Defining Record Type Filters

A record type filter defines criteria used to determine if a record within a data file will use the particular record type associated with the filter. A relational table that is based on a record type will only contain rows corresponding to records that evaluate true against the record type filter.

A record type filter is edited in a right-hand pane of the Designer window, and the filter itself is depicted as a tree. The top level of the tree is a list of AND-based factors; each factor must be TRUE for the filter to include the record with the record type.

Each factor can be a single condition, or it can be an OR list containing multiple conditions, one of which must be TRUE for the factor to be true.

A condition consists of a condition item (an 88-level item) and a desired result: TRUE when the condition item is TRUE, or FALSE when the condition item is not true (effectively, a NOT operation).

The following example shows a filter with two factors: one factor contains a single condition (CASH = True) and the other factor is an OR list that contains two conditions (FARGO-FREIGHT = True and WILD-GOOSE-EXPRESS = True). Both factors must evaluate TRUE for the record to be used in the record type: that is, CASH must evaluate to TRUE, and either FARGO-FREIGHT or WILD-GOOSE-EXPRESS must evaluate TRUE.