 |
HP COBOL
DBMS Database Programming Manual
2.2.4 DB-KEY
The reserved word DB-KEY names a database register. It consists of a
quadword COMP item represented by PIC S9(18) USAGE IS COMP. The
execution of the COBOL data manipulation language (DML) statements
FETCH, FIND, and STORE causes the Database Control System (DBCS) to
place a value in this register.
The DB-KEY special register contains three values. To access the
individual values, move the DB-KEY register to a record as follows:
01 DATABASE-KEY
02 LINE-NUMBER PIC 9(4) USAGE IS COMP.
02 PAGE-NUMBER PIC 9(9) USAGE IS COMP.
02 AREA-NUMBER PIC 9(4) USAGE IS COMP.
|
2.2.5 DB-UWA
The reserved word DB-UWA names a database register. It consists of 108
alphanumeric characters represented by PIC X(108) USAGE IS DISPLAY. The
Database Control System (DBCS) makes data items available to your
program through the DB-UWA record delivery area. The DB-UWA register
can be used with callable DBQ routines. Procedure Division statements
can access the values in this register; however, only the DBCS can
change the value. For more information, refer to the the discussion of
the DBO/WORK_AREA command in the Oracle CODASYL DBMS documentation set.
Chapter 3
Data Division
The logical and physical concepts that apply to the Data Division in
any COBOL program also apply to your database programs. The
HP COBOL Reference Manual presents the general formats for Data Division entries
and clauses, describes their basic elements, and lists rules of use.
Information of special interest to the database programmer will be
found here.
The Data Division defines the data processed by your COBOL program in
both physical and logical terms. It also specifies whether the data is
contained in files, a database, or Oracle CDD/Repository, or is developed only
for local use in your program.
The Subschema Section specifies data contained in a database or
Oracle CDD/Repository. The Working-Storage and Linkage Sections contain data
description entries, which describe characteristics of data developed
for use in your program.
3.1 DATA DIVISION General Format and Rules
Function
The Data Division describes data the program creates, receives as
input, manipulates, and produces as output.
Syntax Rules
- The Data Division follows the Environment Division.
- The reserved words DATA DIVISION, followed by a separator period,
identify and begin the Data Division.
General Rules
- The Data Division has six sections. These sections must be in the
following order:
SUB-SCHEMA SECTION.
FILE SECTION.
WORKING-STORAGE SECTION.
LINKAGE SECTION.
REPORT SECTION.
SCREEN SECTION. (Alpha, I64)
Subschema Section
- The Subschema Section names the subschema you want to use. It
describes a logical view of the database as it is to be accessed by the
COBOL program. It begins with the Subschema Section header containing
the reserved words SUB-SCHEMA SECTION followed by a period (.)
separator character.
- The Subschema entry follows the Subschema Section header.
- The Subschema entry consists of a level indicator (DB), a
sub-schema-name, the reserved word WITHIN, a schema-name, and an
optional database clause.
- A DB entry specifies the following:
- Which subschema contains the descriptions of the data to be made
available to the COBOL program
- Which schema contains the subschema definition in Oracle CDD/Repository
- The name of the database that the program is to be run against
- The stream name
- A run unit can declare only one Subschema (DB) entry.
- Keeplist entries follow the Subschema (DB) entry. A keeplist is a
table containing an ordered list of database key values.
- Each database key value represents a record occurrence in the
Oracle CODASYL DBMS database. A keeplist can contain the same database key
value any number of times. Any number of keeplists can contain the same
database key value any number of times.
- A database key is added to the end of a keeplist by the KEEP or the
FIND ALL statement and selectively removed by the FREE statement. One
FIND ALL statement can add multiple database keys to a keeplist.
- The number of entries in a keeplist is called the
cardinality of the keeplist. A keeplist with zero
entries is called an empty keeplist. The order of the
entries in the keeplist reflects their sequence of insertion. As
entries are added, the cardinality of the keeplist increases. As
entries are removed, the cardinality decreases.
- A COBOL program can reference a keeplist by using a database key
identifier.
- The Database Control System (DBCS) creates and maintains all
keeplists.
- No other run unit can update a record whose database key value is
in a keeplist. The DBCS locks each record that is in a keeplist. The
record remains locked until the program frees it or until the program
executes a COMMIT (without the RETAINING option) or ROLLBACK statement,
or until the program terminates.
File Section
Working-Storage Section
Linkage Section
Report Section
Screen Section (Alpha, I64)
The last five COBOL sections are described and discussed in the
HP COBOL Reference Manual chapter on Data Division.
Additional References
Refer to the HP COBOL Reference Manual for:
- REPORT clause
- VALUE IS clause
- CALL statement
- User-defined words
DB (Subschema Description)
Function
The Subschema entry allows a program to access a subschema in
Oracle CDD/Repository under the schema name.
sub-schema-name
references a subschema name in Oracle CDD/Repository under the schema name. See
Technical Notes for more information.
schema-name
references a schema name in Oracle CDD/Repository. See Technical Notes for more
information.
database-name
represents a complete or partial file specification defining the
database occurrence. A database occurrence is a root file. At run time,
database-name points the Database Control System (DBCS) to the root
file. See Technical Notes for more information.
stream-name
represents the name of the stream. See Technical Notes for more
information.
General Rules
- The compiler finds the schema-name within Oracle CDD/Repository
under your CDD$DEFAULT and includes these sub-schema-name
definitions in your program: realm names, set names, record names,
database data-names. These definitions are user-defined words. Other
definitions include the database registers DB-CONDITION,
DB-CURRENT-RECORD-NAME, DB-UWA, DB-CURRENT-RECORD-ID, and DB-KEY.
Subschema definitions are also called database objects.
- Database objects implicitly have the global attribute. Therefore,
any program contained within a program declaring the DB statement can
access any of the declaring program's database objects.
