Attribute definition table (ADT)
An attribute definition table defines for each attribute
a token to reference the attribute by, sets of valid operations
and filters that may be performed on the attribute, and the
attribute's value structure.
Attribute definition table
get add remove
get replace replaceWithDefault
eq le ge
In the example, the attribute gid (token: 1) may be operated upon
by get, add,
or remove and evaluated by the filters
eq and present;
paired values are structured as integer sets.
1st column: name
made up of characters [a-z_]* and used by
the Server API to identify the attribute.
2nd column: token
an integer used by the Server API and OSAs
to reference the attribute. Identifiers can be
in any order and in any positive range (negative
integers are reserved for error conditions; zero
is reserved for the NULL token).
In languages (such as Tcl) which have only the string
data type, the token is identical to the name.
3rd column: valid operations
all operations that may be performed on the attribute.
Can include: get, replace,
(create is always allowed to include any
attribute, so it is implied to be listed for every attribute.)
4th column: valid filters
all filters that may process the attribute.
Can include: eq, le, ge, subset, superset, present, intersect.
5th column: value structure
Identifies for OSAs and Server API how the value(s)
paired with attribute should be interpreted.
In Tcl every data type is a string, so the CDT does not
specify the data type for this pair of values. In C/C++, strings are
represented as char *, integers as long, and
floating point numbers as doubles.
For more information, see
``Example Attribute Definition Tables (ADT)''.
Operation definition table (ODT)
Creating class definition tables
© 2003 Caldera International, Inc. All rights reserved.
SCO OpenServer Release 5.0.7 -- 11 February 2003