An identifier has no intrinsic meaning. It is used to give a
symbolic name to an entity within the Enterprise Database Server.
An identifier can be written as shown in the following diagram:
<double-byte identifier>
┌◄───────────────────────────────────────────┐
│ │
──<SDO>───┴────/14\───────<16-bit character>───────────┴─────<EDO>─────┤<identifier>
──<letter>─┬────────────────────────────────────────┬──────────────────┤
├───────────────────────────┬─┬─<letter>─┤
│ ┌◄──────────────────────┐ │ └─<digit>──┘
└─┴─/28\─┬─<letter>─────┬─┴─┘
├─<digit>──────┤
└─ hyphen (-) ─┘Examples
The following examples illustrate identifiers:
A
EMPLOYEE
AS-OF-DATE
B-1
Explanation
Start double octet (SDO) is a key found on a Kanji keyboard
that is used to signal the beginning of a 16-bit character string.
End double octet (EDO) is a key found on a Kanji keyboard that is
used to signal the end of a 16-bit character string.
A double-byte (16-bit) identifier is composed of a maximum of
14 double-byte ~(16-bit) characters embedded within the SDO or EDO
Kanji characters. A double-byte identifier cannot contain Kanji blanks
(code value A1A1) and is only supported by a COBOL85 compiler. If
you access the double-byte identifier by using any other compiler,
a syntax error results.
DASDL keywords, except for the word COMMENT, are not reserved
and can be used as identifiers. The COBOL reserved words can also
be used as identifiers; however, DASDL issues a warning message for
each COBOL reserved word encountered.
DASDL permits duplicated names provided that the names are qualifiable
when used in programs. The complete list of restrictions appears in Identifier Restrictions.
Table 3. Identifier Restrictions
|
Structure
|
Restrictions
|
|
Database names
|
The following restrictions apply:
-
Physical and logical database names must be unique.
-
Database names cannot exceed 17 characters.
-
Databases cannot share the same name under the same
usercode or under no usercode (*).
|
|
Disjoint structures
|
The following restrictions apply:
-
Names in the physical database can be duplicated only
in the logical database.
-
Names of data sets and remaps in logical databases
can only be duplicated as disjoint structures in the physical database
and within different logical databases.
-
Sets, subsets, and accesses in logical databases can
have the same name as physical disjoint structures, set names associated
with different data sets in the same logical database, and names in
different logical databases.
-
The restart data set cannot be named RECOVERY.
-
Disjoint structure names cannot exceed 17 characters.
-
In the physical database, database disjoint data sets,
sets, subsets, Accesses, remaps, and global items must not have the
same name.
|
|
Global data and global data remap items
|
These names can be duplicated only in different remaps
of the global data.
|
|
Data items
|
Items within disjoint and embedded structures must not
have the same name as
-
Any item anywhere below its immediate owner. (However,
items in variable-format parts can have the same name as identical
items in variable-format parts of the same data set.)
-
Any item owned directly by the current item's owner
or by any owner's owner, up to the disjoint level.
-
Any data set or remap name used within a logical database.
-
Data item names cannot exceed 30 characters.
|
|
Sets, subsets, Accesses
|
The following restrictions apply:
-
No two structures in the same data set can have the
same name, even if these names are otherwise qualifiable.
-
No set or subset can be called DATA.
-
Set, subset, and Access names cannot exceed 17 characters.
|
|
Structure
|
Restrictions
|
|
Data sets
|
The following terms are considered synonyms when they
are used as DASDL keywords: DATA SET, DATASET, DATA-SET, and DATA.
In this manual, the keyword DATA SET is used in the railroad diagrams.
Any of the allowed synonyms can be used.
|
