Question

I noticed a hash object was once defined as in the following:

 my  %data = ();
 $data{file}  = $file;
 $data{concept} = $#row;
 $data{line1} {$cell[0]} = $cell[1];

What does this hash construction process try to achieve? Or what is the difference between

$data{concept} = $#row;

and

$data{line1} {$cell[0]} = $cell[1];

?

Answer 1

#!/usr/bin/perl
use strict;
use warnings;
use Data::Dumper;

my $file = "contents of my file";
my @row = qw(some random data);
my @cell = qw(key value);

my %data = ();
$data{file}  = $file;
$data{concept} = $#row;
$data{line1} {$cell[0]} = $cell[1];

print Dumper \%data;

Output:

$VAR1 = {
      'file' => 'contents of my file',
      'line1' => {
                   'key' => 'value'
                 },
      'concept' => '2'
};

I think $data{line1} {$cell[0]} is better written as $data{line1}{$cell[0]} or (my preference) $data{line1}->{$cell[0]}.

I included the scalar $file and the arrrays @row and @cell to demonstrate what your code means.

$data{file} = $file;

adds the contents of $file to your hash with the key file.

$data{concept} = $#row;

adds the last index of @row to your hash with the key concept. In my example the last index is 2, since the indexes in @row are 0, 1 and 2.

$data{line1} {$cell[0]} = $cell[1];

adds a hash ref to your hash with the key line1 (through autovivification) and adds the element $cell[1] to this hash ref with the key $cell[0]. Autovivification in this case means that Perl associates line1 with a hash ref and creates it, because you’re accessing it with {$cell[0]}. That saves you the trouble of having to write:

$data{line} = {};
$data{line}{$cell[0]} = $cell[1];