NOTE: CLiarith is primarily for my practice.
CLiArith is a simple trial implementation of the interval arithmetic in Common Lisp based on the simple implementation coded by C/C++ (Simple Interval Arithmeric Library).
CLiArith is available only in SB CL.
- mskashi/kv Github
- Marco Antoniotti, Why You Cannot (Yet) Write an "Interval Arithmetic" Library in Common Lisp
To be described.
Set *read-default-float-format* to (fp-type)-float, and bind the values of (fp-type)-float-epsilon and (fp-type)-float-negative-epsilon to *positive-epsilon and *negative-epsilon respectively
(cliarith:set-float-format 'double-float)Addition:
(cliarith:+[] 0.1 0.1 0.1 0.1 0.1)
;; => #S(CLIARITH:[] :LOW 0.49999999999999994 :HIGH 0.5000000000000001)Subtraction:
(cliarith:-[] 0.1 0.1 0.1 0.1 0.1)
;; => #S(CLIARITH:[] :LOW -0.30000000000000004 :HIGH -0.3)Multiplication:
(cliarith:*[] 0.1 0.1 0.1 0.1 0.1)
;; => #S(CLIARITH:[] :LOW 1.0e-5 :HIGH 1.0000000000000008e-5)Division:
(cliarith:/[] 0.1 0.1 0.1 0.1 0.1)
;; => #S(CLIARITH:[] :LOW 1000.0 :HIGH 1000.0000000000005)Extend a point and an interval with *negative-epsilon and *positive-epsilon
(cliarith:extend[] 1.0)
;; => #S(CLIARITH:[] :LOW 0.9999999999999999 :HIGH 1.0000000000000002)Put this repository, for instance, got by git clone in your ASDF path and load it:
(asdf:load-system :cliarith)
or, if you use roswell, install this repository with
ros install biyori-sh/cliarith
and load it:
(ql:quickload :cliarith)