racket/collects/scribblings/guide/numbers.scrbl
Eli Barzilay debd1f9f1e Recketizing much in `net/*', mass ".ss" -> ".rkt" conversion in .scrbl files.
(Some other minor things here and there.)
2011-06-20 04:27:14 -04:00

148 lines
4.7 KiB
Racket

#lang scribble/doc
@(require scribble/manual scribble/eval "guide-utils.rkt")
@title[#:tag "numbers"]{Numbers}
A Racket @deftech{number} is either exact or inexact:
@itemize[
@item{An @defterm{exact} number is either
@itemize[
@item{an arbitrarily large or small integer, such as @racket[5],
@racket[99999999999999999], or @racket[-17];}
@item{a rational that is exactly the ratio of two arbitrarily
small or large integers, such as @racket[1/2],
@racket[99999999999999999/2], or @racket[-3/4]; or}
@item{a complex number with exact real and imaginary parts
(where the imaginary part is not zero), such as @racket[1+2i] or
@racket[1/2+3/4i].}
]}
@item{An @defterm{inexact} number is either
@itemize[
@item{an IEEE floating-point representation of a number, such
as @racket[2.0] or @racket[3.14e87], where the IEEE
infinities and not-a-number are written
@racket[+inf.0], @racket[-inf.0], and @racket[+nan.0]
(or @racketvalfont{-nan.0}); or}
@item{a complex number with real and imaginary parts that are
IEEE floating-point representations, such as
@racket[2.0+3.0i] or @racket[-inf.0+nan.0i]; as a
special case, an inexact complex number can have an
exact zero real part with an inexact imaginary part.}
]}
]
Inexact numbers print with a decimal point or exponent specifier, and
exact numbers print as integers and fractions. The same conventions
apply for reading number constants, but @litchar{#e} or
@litchar{#i} can prefix a number to force its parsing as an exact
or inexact number. The prefixes @litchar{#b}, @litchar{#o}, and
@litchar{#x} specify binary, octal, and hexadecimal
interpretation of digits.
@refdetails/gory["parse-number"]{the syntax of numbers}
@examples[
0.5
(eval:alts @#,racketvalfont{#e0.5} 1/2)
(eval:alts @#,racketvalfont{#x03BB} #x03BB)
]
Computations that involve an inexact number produce inexact results,
so that inexactness acts as a kind of taint on numbers. Beware,
however, that Racket offers no ``inexact booleans,'' so computations
that branch on the comparison of inexact numbers can nevertheless
produce exact results. The procedures @racket[exact->inexact] and
@racket[inexact->exact] convert between the two
types of numbers.
@examples[
(/ 1 2)
(/ 1 2.0)
(if (= 3.0 2.999) 1 2)
(inexact->exact 0.1)
]
Inexact results are also produced by procedures such as @racket[sqrt],
@racket[log], and @racket[sin] when an exact result would require
representing real numbers that are not rational. Racket can represent
only rational numbers and complex numbers with rational parts.
@examples[
(code:line (sin 0) (code:comment @#,t{rational...}))
(code:line (sin 1/2) (code:comment @#,t{not rational...}))
]
In terms of performance, computations with small integers are
typically the fastest, where ``small'' means that the number fits into
one bit less than the machine's word-sized representation for signed
numbers. Computation with very large exact integers or with
non-integer exact numbers can be much more expensive than computation
with inexact numbers.
@def+int[
(define (sigma f a b)
(if (= a b)
0
(+ (f a) (sigma f (+ a 1) b))))
(time (round (sigma (lambda (x) (/ 1 x)) 1 2000)))
(time (round (sigma (lambda (x) (/ 1.0 x)) 1 2000)))
]
The number categories @deftech{integer}, @deftech{rational},
@deftech{real} (always rational), and @deftech{complex} are defined in
the usual way, and are recognized by the procedures @racket[integer?],
@racket[rational?], @racket[real?], and @racket[complex?], in addition
to the generic @racket[number?]. A few mathematical procedures accept
only real numbers, but most implement standard extensions to complex
numbers.
@examples[
(integer? 5)
(complex? 5)
(integer? 5.0)
(integer? 1+2i)
(complex? 1+2i)
(complex? 1.0+2.0i)
(abs -5)
(abs -5+2i)
(sin -5+2i)
]
The @racket[=] procedure compares numbers for numerical equality. If
it is given both inexact and exact numbers to compare, it essentially
converts the inexact numbers to exact before comparing. The
@racket[eqv?] (and therefore @racket[equal?]) procedure, in contrast,
compares numbers considering both exactness and numerical equality.
@examples[
(= 1 1.0)
(eqv? 1 1.0)
]
Beware of comparisons involving inexact numbers, which by their nature
can have surprising behavior. Even apparently simple inexact numbers
may not mean what you think they mean; for example, while a base-2
IEEE floating-point number can represent @racket[1/2] exactly, it
can only approximate @racket[1/10]:
@examples[
(= 1/2 0.5)
(= 1/10 0.1)
(inexact->exact 0.1)
]
@refdetails["numbers"]{numbers and number procedures}