Use the Properties view to set the transformation properties of
generalization objects before you transform a logical data model into a physical
data model.
Prerequisite: Before you modify the properties of logical data
model objects, you must open a logical data model in the logical data model
editor.
To set generalization transformation properties for logical-to-physical
data model transformation:
- Select a generalization object in the Data Project Explorer, or
in a diagram.
- Click to open the
Properties view, if the view is not already open.
- Optional: On the General tab in the
Properties view, modify the Transform as property by
selecting one of the following options:
Option |
Description |
Separate table |
Select to specify that subtypes
and supertypes are transformed to separate tables. |
Roll up |
Select to specify that subtype
attributes are transformed to be contained in the supertype table. |
Roll down |
Select to specify that supertype
attributes are transformed to be contained in subtype tables. |
- Optional: On the General tab
in the Properties view, modify the Defining Attribute property
by selecting a supertype entity attribute from the from the list. The attribute
that you select is defined as a type discriminator for the generalization.
When you transform to a physical model, a typed table is added only when no
type discriminator is defined on the generalization and the Roll
Up option is selected in the Transform as field.
- Optional: Modify the order of specializations in a
generalization hierarchy. The order of specializations is important
for roll-up denormalization of a generalization.
- Select the entity that contains the generalization object in
the Data Project Explorer, or in a diagram.
- On the Generalizations tab, modify the
order of the sub types that are contained in each generalization object.
The next time that you transform the logical data model using
the Roll Up option, the column order that you specified is preserved in the
resulting physical data model table.
- Save the logical data model.