On Aug 10, Michel Lambert said:
>Fun with Perl's Regexes, to be more precise. They're a lot more fun than
>regular regexes. Here's a little quiz I made up while falling asleep last
>night.
Mike, I'd call you sick, but I'd hate to think what adjective that would
leave for me. ;)
>1) What's the fastest way to strip trailing whitespace ( as defined by
>/\s/ ) from a decent-sized string? Decent-sized means this email, or 100 of
>these emails concatenated together.
I'm standing behind my sexeger approach:
$_ = reverse;
s/^\s+//;
$_ = reverse;
The reason s/\s+$// is so bad is because it isn't optimized, and therefore
the \s+ matches at EVERY chunk of whitespace in your string, no matter
where it is, before it gets to the very end. Ick-o.
>2) Is it possible to write a function which can tell me if a regex matches
>the empty string? (Since the code would most likely be long, a simple why or
>why not will do ;)
I stand behind the
sub empty { "" =~ shift }
approach. But, if you mean "can you inspect a regex and determine if it
matches the empty string", I, being a regex junkie, will say yes.
Here are the following ways a regex can match an empty string:
* all of its tokens have a {0,X} quantifier
? and * and {0,4}
* it contains no text-matching tokens
/^$/ matches ""
* it contains alternation to an empty string at each level
/(abcd|)efgh|/
I think that covers it. And that can be determined in parsing.
>3) Write a regex that uses variable-width-lookahead without using the
>various look-ahead operators, (?=..), (?!..), (?<=..), or (?<!..)
If Perl wouldn't be such a hag about re-entering the engine, I'd do it in
a second:
(?=...) (?(?{ $' =~ /\A.../ })|(?!))
(?!...) (?(?{ $' =~ /\A.../ })(?!))
(?<=...) (?(?{ "$`$&" =~ /...\z/ })|(?!))
(?<!...) (?(?{ "$`$&" =~ /...\z/ })(?!))
I'd prefer to use sexeger approaches for those last two, but that would be
getting ahead of myself. ;)
Oh, and don't yell at me for using (?!) as the failure. I suppose I could
have put something heinous in there like -\A instead (I'd like to see YOU
match a dash before the beginning of a string!).
--
Jeff "japhy" Pinyan [EMAIL PROTECTED] http://www.pobox.com/~japhy/
RPI Acacia brother #734 http://www.perlmonks.org/ http://www.cpan.org/
** Look for "Regular Expressions in Perl" published by Manning, in 2002 **
