the UTF-8 session character set. They will face
the following issues with the Unicode driver:
Supplying Latin or Kanji object names through the ANSI API will fail because the Driver
Manager cannot translate to UTF-16 using the application code page as the string will
contain invalid characters. For example, working on a Japanese PC, the Driver Manager
will attempt to convert UTF-8 characters thinking it is SJIS.
Supplying Latin or Kanji data in the SQL request, for example: “INTO T values …” will
fail because the Driver Manager cannot translate to UTF-16 using the application code
page.
Character data of type SQL_C_CHAR. The new driver will convert data to and from the
UTF-8 session character set using the applications code page. This will fail for non-ASCII
characters.
Therefore, supplying and retrieving UTF-8 data using SQL_C_CHAR data binding is not
supported.