Copyright © Cloud Software Group, Inc. All Rights Reserved

Task C: Specify Header Information

An Adabas file has a specific structure that must be mapped to one or more ADA tables. The primary key is made up from a field representing the record's Internal Sequence Number (ISN), which is a unique Adabas identifier and, optionally, a second key representing the occurrence number for a periodic group (PE) or a multiple-value field (MU). A third key is required when mapping a multiple-value field within a periodic group.

Representing Adabas Data in TIBCO Object Service Broker Tables

In this example, the Adabas file layout is represented by pseudo-COBOL storage definitions.

For example:

01 AA

01 AB,PE            (Periodic Group)

   03 AC,PE

   03 AD,PE

      05 AE,MU,PE   (Multiple-Value Field)

01 AF

01 AG,MU

This structure must be mapped to four TIBCO Object Service Broker tables, each representing a different set of fields:

1.	The non-periodic groups and multiple-value fields (AA and AF):

ISN

AA

AF

2.	The fields of the periodic group fields AC and AD:

ISN

CNT1

AC

AD

3.	The multiple-value field AE within a periodic group:

ISN

CNT1

CNT2

AE

4.	The multiple-value field AG:

ISN

CNT1

AG

The ISN and count (CNT) fields are maintained by TIBCO Object Service Broker and do not exist in the Adabas data as defined fields. They are used in this example to identify the current Adabas record (ISN) and the relative occurrence number of a repeating group and or a multiple value field, so that the data can be represented as a relational view. The field names specified in the example are the default values assigned by the Table Definer but these can be altered. These fields can be specified in search arguments to retrieve a specific Adabas occurrence.

Specifying ADA Table Header Fields

The following fields are in the header segment of the ADA table definition screen:

Table	Displays the table name specified when you invoked the Table Definer. Type a new name to save the definition of the current table under a new name. For more information on how to copy TIBCO Object Service Broker objects, refer to TIBCO Object Service Broker Shareable Tools. Valid entries: A character string of up to 16 characters beginning with a letter (A - Z) or a special character ($ or #), and continuing with more letters, special characters, digits (0 - 9), or underscore characters (_). A table name starting with an @ symbol denotes a table supplied by TIBCO Object Service Broker.
Type	Displays the table type ADA, which you changed in Task B: Invoke the Table Definer.
Unit	Displays the user unit associated with the table. The unit marks a table as belonging to a particular application or to a logical unit, for example, utilities, accounting, or network control. The default unit for your user ID is specified in your TIBCO Object Service Broker user profile. Valid entries: A character string of a maximum of 8 characters. These are typically provided by your system administrator, for example, ACC.
Server ID	Identifies a group of instances of the server with common characteristics, and must match the SERVERID startup parameter specified in the server JCL. Type the ID for the server or group of servers to use when accessing the table you are defining. Refer to Supplying the Startup Parameters for more information. The SERVERID parameter can be overridden at runtime. Refer to Dynamically Changing the Parameters for more information. Valid entries: A character string up to eight characters.
DBID	Specifies the Adabas database number in which the Adabas file resides.
FILE No.	Specifies the Adabas file number of the file for which the table is being defined.
Server->EE Block	Specifies the number of rows passed to the Execution Environment in any one interaction with the server. If set to 0, a full 31 KB buffer is used. If set to the number of rows that satisfy the request, the buffer is less than 31 KB. Setting a large value for this field means that a 31 KB buffer is set when required. Refer to Taking Advantage of Adabas Features for examples of how to use this field.
ISN Assigner	Specifies whether the Internal Sequence Number (ISN) of the Adabas file is assigned by Adabas or the user application. Valid entries: SYSTEM or USER.
Use GFID	Specifies that when the table is bound, a generated Global Format ID is to be used when accessing the table.
Name of countfield	Determines how many occurrences are in a periodic group or multiple-value field. The field is used to turn on the count function. If a definition contains fields from more than one periodic group, choose the value for this field carefully.
Repeat Type	Determines the type of repeating construct to be mapped for tables that map a multiple field or a periodic group. Valid entries: MU or PE.
Mu/Special field in Grp	When set to N, the access to the Adabas file can be optimized. This field has two meanings depending on whether a periodic group (PE) or multiple field (MU) is being mapped. For a periodic group, this field can be set to N if: • The definition contains all the fields that are defined in the Adabas group • The fields are in the same order as that defined to Adabas • The group does not contain a multiple value field Otherwise, set it to Y. For a multiple value field, the field must be set to N if the MU field being mapped is not part of a PE group. If it is an MU within a PE group the field must be set to Y.
Occurs/Read	Sets the number of occurrences of a PE group or MU field that are retrieved from Adabas for every Adabas call (occurrences per read). By selecting a value based on the number of occurrences within your own data structures, you can optimize the Adabas access that is performed. Set this field to the average number of occurrences expected to be read for each row or ISN.

