Field group design: Difference between revisions

From m204wiki
Jump to navigation Jump to search
m (added See also)
Line 154: Line 154:
<ul>
<ul>
<li>[[Field group (File architecture)]]</li>
<li>[[Field group (File architecture)]]</li>
<li>[[Operations on multiply occurring fields]] </li>
</ul>
</ul>


[[Category:File management]]
[[Category:File management]]

Revision as of 23:46, 2 April 2014

Overview

Model 204 supports non-relational, de-normalized data structures. Many Model 204 sites have enjoyed significant cost and performance benefits from efficiently processing multiply occurring fields. This concept has been enhanced to introduce physical field groups that let you view and process groups of fields as a logical entity.

This feature is available in Model 204 version 7.5.

Traditional approaches to multiply occurring fields

Take the following approaches to the same data. A customer might have a number of different addresses (for example: home and postal); with the same data: house number; street; city; state; and postal code for each.

So, you could define the following fields:

HOME_HOUSE_NUMBER
HOME_STREET
HOME_CITY
HOME_STATE
HOME_POSTAL_CODE
POST_HOUSE_NUMBER
POST_STREET
POST_CITY
POST_STATE
POST_POSTAL_CODE

Or you could define the following fields

ADDRESS_TYPE
HOUSE_NUMBER
STREET
CITY
STATE
POSTAL_CODE

and, by identifying an address type for each address, have your application code maintain the occurrences of the repeatable fields so that they could be managed as sets. Add historical data to this (as the customer might move during his lifetime, and you may want to track this history) maintaining these sets becomes problematic.

As the records grow, there might also be performance issues because, although the application treats them as a group, it still must update and retrieve the fields individually.

Field group approach to multiply occurring fields

As of Model 204 version 7.5, there is a third option: you can define these fields as part of a field group. This option natively treats the fields as a single entity for retrieving and updating.

The effect of this is that you can define a Model 204 Record as a container for different relational tables, with the advantages of having them pre-joined (in the one physical record) and being able to maintain them in whatever relative order the application requires.

Note: This option is only available by setting the FILEORG parameter to X'100'.

This enhancement changes both the file DDL (as follows) and the User Language DML.

Field groups and Table B storage considerations

Every record in a FILEORG=X'100' file contains the 4-byte highest allocated field group ID. Every occurrence of a field group has a unique binary ID that occupies from two to five bytes, thus supporting up to four gigabytes of field group IDs.

As is true of all FILEORG=X'100' files, the field name representation (as held in Table A and thus in the record, are three bytes in length. Depending on the definitions of the fields within the group, this may be offset by the physical absence of fields defined as part of the group.

Defining a field group

You must define a field group before you define the fields that it contains. As you define them, keep track of the names you assign, because both field groups and fields are stored in Table A of the Model 204 file, so each name must be unique.

Syntax

The following syntax is used to define a field group:

DEFINE FIELDGROUP fieldGroupName

Where fieldGroupName specifies the name of the field group you are defining. The rules for the field group name are the same as for fields.

Defining a field inside a field group

Use the following syntax to define the field(s) for a field group:

Syntax

DEFINE FIELD fieldName WITH FIELDGROUP fieldGroupName [AND [(attribute1 attribute2 ...)]]

Where:

  • fieldName specifies a unique name for a field.
  • The WITH FIELDGROUP fieldGroupName clause identifies the containing field group. When fieldGroupName is indicated as an asterisk (*), it means that the field is included into all field groups defined for this file.
  • attribute1 attribute2 ... specifies other field attributes.

Example

To define a field group MAKEMODEL with fields MAKE and MODEL in it, you would write the following definition:

DEFINE FIELDGROUP MAKEMODEL DEFINE FIELD MAKE WITH FIELDGROUP MAKEMODEL AND ORDERED CHARACTER DEFINE FIELD MODEL WITH FIELDGROUP MAKEMODEL

Usage

Defining fields after defining the containing field group makes it possible to later add fields to an already defined field group, for example:

DEFINE FIELD MPG WITH FIELDGROUP MAKEMODEL

The keyword AND on field definitions means that a field group name can contain blanks, just like a field name. You must use the AND keyword to separate the field group name from subsequent attributes:

DEFINE FIELD COLOR WITH FIELDGROUP MAKEMODEL AND ORDERED CHARACTER

The AND is unnecessary if the FIELDGROUP fieldGroupName clause is the last field attribute:

DEFINE FIELD COLOR WITH ORDERED CHARACTER FIELDGROUP MAKEMODEL

Note: The MAKEMODEL field group must be defined before defining the field(s) and/or field group(s) that belong with it.

FIELDGROUP (FG) attribute

The FIELDGROUP attribute specifies the name of the field group that the defined field is associated with (contained in). Once you define a FIELDGROUP value for a field, you cannot redefine the FIELDGROUP value.

Syntax

FIELDGROUP [fieldgroupname | *]

The FIELDGROUP attribute does not allow:

  • Record security
  • Use for SORT or HASH file
  • 1NF file model

The FIELDGROUP attribute can be used with the STORE-NULL LITERAL attribute.

Using FIELDGROUP*

The FIELDGROUP * attribute means that the field will be included into all field groups.

The EXACTLY-ONE attribute conflicts with the FIELDGROUP * attribute.

Other attributes to consider

The default attribute for frequency of occurrence for fields in a field group is EXACTLY-ONE (instead of the default for individually defined fields of REPEATABLE). The other options REPEATABLE and AT-MOST-ONE can also be used.

Because field groups contain sets of data, the setting of a Default Value (DV), the Store Default (SD), and the Store Null (SN) settings can (for field types other than REPEATABLE) overcome the added space overhead described above.

Displaying field group definitions

Field groups can be displayed in alphabetical order as follows:

IN POLICIES DISPLAY FIELDGROUP ALL CLAIM (FIELDGROUP VEHICLE) DRIVER () VEHICLE ()

Because the "container group" must be specified before its members, the previous command's output cannot be fed back into Model 204 to define the field groups in an empty file. Instead, use the DDL option with the display command:

IN POLICIES D FIELDGROUP (DDL) ALL *** DDL FOR FILE POLICIES *** DDL REQUEST DATE/TIME: 10.102 APR 12 12.34.01 DEFINE FIELDGROUP DRIVER - () DEFINE FIELDGROUP VEHICLE - () DEFINE FIELDGROUP CLAIM - (FIELDGROUP VEHICLE)

See also