Simple way to parse whitespace separated, optionally quoted words out of a string

[Aristotle]

I recently needed to parse words separated by whitespace out of a string, where a doublequoted word may contain spaces itself but no escaped doublequotes. The following regex will do, is quite tidy and also hardly backtracks (occasionally by a few characters, not more).

my @words = /"?((?<!")\S+(?!"\s)|[^"]+)"?\s*/g;

[japhy]

Your method has its flaws. It does not require proper balancing of quotes, and it breaks q{"oops"I did it again"} into q{oops} and q{I did it again} -- that is, two strings, where it should be five. It also returns q{""} for the string q{""}, when it should probably return just an empty string (q{}).

_____________________________________________________
Jeff[japhy]Pinyan: Perl, regex, and perl hacker, who'd like a job (NYC-area)
s++=END;++y(;-P)}y js++=;shajsj<++y(p-q)}?print:??;

[Aristotle]

Thanks for your points. The latter was easy to fix, but the first one caused me a fair bit of work and I wasn't able to come up with a solution that worked with only a single capturing pair of parens (my original goal). The following shouldn't fail regardless of how pathological the case you throw at it gets, but requires a grep.

my @word = do {
    my $i=0;
    grep $i++&1, m/ (")? ((?(1) [^"]* | \S+)) \1\s* /xg
};

Correct, but not nearly as neat as before. :-/ It does have the advantage that it can easily be extended to accept single quotes as well, though.

Makeshifts last the longest.