SQL/MX 2.x Reference Manual (G06.24+, H06.03+)

SQL/MX Language Elements

HP NonStop SQL/MX Reference Manual—523725-004

6-5

Collations

A collation is an object that contains rules for a collating sequence (the sequence in

which characters are ordered for sorting), case, and character class and character

string equivalence.

Every character set has a collation. See Character Sets on page 6-3. To be compared,

character strings must be from the same character set. When two strings are

compared, the comparison is made with a temporary copy of the shorter string that has

been padded on the right with blanks to have the same length as the longer string.

You create an SQL/MP collation with the SQL/MP CREATE COLLATION statement. A

collation name must be a Guardian name. For more information, see the SQL/MP

Reference Manual.

You cannot use SQL/MP collations on SQL/MX tables and you cannot create collations

for SQL/MX tables. SQL/MX Release 2.x supports only the DEFAULT collation.

DEFAULT is based on binary ordering and is the default collating sequence for CHAR

and NCHAR data types.

Binary collation is a collating sequence based on binary ordering. A binary collation

comparison of two equal length strings, s1 and s2, compares the values of the

corresponding characters of s1 and s2 until it finds a difference. If a difference is found

and the differing character value of s1 is less than that of s2, s1 is considered to come

before s2. If there is no difference, s1 is considered equal to s2. Otherwise, s1 comes

after s2. If the two strings are not equal in length, the shorter one is padded with

spaces in the corresponding character set.

Comparison of two identical strings associated with the same character set will always

be evaluated as equal by the DEFAULT collation. However, the DEFAULT collation

does not necessarily yield sorting orders that are culturally correct.