Specifying Optional Location Parameter Information

The location parameter segment of the screen is where you define a location parameter for the ADA table. A location parameter is required only if you want to access Adabas data through a peer server associated with a different Data Object Broker (remote node). If you do not require a location parameter, position your cursor in this segment, and then press PF4 and use the D line command to delete the parameter. If you always access the Adabas table remotely, the node from which you request the access can have either a minimal or a full definition.

Minimal Definition

A minimal definition consists of the following:

•	The table name, which must be the same at both locations

•	The location parameter, which must be the same at both locations

The name of the remote node where the full definition is located must be supplied through the use of the Default field. Data parameters are defined on the full definition, not a minimal definition.

A minimal definition with a location parameter means you always access data at a remote node. The table type specified in a minimal definition need not match the table type of the full definition on the remote node.

Full Definition

A full table definition with a location parameter indicates you can access data at either the local node or a remote node.

The table type specified in a full definition must match the data on the local node. For example, a full definition of type TDS used to access TDS data on the local node can also be used to access an ADA table with the same name on a remote node.

Specifying Optional Event Rule Information

Use the event rule segment of the Table Definition screen to provide additional controls over the access to a table, and then define event rules based on these accesses.

Event rules are always called when the table is accessed in the access type specified. All the rules that apply to a specific access are executed in the order in which they are entered in the event rule segment. They cannot access tables on a remote node.

Types of Event Rules

You can define two types of event rules, as follows:

Validation	Verify the value of an occurrence when the table is being modified, such as checking the validity of a field value.
Trigger	Cause additional processing to take place when a table is accessed. For example, a trigger rule can be used to create an audit trail or update other tables.

Field Definitions

The event rule information is entered in the scrollable event rule segment. To define event rules, position your cursor in this segment and press PF4 to bring up a definition screen. Complete the following fields:

 

Line#	Type in a line number, starting at 1 for the first line, with one event rule per line. The line numbers must be numbered consecutively.
Typ	The type of event rule. Valid entries: V – Validation Rule. No database updates are allowed during the validation process. The rule must be a function that returns Y (yes), the validation was successful, N (no), the validation was not successful, or a message explaining why it was not successful. T – Trigger Rule. A trigger rule cannot be a function or change the contents of the triggering row, and cannot use the TRANSFERCALL statement. Nested triggers are permitted.
Acc	The type of access (or manipulation) to be performed on the data, causing the event to be executed. Valid entries: Validation and Trigger Rules: W - Any write (insert, replace, delete) I - Only insert R - Only replace D - Only delete   Trigger Rules only: G - Any retrieval

See Also

•	TIBCO Object Service Broker Managing Data for more information on location parameters, event rules and minimal table definitions.

•	TIBCO Object Service Broker Programming in Rules for more information on the TRANSFERCALL statement.

Copyright © Cloud Software Group, Inc. All Rights Reserved