<?xml version="1.0" encoding="windows-1252"?>
<node id="1002969" title="Re: My first perl script is working, what did I do wrong?" created="2012-11-08 14:52:53" updated="2012-11-08 14:52:53">
<type id="11">
note</type>
<author id="622051">
toolic</author>
<data>
<field name="doctext">
If your part numbers will never have  spaces in them, you could [doc://split] your lines and build up a [doc://perldsc|hash of arrays]:

&lt;c&gt;
use strict;
use warnings;

print "ARGV: $#ARGV\n";

if ($#ARGV != 0) {
     print "One, and only one (not $#ARGV), command line parameter is expected\n";
     exit;
}
my $inputFile=$ARGV[0];

if ( not -e $inputFile) {
     die "$inputFile doesn't exist!!!\nExiting.\n";
}

print "Attempting to open $inputFile\n";

open my $file, $inputFile or die "Could not open $inputFile: $!";

my %parts;
while (&lt;$file&gt;) {
    next if /^Part/;
    my ($part, $info) = split /\s+/, $_, 2;
    $info =~ s/\s//g;
    push @{ $parts{$info} }, $part;
}

for my $info (sort keys %parts) {
    print "@{ $parts{$info} } - $info - ", scalar @{ $parts{$info} }, "\n";
}

__END__

ARGV: 0
Attempting to open 1.txt
P10 - CircleBlue4 - 1
P9 - CircleGreen3 - 1
P1 P4 - CircleRed1 - 2
P7 - RectangleBlue4 - 1
P3 - RectangleRed4 - 1
P5 - SquareBlue1 - 1
P2 P6 - SquareGreen3 - 2
P8 - SquareRed2 - 1
&lt;/c&gt;

&lt;p&gt;With a little more work, you could [doc://sort] on hash [doc://values] to get the output in the order you want.</field>
<field name="root_node">
1002964</field>
<field name="parent_node">
1002964</field>
</data>
</node>
