Quality Driven Transformation Compositions for Object Oriented Migration

Ying Zou, Kostas Kontogiannis

Abstract: Reengineering legacy software systems to object oriented platforms has received significant attention over the past few years. In this paper, we propose a goal driven software migration framework that aims to identify and extract a quality object model from a procedural system and to generate quality object oriented code. The framework is composed of analysis tools, transformation rules, and non-functional requirement models for the target migrant system. Specifically, to facilitate the design and development of such goal driven migration framework, source code transformation rules are associated with a degree of belief that they contribute towards enhancing a desired property for the target system. The migration process applies a search algorithm that is guided by the source code analysis to select a transformation sequence that has the highest likelihood of yielding such a target system. The migration of a selected set of gnu AVL libraries to a new object oriented platform is presented as a proof of concept for the proposed technique.