Data Transformation Engine Type Designer Reference Guide

Chapter 9 - Partitioning Partitioning Types
♦ It is easy to read this map rule and understand the mapping function being
performed. For example, if the state belongs to the list of states in the eastern
region, execute the MapEast functional map.
♦ The partitioning method is easier to maintain. If a value for State is added or
moves from one region to another, it can be easily changed in the type tree
and automatically reflected in any mapping rules that reference the partitioned
object.
♦ Partitioning using a restriction list used with the Ignore Case setting
eliminates the need for
PROPER, LOWERCASE, or UPPERCASE functions to
compare each state with a literal.
Partitioning Types
When data objects of different types appear in the same place in the data, the
types must be distinguishable. This means that the data needs to be
distinguishable by their definitions in the type tree.
When the data object at any given point in the data may belong to any of a
number of different types, there must be some way to tell the difference between
them. To do this, you create a type and define a mutually exclusive subtype for
each data object that may appear in the same place in the data. Once the
subtypes are created, the subtypes also need to be distinguishable. Subtypes are
distinguishable based on a value in the data or in the syntax of the different types.