Path: utzoo!utgpu!news-server.csri.toronto.edu!bonnie.concordia.ca!uunet!bu.edu!nntp-read!composer
From: composer@chem.bu.edu (Jeff Kellem)
Newsgroups: comp.lang.perl
Subject: Re: Learning perl -- generlized file loading w/var assignment
Message-ID: <COMPOSER.91Feb16101423@chem.bu.edu>
Date: 16 Feb 91 15:14:23 GMT
References: <16785@venera.isi.edu>
Sender: news@bu.edu.bu.edu
Reply-To: composer@chem.bu.edu
Distribution: comp
Organization: Boston University Chemistry Department
Lines: 75
In-reply-to: jas@ISI.EDU's message of 15 Feb 91 19:04:40 GMT

In article <16785@venera.isi.edu> jas@ISI.EDU (Jeff Sullivan) writes:

 > To that end, I have created a set of files which each contain a list
 > of commands or terms which I want the user to be able to abbreviate.
 > I then want to be able to read in this list of files and assign each
 > file its own associative array filled out by abbrev.pl.  However, my
 > test doesn't seem to work.
 >
 > Here's a sample of the code:
[ ..deleted.. ]
 > #Set abbreviation eval line (run for each abbr file.).
 > #$filestub will be set to the root of the file name, which is always
 > #of the form <file-contents>-LIST.
 > $eval_line = "%$filestub = ();\n&abbrev(*$filestub, @tmp);\n";

First off, there are a few problems with the above $eval_line.  Since you've
enlosed the RHS of the $eval_line assignment in double quotes, the $filestub,
and @tmp will be evaluated immediately.  Since you haven't assigned anything
to $filestub yet (and you want to use the name @tmp as an arg) they get
replaced by their current contents, which happen to be nothing at this point.

When creating code on the fly, it's always a good idea to add:

	print $eval if $debug;

before the actual eval.  Anyways, you probably want to move the
$eval_line assignment into the "foreach $filename" loop after the
$filestub assigment.  Also, you don't need the newlines in this case.
So, with double quotes, the above should be written as:

	$eval_line = "%$filestub = (); &abbrev(*$filestub, \@tmp);";

You want the "@tmp" to stay, as is, so you need to escape the @-sign.

 > #Load abbreviation files.
 >
 > foreach $filename ("ATTR-LIST", "HELP-LIST", "TYPE-LIST", "CMD-LIST") {
 >
[ ..deleted.. ]
 >     $filename =~ /(\w+)-LIST/;
 >     $filestub = $1;
[ ..deleted.. ]
 > while (<>) {
 >     chop;
 >     $long = $cmd{$_};
 >     print "$long\n" if length($long);
 > }
 >
 > The last bit is pulled vrom the example using abbrev.pl.  The $cmd{}
 > array (should) be created by the eval_line when $filename is
 > "CMD-LIST" and $filestub is "CMD".  Right?

Except that variable names are case-sensitive.  So, assuming the code
worked, you would be producing an associative array named %CMD, instead
of %cmd.  If you want to lowercase the part of $filename you're trying
to grab, there are a bunch of ways to do it.  Here's one:

($filestub = $filename) =~ s/^(\w+)-list/(($lc = $1) =~ tr|A-Z|a-z|),$lc/ie;

Okay, maybe you want something a little easier to read for the
beginner.. ;-}  Following your start:

	$filename =~ /(\w+)-LIST/i;
	($filestub = $1) =~ tr/A-Z/a-z/;

You may want the i on the end of the match, //i, so you can match
CMD-LIST and CMD-list and so on... in otherwords, the match ignores
case.

Hope that helps a little...Enjoy Perl!

			-jeff

Jeff Kellem
Internet: composer@chem.bu.edu