Notes about using the Linux edition of the HiRDB ODBC 3.5 driver

The Linux edition of the HiRDB ODBC 3.5 driver processes one character of the SQL_C_WCHAR type as two bytes in the UTF-16LE format. Therefore, when you pass data of the SQL_C_WCHAR type to a Linux edition of the HiRDB ODBC 3.5 driver, make sure that the data is in UTF-16LE format.

When an ODBC function supporting Unicode is used, the Linux edition of the HiRDB ODBC 3.5 driver checks the character codes of the connection-target HiRDB server, and converts a character string received from the driver manager from UTF-16LE format to the HiRDB server's character codes. This driver converts data received from the HiRDB server from the HiRDB server's codes to UTF-16LE format and then returns the data to the driver manager. However, if the PDCLTCNVMODE client environment definition is specified, the Linux edition of the HiRDB ODBC 3.5 driver converts character codes according to the PDCLTCNVMODE specification regardless of the character codes of the connection-target HiRDB server. The following table shows the combinations of character codes that can be used when ODBC functions supporting Unicode are used, and the character code conversion methods.

Table 14-25 Combinations of character codes that can be used when ODBC functions supporting Unicode are used and the character code conversion methods

PDCLTCNVMODE specification	ODBC driver	Client library	HiRDB server
UCS2_UTF8	UTF-16LEUTF-16LE	UTF-16LEUTF-8	UTF-8
UCS2_UJIS	UTF-16LEUTF-16LE	UTF-16LEEUC	EUC
UCS2_HJ	UTF-16LEUTF-16LE	UTF-16LEEUC-HJ	EUC-HJ
UTF8	UTF-16LESJIS	SJISUTF-8	UTF-8
UJIS	UTF-16LESJIS	SJISEUC	EUC
EUCHJ	UTF-16LESJIS	SJISEUC-HJ	EUC-HJ
Omitted	UTF-16LESJIS	SJISSJIS	SJIS
	UTF-16LEUTF-8	UTF-8UTF-8	UTF-8
	UTF-16LEEUC	EUCEUC	EUC

When an ODBC function that does not support Unicode is used, the Linux edition of the HiRDB ODBC 3.5 driver does not convert character strings passed as arguments. In such a case, you must ensure that the character codes used in the UAP execution environment match the character codes used on the connection-target HiRDB server. If the PDCLTCNVMODE client environment definition is specified, character codes are converted according to the PDCLTCNVMODE specification.
When a connection function of an ODBC function that supports Unicode is used, the Linux edition of the HiRDB ODBC 3.5 driver converts a data source name to the character codes specified in the LANG environment variable in the execution environment. Therefore, if the character codes in the execution environment's LANG environment variable do not match the character codes for the registered data source name, the driver might not be able to read a data source name that is not in ASCII characters.
The Linux edition of the HiRDB ODBC 3.5 driver does not support display of dialog boxes. Therefore, if SQLDriverConnect is called with a value other than SQL_DRIVER_NOPROMPT specified when connection is established with the HiRDB server, the driver returns SQL_ERROR.
The Linux edition of the HiRDB ODBC 3.5 driver does not support registration of data sources that use the unixODBC-provided GUI tool ODBC Config. When you register data sources, you must follow the procedure described in 14.3.1(4) Installation procedure (in the Linux edition).

14.13 Notes about using the Linux edition of the HiRDB ODBC 3.5 driver