Simon Thompson and Claus Reinke
Refactoring Functional Programs
Technical Report 16-01, Computing Laboratory, University of Kent at Canterbury, October 2001. (PDF)

Abstract: Refactoring is the process of redesigning existing code without changing its functionality. Refactoring has recently come to prominence in the OO community. In this paper we explore the prospects for refactoring functional programs.

Our paper centres on the case study of refactoring a 400 line Haskell program written by one of our students. The case study illustrates the type and variety of program manipulations involved in refactoring. Similarly to other program transformations, refactorings are based on program equivalences, and thus ultimately on language semantics. In the context of functional languages, refactorings can be based on existing theory and program analyses. However, the use of program transformations for program restructuring emphasises a different kind of transformation from the more traditional derivation or optimisation: characteristically, they often require wholesale changes to a collection of modules, and although they are best controlled by programmers, their application may require nontrivial semantic analyses.

The paper also explores the background to refactoring, provides a taxonomy for describing refactorings and draws some conclusions about refactoring for functional programs.