i was writing a unit test which had something analogous to the following:
(check-equal? #rx"a" #rx"a") => raises exn:test:check
so i checked and sure enough:
(equal? #rx"a" #rx"a) => #f
which led me to look for a regexp-equal? so i could do (check regexp-equal?
#rx"a" #rx"a")
that doesn't exist, so i wrote one:
(define (regexp-equal? a b)
(and (and (regexp? a)
(regexp? b))
(or (and (pregexp? a)
(pregexp? b))
(and (not (pregexp? a))
(not (pregexp? b))))
(equal? (object-name a)
(object-name b))))
why not just have (equal? #rx"a" #rx"a") => #t ?
anticipated objection: "what should equal? mean for two regexps?"
it should mean that the patterns are identical, totally ignoring that two
non-identical patterns might match exactly the same set, like #rx"(a|b)"
and #rx"(b|a)".
i see in the docs that there is an internal regexp value. if those are
what i think they are, i propose equal? just compares those for regexes.
_________________________________________________
For list-related administrative tasks:
http://lists.racket-lang.org/listinfo/users
