Software Internationalization Guide

The HP Internationalization Subsystem
Software Internationalization Guide—526225-002
4-13
Internationalization Functions and HP NonStop
SQL/MP
SQL functions and predicates; however, you cannot use KANJI or KSC5601 character
sets for character columns of SQL/MX tables.
When you install SQL/MX, you can set the national character set. The national
character set is associated with NCHAR and NATIONAL CHARACTER data types and
with N string literals. If you do not specify a national character set, the default is UCS2.
For information about setting the national character set, see the SQL/MX Installation
and Management Guide.
Date-Time Formats
SQL/MX accepts date-time values in three formats, which differ from OSS. SQL/MX
includes a function that converts date-time values from one SQL/MX format to another.
If you store date-time values in an SQL/MX or SQL/MP table, you must use the
SQL/MX DATEFORMAT function to retrieve and format the values, not the OSS
functions that vary by locale. SQL/MX does not recognize values from the OSS
functions as date-time information. For more information about the DATEFORMAT
function, see the SQL/MX Reference Manual.
Message Text
To receive message text in your embedded C, C++, or COBOL85 application in
Unicode or ISO 8859-1 character set, you must declare the host variable with either of
the two character sets UCS2 or ISO88591. SQL/MX does not allow KANJI or
KSC5601 in this situation. Unlike SQL/MP, SQL/MX does not provide a DEFINE to
specify an alternate message file that contains messages in another language.
Internationalization Functions and HP NonStop SQL/MP
SQL/MP is not locale-sensitive and does not use the OSS internationalization
functions.
Character Sets and Collating Sequences
SQL/MP supports a subset of the character sets supported by OSS. For more
information about SQL/MP character sets, see the SQL/MP Reference Manual. For
information about using character sets in SQL/MP applications, see the SQL/MP
Programming Manual for C and the SQL/MP Programming Manual for COBOL85.
Single-byte Character Sets SQL/MP supports nine single-byte character sets, ISO
8859-1 through ISO 8859-9. Users define collating sequences for a column. Each
column can have a different collating sequence; a collating sequence is defined by the
CREATE COLLATION command in the SQLCI utility.
Each column of character data in a SQL/MP database is permanently associated with
a character set and a collating sequence when the column is created. If the column is
part of the key for a table or index, the collating sequence for the column determines
the storage order for rows within the table. You can specify a different character set
and collating sequence for different columns in a table, but you cannot change the