NonStop SQL/MP Reference Manual

NonStop SQL/MP Reference Manual—142115
C-33
The LC_TDMCODESET Section of a Collation
Definition
char is defined in the discussion of the LC_COLLATE section. lower specifies a
lowercase char; upper specifies an uppercase char.
If neither char in a pair has appeared previously in the toupper clause, the first
character in the pair is treated as a lowercase character in the collation and the
second character in the pair is treated as the corresponding uppercase character.
If you omit the toupper clause, SQL associates lowercase characters “a” to “z” with
uppercase characters “A” to “Z”.
You can define a character as the uppercase version of more than one lowercase
letter, or as the lowercase version of more than one uppercase letter. This is useful,
for example, if you define a collation for the French language, in which accented
lowercase letters often lose their accents if upshifted.
If the upper character in a pair appeared in a previous toupper pair, the latter pair
defines an upshift only. For example, in the following toupper clause (which defines
case shifting for the French e-grave character):
toupper ( e, E);\
( \d232, E;)\
( \d232, \d200) # 232 is e-grave; 200 is E-grave
the toupper pair (\d232, E) specifies that e-grave upshifts to E but does not specify
how to downshift E; the toupper pair (\d232,\d200) specifies that E-grave
downshifts to e-grave, but does not specify how to upshift e-grave.
Similarly, if the lower character in a toupper pair appears in a previous pair, the
latter pair defines a downshift only.
At least one of the characters in each toupper pair must not have appeared in an
earlier pair.
END LC_CTYPE
ends an LC_CTYPE section.
The LC_TDMCODESET Section of a Collation Definition
The LC_TDMCODESET section specifies the character set for the collation. If you do
not specify the LC_TDMCODESET clause, SQL uses the ISO88591 character set for
the collation.