SQL/MP Reference Manual

HP NonStop SQL/MP Reference Manual523352-013
S-81
Considerations—String Literals
_UNKNOWN
specifies that data in the string literal belongs to an unknown character set.
Specifying UNKNOWN is equivalent to omitting the character set specification.
NonStop SQL/MP handles the data as if it were 8-bit data.
N
associates the system default multibyte character set with the string literal. (Kanji is
the standard system default multibyte character set, but the default can be different
at your site. For more information, see Multibyte Character Sets on page M-44)
string
is a series of single-byte or double-byte characters.
Considerations—String Literals
As indicated in the syntax diagram, you must enclose string literals in either single
(') or double (") quotation marks, although the value of the string literal does not
include the quotation marks. To specify the delimiter character within the literal, use
two consecutive quotation marks.
A string literal can be as long as a character column. For limits on data length, see
Limits on page L-6.
To specify a long string literal in SQLCI, separate the literal into several smaller
string literals and use line-continuation characters (&) to connect them.
In host programs, rules for breaking string literals across lines depend on the
language being used.
In string comparisons, lowercase letters are not equivalent to the corresponding
uppercase letters.
Strings associated with double-byte character sets must contain an even number
of bytes. An error occurs if such a string contains an odd number of bytes.
You should not mix single-byte and double-byte characters in the same string, but
NonStop SQL/MP does not prevent you from doing so.
NonStop SQL/MP does not perform data validation to ensure that characters in a
string literal belong to the character set associated with the string literal. For
example, this string literal is valid:
_KANJI"abcdef"