[REBOL] Re: Parse versus Regular Expressions
From: joel:neely:fedex at: 4-Apr-2003 17:13
Hi, Gregg,
Gregg Irwin wrote:
> << 2) It totally breaks normal usage that "or" is symmetrical, even
> though we read the | as "or". >>
>
> It's like the ANY function, so I don't know that I'd say it "totally
> breaks" normal usage. Isn't OR often optimized to work this way; i.e.
> shortcut evaluation?
>
In REBOL (at least) OR does *not* shortcircuit, but ANY does:
>> any [do [print "1" true] do [print "2" true]]
1
== true
versus
>> (do [print "1" true]) or (do [print "2" true])
1
2
== true
However, that's beside the point if we're talking about symmetry; in
REBOL *both* OR and ANY are symmetric (over pure logic values),
in the sense that the result is independent of the order of the
arguments:
foreach a reduce [true false] [
foreach b reduce [true false] [
print [a b equal? any [a b] any [b a] equal? (a or b) (b or
a)]
]
]
yields
true true true true
true false true true
false true true true
false false true true
So it's not like ANY in that respect.
> I think the most important thing, barring actual "bugs" in PARSE's
> design is documenting how these things work so grammars can be designed
> correctly (e.g. longest match first; greedy grammars? :).
>
I totally agree with that statement.
-jn-
--
----------------------------------------------------------------------
Joel Neely joelDOTneelyATfedexDOTcom 901-263-4446
Counting lines of code is to software development as
counting bricks is to urban development.
