Customize - Data Types

You are here: Customizing Quality Knowledge Bases>Overview of Customize Features>Data Types

DataFlux Data Management Studio 2.5: User Guide

Customize - Data Types

A data type defines a category of content, such as names or addresses. Data types have sub-fields (tokens) that correspond to logical or semantic units in the data.

For example, a data type intended to describe a person's name might contain the following tokens:

Prefix (Mr., Mrs., and so on)
First name
Middle name
Last or family name
Suffix (Jr., IV, and so on)

Data types can be inherited down the locale hierarchy.

Using Customize, you can add, modify, and delete data types.

Create a Data Type

You can create data types from the Customize Data Types and Definitions pane, or from the context menu within the tree. Your new data type will be assigned to the selected locale. If you use the context menu, this is the locale in which the context menu is activated.

In the Data Types and Definitions pane, specify a name for the new data type. There are some rules when naming data types and tokens.

Data Type Names Rules:

Names are case-insensitive for the purpose of comparison.
A name may not contain non-alphanumeric characters, including commas.
Within the locale, no two data types may have the same name.
A locale may have a data type of the same name as one in an ancestor locale; in this case, the ancestor's data type will be overridden. For example, the ancestor's data type will no longer be available for use within the child locale.

Note Note: You must also specify at least one token.

Token Name Rules:

Names are case-insensitive for the purpose of comparison.
A name may not contain non-alphanumeric characters, including commas.
Within the data type, no two tokens may have the same name.
Tokens in different data types, even data types with the same name in different locales, have no relation to each other.

Tokens can be added, edited, deleted, or rearranged. The order of tokens usually corresponds to the typical order of occurrence in real data, but it is not required.

Note Note: If you delete a token, extraction schemes or sought categories that reference that token are removed as well. As a result, you could potentially make a definition invalid by removing a token. Customize marks these items with a red "X".

You can also add comments in the Notes section (optional).

Copy a Data Type

You can copy a data type from the Data Types and Definitions pane or from the context menu within the tree. When you copy a data type, all of the definitions are copied. Each individual copy belongs to the Locale where its original belongs.

For example, if you select a data type that comes from L1, that has two definitions under it (A and B) then definition A belongs to EN and definition B belongs to ENUSA. The copied data type is placed in L1. The copy of definition A is placed in EN and the copy of definition B in ENUSA.

If there is already a data type with the same name within the target locale, the copy will be named similarly to the original, with a numeric suffix added (for example, Copy of Address 2). You can rename the copy. The same principle applies to the definitions copied along with the data type.

You also have the option to copy the Data Type without any definitions.

Delete a Data Type

Select the data type you want to delete and click the Delete icon from the Data Types and Definitions toolbar. You can also delete a data type from the context menu within the tree.

Important: Only items that are not in use by any definitions may be deleted.

Edit a Data Type

To edit a data type, click the data type you want to edit. Click the Properties or Notes tab to make changes.

You can rename a data type from the Properties tab, using the data type name rules. Any existing definitions that use the renamed item, either in the current locale or its children, are notified of the new name.

In the Properties tab, click the icons to add, edit (rename), or delete tokens. Refer to the token name rules, which are enforced by error messages. If you delete a token from a data type, any definition that uses that token is updated so that the token is no longer used. However, if there is a definition that only uses one token from the data type, that token may not be deleted from the data type, regardless of how many other tokens are left in the data type.

For example:

Data Type A has tokens First, Middle, and Last
Definition B uses First and Last from Data Type A
Definition C uses First from Data Type A
No other Definitions use Data Type A

Deletion of First from Data Type A is not allowed, because Definition C uses it as its only token, and deleting the token would make Definition C both useless and invalid. Middle can be deleted from the data type, and this has no effect since no definition uses it. Last can also be deleted, and Definition B will be updated to no longer use it.