Path: utzoo!censor!geac!torsqnt!news-server.csri.toronto.edu!cs.utexas.edu!tut.cis.ohio-state.edu!snorkelwacker.mit.edu!bloom-beacon!eru!hagbard!sunic!dkuug!freja.diku.dk!rimfaxe.diku.dk!thorinn
From: thorinn@rimfaxe.diku.dk (Lars Henrik Mathiesen)
Newsgroups: comp.unix.shell
Subject: Re: What is 'expect'
Message-ID: <1990Nov22.164147.22894@diku.dk>
Date: 22 Nov 90 16:41:47 GMT
References: <1990Nov13.212403.11129@iwarp.intel.com> <7316:Nov1408:33:0390@kramden.acf.nyu.edu> <108721@convex.convex.com> <1990Nov15.054937.27996@midway.uchicago.edu> <5808@stpstn.UUCP>
Sender: news@diku.dk (The Netnews System)
Organization: Department Of Computer Science, University Of Copenhagen
Lines: 45

lerman@stpstn.UUCP (Ken Lerman) writes:
>In article <1990Nov15.054937.27996@midway.uchicago.edu> goer@quads.uchicago.edu (Richard L. Goerwitz) writes:
>[...]

>Shouldn't you write:.........................then I will most definitely
> want to use it in place of OTHER ugly conglomerations of sed, awk, and sh.

>No, I am not disparaging perl, because I don't know enough about it.
>But from the little I know about perl, I don't believe it solves what
>I perceive to be the problem.  What we (or I) need is a simple,
>elegant, extensible tool that is easy to learn and use.  Do people
>seriously claim that perl is it?  If so, then perhaps it is time for
>me to spend some time with my perl manual.

Perl is a somewhat complex, somewhat kludgy tool which allows you to
do (almost?) everything you can do in a Bourne shell script using sed,
awk, test and expr. It is extensible in much the same way, by piping
to or from other programs.

It seems to me to have a good balance between simplicity and features,
and between elegance and usefulness. The goal, after all, is to allow
a system programmer to express the algorithms they need with a minimum
of fuss.

My first Perl script took me an afternoon of trying out versions with
manual in hand. It involved taking a list of names in upper case,
properly capitalizing them, and merging them into the gecos field of
the password file. In this case, I might have been able to make an awk
script in half the time, but next time I don't think Perl will be
harder to write for.

My feeling is that I will almost always be more sure of what a Perl
script means than of the meaning of a shell script of similar
complexity. A large part of this difference comes from the almost
quoting-free regular expressions and the ``unifying'' of testing and
control. And with the reputation of Larry Wall, I trust the core
features of Perl to ``do what it means'' at least as much as I trust a
4.3BSD awk, for instance.

So, elegant and simple: no. Simpler and easier to learn, write and
maintain than sh, sed, awk, expr, test: yes, indeed.

--
Lars Mathiesen, DIKU, U of Copenhagen, Denmark      [uunet!]mcsun!diku!thorinn
Institute of Datalogy -- we're scientists, not engineers.      thorinn@diku.dk