NetNews Usenet Archive 1992 #30/NN_1992_30.iso/spool/comp/lang/perl/7455

Text File | 1992-12-16 | 1.9 KB | 55 lines

Newsgroups: comp.lang.perl
Path: sparky!uunet!zaphod.mps.ohio-state.edu!menudo.uh.edu!sugar!claird
From: [e-mail] (Cameron Laird)
Subject: In search of good idioms.
Organization: NeoSoft Communications Services -- [phone]
Date: Wed, 16 Dec 1992 14:24:52 GMT
Message-ID: <[e-mail]>
Lines: 45

I'm a newbie perl-plucker.  I'm working through some
exercises I've set myself, in an effort to develop
some taste in matters of perl style.  Current problem:
write a permute program.  Such a program would, for
example, when presented with input

    permute a1 zebra third

generate output something like

    a1 zebra third
    a1 third zebra
    zebra a1 third
    zebra third a1
    third a1 zebra
    third zebra a1

I don't like any of the solutions I've conceived, and
seek the judgment of those with more experience.  My
current summary of the situation goes something like
this:

The natural expressions of this problem involve an
abstract data type, "collection of (sub)permutations",
or its proceduralization by way of a parametrized re-
cursion.  Procedure calls with more than one formal
parameter look gross in perl (don't they?  Is there a
way to handle them without clutter?), and there is no
intrinsic support for data abstraction.  Therefore,
all perl solutions of this problem look clunky.

Is this true?  Is there some slick way to use associative
arrays to solve this?  Have I inadvertently hit on a
well-known demo that's available on some ftp site I
haven't yet visited?  I suspect that working through this
to some conclusion will help me better understand the
<<esprit de perl>>; thanks for any assistance.

I'm also having problems with wafe that the FAQ doesn't
answer, but I'll save that for another time ...
--

Cameron Laird
[e-mail] ([e-mail])    [phone]
[e-mail] ([e-mail])      [phone]