Fusing Logic and Control with Local Transformations: An Example Optimization

Abstract

Abstract programming supports the separation of logical concerns from issues of control in program construction. While this separation of concerns leads to reduced code size and increased reusability of code, its main disadvantage is the computational overhead it incurs. Fusion techniques can be used to combine the reusability of abstract programs with the efficiency of specialized programs. In this paper we illustrate some of the ways in which rewriting strategies can be used to separate the definition of program transformation rules from the strategies under which they are applied. Doing so supports the generic definition of program transformation components. Fusion techniques for strategies can then be used to specialize such generic components. We show how the generic innermost rewriting strategy can be optimized by fusing it with the rules to which it is applied. Both the optimization and the programs to which the optimization applies are specified in the strategy language Stratego. The optimization is based on small transformation rules that are applied locally under the control of strategies, using special knowledge about the contexts in which the rules are applied.