Complex Attribute Manipulation in TGGs with Constraint-Based Programming Techniques

Abstract

Model transformation plays a central role in Model-Driven Engineering (MDE) and providing bidirectional transformation languages is a current challenge with important applications.  Triple Graph Grammars (TGGs) are a formally founded,  bidirectional model transformation language shown by numerous case studies to be quite promising and successful.  Although TGGs provide adequate support for structural aspects via object  patterns in TGG rules, support for handling complex relationships between different attributes is still missing in current implementations.  For certain applications, such as bidirectional model-to-text transformations, being able to manipulate attributes via string manipulation or arithmetic operations in TGG rules is vital.  Our contribution in this paper is to formalize a TGG extension that provides a means for complex attribute manipulation in TGG rules.  Our extension is compatible with the existing TGG formalization, and retains the "single specification'' philosophy of TGGs.