Path: utzoo!utgpu!news-server.csri.toronto.edu!bonnie.concordia.ca!uunet!zaphod.mps.ohio-state.edu!magnus.acs.ohio-state.edu!csn!arrayb!wicklund
From: wicklund@intellistor.com (Tom Wicklund)
Newsgroups: comp.lang.perl
Subject: Re: regular expressions
Message-ID: <1991Jun6.153210.21305@intellistor.com>
Date: 6 Jun 91 15:32:10 GMT
References: <1991Jun6.053324.19841@uvaarpa.Virginia.EDU>
Organization: Intellistor
Lines: 14

As I might have expected, the regexp I posted wasn't correct.  The
following seems closer.

/^([^b]|b+[^ab]|b+a+[^barf]|(b|aa+|ab)+r)*(foo|[ba]+foo)([^b]|b+[^ab]|[ba]+[^bar]|(b|aa+|ab)+r)*[ba]*$/

Note that checking for not of a 2 character string has the form
(x|y)z, while checking not a 3 character string above has the form
(a|b|c|d)e.  I imagine longer strings will grow exponentially, so in
practice the "/foo/ && !/bar/" form is much more practical.  I imagine
the regexp definition could be expanded to include a not function (and
it might not be too hard to implement from what I remember of looking at
the PERL/GNU/Henry Spenser regular expression implementations).  It's
probably also possible to derive a means of inverting a regular
expression, but I wouldn't want to try it.