ice.NET Key Concepts: Relationship Types

Relationship Types types describe the characteristics of relationships. Relationship Types share a set of common aspects:

• Relationship Types (like all schema elements) are organized in packages.
• Every Relationship Types is identified by its Relationship Types ID. The ID does not change and exists until the object type is deleted. The ID is unique within the scope of the current system context (i.e. normally within a database or an XML document).
• Furthermore, every Relationship Type is identified by its Relationship Type name. This name has to be choosen by the developer who creates the Relationship Types and should reflect the semantics of the Relationship that conform to this type. The Relationship Type name has to be unique. The character set of Relationship Type names is limited to the capability of the XML QName specification in order to support programming concepts like XML Schema or code generators.
• Every Relationship Type has a display name and a description. The display name is used for UI and other visualization purposes and provides the possibility to assign a user-friendly name to the object type. The display name does not have to be unique (though it is recommended to ensure uniqueness within a reasonable scope). It can contain any unicode characters (e.g. umlauts or scandinavian special characters).
• Relationship Types reference Object Types in From and To direction. Relationships that conform to a specific Relationship Type can connect objects that are of these types (or their subtypes).
• Both From and To directions have a Cardinality assigned. This cardinality defines how many objects can be connected usinf relationships of this type in the specific direction.