Data Rules: Enforcing Data Quality

A data rule is a definition of valid data values and relationships, which determine legal data within a table or legal relationships between tables. Data rules can be applied to tables, views, dimensions, cubes, materialized views, and external tables. They are used in many situations including data profiling, data and schema cleansing, and data auditing.

The metadata for a data rule is stored in the workspace. To use a data rule, you apply the data rule to a data object. For example, you could create a data rule called gender_rule, which could specify that valid values are 'M' and 'F'. You could then apply this data rule to the emp_gender column of the Employees table. Applying the data rule ensures that the values stored for the emp_gender column are either 'M' or 'F'. You can view the details of the data rule bindings on the Data Rule tab of the Data Object Editor for the Employees table.

A data rule can be derived from the results of data profiling, or it can be created using the Data Rule Wizard or OMB*Plus scripting commands.