Which national character sets are supported by SQL?

Support Center

Title: Which national character sets are supported by SQL?

Object Name: mmr_ns-1.0.1788627.2671686

Document Type: Support Information

Original owner: NonStop Client Server

Disclosure level: Public

Version state: final

Environment

FACT:SQL/MP

FACT:SQL

FACT:ISO

FACT:Character Set

FACT:ISO8859

FACT:SQL/MX

FACT:ISO88591, UCS2, KANJI and KSC5601

FACT:UNICODE

FACT:Manual G06.30, SQL/MX Reference Manual, E. Standard SQL and SQL/MX, ch. "Character Set Support" says:

NonStop SQL/MX supports a limited number of national, international, and vendorspecific encoded character set standards: ISO88591, UCS2, KANJI and KSC5601.
Unicode is a universal encoded character set that lets you store information from any language using a single character set. Modern standards, such as XML, Java, Java Script, and LDAP, require Unicode. Unicode complies with ISO/IEC standard 10646. To obtain a copy of ISO/IEC standard 10646, see the Web sites listed under ISO Standards on page E-2.
SQL/MX Release 2.x complies fully with the Unicode 2.1 standard. For information about this standard, see the Web site of the Unicode Consortium:  
 
http://www.unicode.org/ 




 
NonStop SQL/MX uses UTF-16BE (16-bit) encoding for the Unicode (UCS2) character set. The full range of UTF-16 characters is allowed, but surrogate pairs are not recognized as characters. NonStop SQL/MX uses single-byte encoding for ISO88591 character set and permits wide-char KANJI and KSC5601 character set host variables in embedded applications.

Questions/Symptoms

GOAL:Which national character sets are supported by SQL?

GOAL:Which are the default character sets for SQL(MP and SQL/MX?

Cause

CAUSE:

Answer/Solution

FIX:The answer is in the SQL/MP Reference Manual, Paragraph "Character Sets":

SQL/MP allows you to associate one of the following character sets with a column, literal, host variable, or parameter:
ISO 8859/1 through ISO 8859/9, Kanji, KS C5601.
You can also define a collation that uses any of the nine ISO 8859 character sets and associate the collation with a column, literal, host variable, or parameter of the same character set. (You cannot define a collation that uses the Kanji or KS C5601 characters sets. SQL always collates characters from those character sets according to the binary value of the characters.)
For compatibility with versions of SQL/MP that do not support multiple character sets, you can specify UNKNOWN to indicate that the character set is unknown. SQL considers this equivalent to omitting the character set specification and treats the data as 8-bit data.

FIX:The default character set is as follows:

SQL/MP:
CHAR: UNKNOWN
NCHAR: KANJI

SQL/MX:
CHAR: ISO88591
NCHAR: UCS2 (if not chosen differently during InstallSqlmx)

Only the default for SQL/MX NCHAR can be chosen by the user, all other defaults are fixed.