It is often the case in practical information systems that information is held in a denormalized format. For example, an employee name may be held in several tables. It is desirable (or essential) that if the name is masked in one column then the other columns in which the information is held are also updated with an identical value. This is called a synchronized update and in the Data Masker software it is implemented by a Cascade rule.
Cascade rules are designed to synchronize any data changes made to a table column with columns in other tables. Thus implementing a Cascade rule requires the knowledge of a join condition between the two tables. The Data Masker software uses this join condition to ensure the appropriate rows are updated correctly.
Cascade rules are added as dependent rules on Substitution Rules. As the Substitution rule executes and replaces the information in the column it is updating, simultaneous updates will be performed by the Cascade rule on the linked columns. There can be any number of Cascade rules on any column in any table in a schema.
Cascade rules do use options - these options are the join condition between the two tables. The Cascade rule uses the join condition to identify the row in the child table to synchronize. The join condition can be the column currently being updated or it can be another column (or set of columns) entirely. After configuring the rule is is possible to edit the join columns for that rule.
For example consider the two tables:
TABLE PARENT COLUMN IDNUM number(10) COLUMN NAME varchar2(40) TABLE CHILD COLUMN IDNUM number(10) COLUMN NAME varchar2(40)
Assume the NAME column must be synchronized between the two tables and the join condition is PARENT.IDNUM=CHILD.IDNUM. If a Substitution rule is placed on the PARENT.NAME column and a Cascade rule is added to the Substitution rule then each CHILD.NAME column will receive values identical to the ones in the PARENT table where the join condition matches.
Important Note: It is important to realize that if there are IDNUM values in the CHILD table that are not present in the PARENT table, the Cascade rule will have no effect on those rows. These columns will NOT be masked.
It is also possible to synchronize masking changes to the PARENT.IDNUM column even thought it is part of the join condition. The Cascade rule is implemented in an identical manner and the Data Masker software will apply the updates automatically. However, if CASCADE rules are present it is essential that the values being used to update the PARENT table IDNUM column are not already present in the CHILD.IDNUM column. The Cascade rule cannot tell the difference between data previously present in the table and newly added data and referential integrity can be lost.
Cascade rules are always added as dependents to a Parent Substitution Rule. To add a Cascade rule select the substitution rule which will act as the parent. The Add Cascade Rule... buttion will activate. Press that button to add the column to synchronize and the join conditions for the Cascade rule.
Note that it is possible to apply any number of Cascade rules to a Substitution rule. The example above shows an existing Cascade rule on the CHILD2.NAME column.