Path: utzoo!utgpu!jarvis.csri.toronto.edu!cs.utexas.edu!tut.cis.ohio-state.edu!purdue!bu.edu!nntp-read!jbw
From: jbw@bucsf.bu.edu (Joe Wells)
Newsgroups: comp.lang.perl
Subject: perl slowness with regexp
Message-ID: <JBW.90Mar5221106@bucsf.bu.edu>
Date: 6 Mar 90 03:11:06 GMT
Sender: news@bu.edu.bu.edu
Organization: Boston University Computer Science Department
Lines: 28

I ran into some interesting perl behavior recently.  The following short
program will illustrate it:

    for (1 .. 20)
    {
      $_ = 'Y' . ('x' x $_);
      print "_: $_\n";
      m/(.+)*Y/;
      print "`: $`, 1: $1, &: $&, ': $'\n";
    }

If you run this program, you will notice that each iteration of the loop
takes much longer than the previous iteration.  Am I correct in guessing
that this is perl's way of telling me don't do that?  Or is this kind of
regular expression optimizable?

For the curious, this is the original regexp from which I distilled the
one in the program above:

    ^\s*(\S+(\s*\S+)*)\s*<.*>$

I was trying to exclude the spaces on the end of what I was looking for.
Since then, I've done that afterwards.

-- 
Joe Wells <jbw@bu.edu>
jbw%bucsf.bu.edu@cs.bu.edu
...!harvard!bu-cs!bucsf!jbw