Copy link to this section Collation TypesĬollation can be specified as a keyword in the definition of a field/property or the definition of an index.Ĭollation can be specified by applying a collation function to a field name in a query clause. You can use the ObjectScript Sorts After operator to determine the relative collation sequence order of two values.Īlso see “ SQL Collation and NLS Collations,” later in this chapter. However, because a timestamp is in ODBC format, the string collation is the same as chronological sequence, if leading zeros are specified.Ī string expression (such as those using the scalar string functions LEFT or SUBSTR) makes its result collation EXACT.Īny comparison of two literals uses EXACT collation. This conversion is for the purposes of collation only in Caché SQL strings are usually displayed in uppercase and lowercase letters, regardless of the collation applied, and the length of a string does not include the appended space character.Ī timestamp is a string, and therefore follows the current string collation. SQLUPPER collation converts all letters to uppercase (for the purpose of collation), and appends a space character to the beginning of the string. The default string collation is SQLUPPER this default is set for each namespace. This creates an order such as the following: null, A, AA, AAA, AAB, AB, B. String collation orders strings by collating on each sequential character. Numeric collation orders numbers based on the complete number in the following order: null, then negative numbers from largest to smallest, zero, then positive numbers from smallest to largest. There are two fundamental collations: numeric and string. Table Field/Property Definition CollationĬollation specifies how values are ordered and compared, and is part of both Caché SQL and Caché Objects.
0 kommentar(er)
