About type trees

A type tree (.mtt) defines the entire contents of at least one input that you intend to map or one output you intend to map. A type tree is the mechanism for defining each element of your data. Similar to a data dictionary, a type tree contains a collection of type definitions.

Because data definitions are defined in a type tree, you should be familiar with the specifications that define your data before attempting to create one.

A data file is a simple example. The file is made up of records and each record is made up of fields. In this case, there are three kinds of data objects: a file, a record, and a field.

In a file of records, think of the data in terms of the three data objects. For example, one type defines the entire file; another type defines the entire record contained in that file. Other types in the same type tree define the data fields of the record.

A type tree has three datatype classifications: group, category, and item. The tree has a root type and other types are connected to the tree through branches of the tree. The root type is the base type from which all other types stem, representing the data objects of all types in the tree. "Root" is the default name when creating a new type tree, however, you can modify any type name.

When viewing a type tree from the Extender Navigator, you can see the different kinds of data objects, but you cannot see the layout or composition of the data. For example, by looking at the tree, you cannot tell that a record consists of fields.

Types within a type tree are listed in alphabetical order by default. To change the order of appearance for new types, modify the root type properties.

IBM Transformation Extender supports large metadata. It supports more than 65,536 types and 65,536 components in a single type tree. To reduce the number of types that maps that use large type trees must process, trim these type trees so that they contain only the required types.