- Only one DB statement is allowed for each separately compiled
program; however, more than one DB statement is allowed in the run unit
if its separately compiled programs define identical DB statements, or
if each separately compiled program uses a stream.
- Database objects either explicitly or implicitly defined in the
subschema section are external objects by default. However, these
objects can only be referenced by a program defining the DB statement
or a program contained within that program.
- When you compile a program containing a DB statement, the resulting
.OBJ file includes the subschema date and time stamp assigned to the
sub-schema-name when it was inserted in Oracle CDD/Repository. At run
time, the DBCS finds the root file and binds the program to the
subschema identified by sub-schema-name. If the program does
not contain the same date and time as the subschema identified by
sub-schema-name, then compile-time subschema definitions may
not be the same as the run-time definitions; therefore, an exception
condition occurs. This security check prevents your program from
accessing the database with obsolete subschema definitions.
- If you include the FOR database-name clause, the DBCS determines if
database-name is a logical name or a file specification. If it
is a partial file specification, the default file type is .ROO.
- If you omit the FOR database-name clause, the DBCS creates a
database-name from schema-name. The DBCS determines
if schema-name is a logical name. If it is, a translation is
done to determine the database-name. If schema-name
is not a logical name, the DBCS creates a database-name by
appending .ROO to the first nine characters of schema-name.
- If a separately compiled program in the run unit uses a stream,
then all separately compiled programs in the same run unit must use
streams. Otherwise, a run-time error occurs.
Technical Notes
- Table 3-1 describes the representation of data types in the
Oracle CODASYL DBMS data definition language (DDL) compiler and the
HP COBOL for OpenVMS Alpha, I64, or VAX compiler. Additionally,
this table lists all data types that can be specified using the DDL
data definition language with the corresponding COBOL data item
picture. Note that COBOL does not have an equivalent specification for
some data types.
Table 3-1 HP COBOL and DDL Utility Data Type Equivalences
| DDL |
COBOL |
|
SIGNED BYTE
|
(a)
|
|
UNSIGNED BYTE
|
(a)
|
|
SIGNED WORD
|
S9(4) COMP
|
|
UNSIGNED WORD
|
(b)
|
|
SIGNED LONGWORD
|
S9(9) COMP
|
|
UNSIGNED LONGWORD
|
(b)
|
|
SIGNED QUADWORD
|
S9(18) COMP
|
|
UNSIGNED QUADWORD
|
(b)
|
|
SIGNED OCTAWORD
|
(a)
|
|
UNSIGNED OCTAWORD
|
(a)
|
|
FLOATING
|
COMP-1
|
|
FLOATING COMPLEX
|
(a)
|
|
D_FLOATING
|
COMP-2 (/FLOAT=D on Alpha, I64,
VAX)
|
|
D_FLOATING COMPLEX
|
(a)
|
|
G_FLOATING
|
COMP-2 (/FLOAT=G on Alpha, I64)
|
|
G_FLOATING COMPLEX
|
(a)
|
|
H_FLOATING
|
(a)
|
|
H_FLOATING COMPLEX
|
(a)
|
|
CHARACTER l
|
X(l)
|
|
UNSIGNED NUMERIC l s
|
9(m)V9(n)
|
|
LEFT SEPARATE NUMERIC l s
|
S9(m)V9(n) LEADING SEPARATE
|
|
LEFT OVERPUNCHED NUMERIC l s
|
S9(m)V9(n) LEADING
|
|
RIGHT SEPARATE NUMERIC l s
|
S9(m)V9(n) TRAILING SEPARATE
|
|
RIGHT OVERPUNCHED NUMERIC l s
|
S9(m)V9(n) TRAILING
|
|
ZONED NUMERIC l s
|
(a)
|
|
PACKED DECIMAL l s
|
S9(m)V9(n) COMP-3
|
Legend:
l---The total number of digits for the item.
s---The decimal offset to 1.
(
a
)
---COBOL has no equivalent for this data
type. A fatal diagnostic is issued for such an item, which is part of a
subschema record description. The compiler treats this item as if it had
been specified as an alphanumeric data item that occupies the same
number of
bytes.
(
b
)
---COBOL has no exact equivalent for this data type. A
warning diagnostic is issued for such an item, which is part of a
subschema record description. The compiler treats this item as if it had
been specified as the equivalent unsigned COMP item.
The method for describing the assumed decimal point is different in
the two products. In a COBOL picture, the decimal position is directly
indicated by the symbol V or implied by the symbol P. In DDL, scaled
numbers are specified by two integers. The first integer represents the
total number of decimal digits that the item represents. The second
integer represents the decimal offset to the first integer.
For
example, the COBOL data item described by PIC 9(4)V99 is equivalent to
the DDL entry TYPE IS UNSIGNED NUMERIC 6 -2. Similarly, the DDL entry
TYPE IS LEFT SEPARATE NUMERIC 6 2 is equivalent to the COBOL
description PIC S9(6)PP SIGN IS LEADING SEPARATE. The items described
using the CDD INDEXED FOR COBOL BY clause become COBOL index-names.
- Schema-name, sub-schema-name,
database-name, and stream-name can be a nonnumeric
literal or a COBOL word formed according to the rules for user-defined
names. Database-name represents a complete or partial file
specification. If any of these names is not a literal, the compiler:
- Translates hyphens in the COBOL word to underline characters
- Translates lowercase alphabetic characters to uppercase
- Treats the word as if it were enclosed in quotation marks
- Most of the information described in the remaining Technical Notes
is included in the subschema map. See Chapter 1, Program Organization and Structure, for information on
the /MAP qualifier.
- You can use this Database Operator (DBO) utility command to display
date and time stamp information on schema-name and
sub-schema-name:
|
