Path: utzoo!attcan!uunet!know!zaphod.mps.ohio-state.edu!usc!elroy.jpl.nasa.gov!jpl-devvax!lwall
From: lwall@jpl-devvax.JPL.NASA.GOV (Larry Wall)
Newsgroups: comp.lang.perl
Subject: Re: Pattern Substitution Question
Keywords: pattern
Message-ID: <10154@jpl-devvax.JPL.NASA.GOV>
Date: 29 Oct 90 20:50:13 GMT
References: <1086@dg.dg.com>
Reply-To: lwall@jpl-devvax.JPL.NASA.GOV (Larry Wall)
Organization: Jet Propulsion Laboratory, Pasadena, CA
Lines: 41

In article <1086@dg.dg.com> blurie@einstein.webo.dg.com writes:
: Here's a question, whose answer I'm sure is going to be trivial, that I've
: been trying to resolve for days.  Given input like the following:
: 
: 	Lurie;Benjamin;1212 Nowhere Road;(617) 923-0000;(w) (000) 202-2343
: 
: I would like to translate it to:
: 
: 	Benjamin Lurie;1212 Nowhere Road;(617) 923-0000;(w) (000) 202-2343
: 
: I've tried a pattern match looking for a semi-colon, but I always get the
: longest pattern that matches, which would include everything up to the "(w)".
: 
: What pattern substitution would accomplish what I want here?  And, in general,
: how would I specify to match only the input up to the first ocurrance of 
: <c>, where <c> is some specific character?

The general solution is to use a negated character class.  For semicolons,
say something like:

    while (<>) {
	s/^([^;]*);([^;]*);/$2 $1;/;
	print;
    }

However, if you're looking for readability it might be cleaner to say
something like:

	while (<>) {
	    ($lastname, $firstname, $rest) = split(/;/, $_, 3);
	    print "$firstname $lastname;$rest";
	}

And, believe it or not, it runs faster, at least on my machine.  You can
shave another 7-8% off by saying

	    print $firstname, ' ', $lastname, ';', $rest;

which avoids some string copying.

Larry