Gradual Refinement: Blending Pattern Matching with Data Abstraction
Meng Wang‚ Jeremy Gibbons‚ Kazutaka Matsuda and Zhenjiang Hu
Abstract
Pattern matching is advantageous for understanding and reasoning about function definitions, but it tends to tightly couple the interface and implementation of a datatype. Significant effort has been invested in tackling this loss of modularity; however, decoupling patterns from concrete representations while maintaining soundness of reasoning has been a challenge. Inspired by the development of invertible programming, we propose an approach to abstract datatypes based on a right-invertible language RINV - every function has a right (or pre-) inverse. We show how this new design is able to permit a smooth incremental transition from programs with algebraic datatypes and pattern matching, to ones with proper encapsulation (implemented as abstract datatypes), while maintaining simple and sound reasoning.
Details
| Book Title |
Mathematics of Program Construction |
| Editor |
Jules Desharnais |
| Note |
See revised journal version "Refactoring Pattern Matching" |
| Series |
LNCS |
| Volume |
6120 |
| Year |
2010 |
Links
DOI (10.1007/978-3-642-13321-3_22)
Related pages
|
People |
|
|
Projects |
|
|
Activities |