A Grammar-based Data-flow Analysis to Stop Deforestation (1994)
| Venue: | CAAP'94: Proceedings of the 19th International Colloquium on Trees in Algebra and Programming, volume 787 of Lecture Notes in Computer Science |
| Citations: | 5 - 0 self |
BibTeX
@INPROCEEDINGS{Sørensen94agrammar-based,
author = {Morten Heine Sørensen},
title = {A Grammar-based Data-flow Analysis to Stop Deforestation},
booktitle = {CAAP'94: Proceedings of the 19th International Colloquium on Trees in Algebra and Programming, volume 787 of Lecture Notes in Computer Science},
year = {1994},
pages = {335--351},
publisher = {Springer-Verlag}
}
Bookmark
 |
 |
 |
 |
 |
 |
OpenURL
Abstract
. Wadler's deforestation algorithm removes intermediate data structures from functional programs, but is only guaranteed to terminate for treeless programs. Chin has shown how one can apply deforestation to all first-order programs: annotate non-treeless subterms and apply the extended deforestation algorithm which essentially leaves annotated subterms untransformed. We develop a new technique of putting annotations on programs. The basic idea is to compute a finite grammar which approximates the set of terms that the deforestation algorithm encounters. The technique extends Wadler's and Chin's in a certain sense. 1 Introduction Modern functional programming languages like Miranda 1 [Tur90] lend themselves to a certain elegant style of programming which exploits higher-order functions, lazy evaluation and intermediate data structures; Hughes [Hug90] gives illuminating examples. While this programming style makes it easy to read and write programs, it also results in inefficient pro...
