用户名: 密码: 验证码:
B枚hm theorem and B枚hm trees for the -calculus
详细信息查看全文 | 推荐本文 |
摘要
Parigot鈥檚 -calculus (Parigot, 1992聽) is now a standard reference about the computational content of classical logic as well as for the formal study of control operators in functional languages. In addition to the fine-grained Curry-Howard correspondence between minimal classical deductions and simply typed -terms and to the ability to encode many usual control operators such as call/cc in the -calculus (in its historical call-by-name presentation or in call-by-value versions), the success of the-calculus comes from its simplicity, its good meta-theoretical properties both as a typed and an untyped calculus (confluence, strong normalization, etc.) as well as the fact that it naturally extends Church鈥檚 -calculus. Though, in 2001, David and Py proved聽 that B枚hm鈥檚 theorem, which is a fundamental result of the untyped -calculus, cannot be lifted to Parigot鈥檚 calculus.

In the present article, we exhibit a natural extension to Parigot鈥檚 calculus, the-calculus, in which B枚hm鈥檚 property, also known as separation property, can be stated and proved. This is made possible by a careful and detailed analysis of David and Py鈥檚 proof of non-separability and of the characteristics of the -calculus which break the property: we identify that the crucial point lies in the design of Parigot鈥檚 -calculus with a two-level syntax. In addition, we establish a standardization theorem for the extended calculus, deduce a characterization of solvability, describe -B枚hm trees and connect the calculus with stream computing and delimited control.

© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号

地址:北京市海淀区学院路29号 邮编:100083

电话:办公室:(+86 10)66554848;文献借阅、咨询服务、科技查新:66554700