Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!swrinde!zaphod.mps.ohio-state.edu!sol.ctr.columbia.edu!cunixf.cc.columbia.edu!ben
From: ben@cunixf.cc.columbia.edu (Ben Fried)
Newsgroups: comp.lang.perl
Subject: Re: Why is this so slow?
Message-ID: <1990Dec10.022814.5349@cunixf.cc.columbia.edu>
Date: 10 Dec 90 02:28:14 GMT
References: <1990Dec7.212127.8161@uvaarpa.Virginia.EDU>
Organization: Columbia University Center for Computing Activities
Lines: 34

In article <1990Dec7.212127.8161@uvaarpa.Virginia.EDU> Dale Worley writes:
>    From: merlyn@iwarp.intel.com (Randal Schwartz)
>
>    $lines_per_page = 66; $max_pages = 100; # you forgot this
>    while (<>) {
> 	   $lineno++;
> 	   $pageno++, $lineno = 0 if /\f/ || ($lineno >= $lines_per_page);
> 	   die "Aborted at page $pageno\n" if $pageno >= $max_pages;
> 	   print;
>    }
> 
> Well, to start with, if there are two form-feeds in one line, $pageno
> will only get incremented by 1.
> 
> I may be wrong, but to find out the number of times a regexp matches a
> string, you have to write an explicit loop.

This is a good point - neither Randal's nor Larry's posted code handles
this case.  

Larry's code for non-plaintext files works just great: in those cases,
the _only_ way that the printer can reach end of page is by a \f or an
\033\f, depending on mode; his idea of setting $/ to \f (and in the
second case matching against /\033\f$/) works great.

This made me wish that perl had some operator for determining the count
of matches of a pattern; is that something that sounds generally useful?

> Dale Worley		Compass, Inc.			worley@compass.com

Ben
--
Benjamin Fried
ben@cunixf.cc.columbia.edu				rutgers!columbia!ben