Multi keyed hash

A quirky multi keyed hash with an anon array for its value.

%modhash = ();
$modhash{ 'first:second:third' } = [ 'bob', 'bill', 'bo' ];
$modhash{ 'Canada:USA:China' } =
                               [ 'beaver','eagle','Mao' ];

# two sample entries, notice the anonymous arrays

foreach $key ( keys %modhash )
{
  @subkeys = split( /:/, $key );
  for( $i = 0; $i < scalar @subkeys; $i++ )
  {
    print "subkey-> $subkeys[$i]\t",
          "subvalue-> $modhash{$key}->[$i]\n";
  }
  print "\n"; # separates different main keys
}
[download]

Produces:
subkey-> Canada subvalue-> beaver
subkey-> USA subvalue-> eagle
subkey-> China subvalue-> Mao

subkey-> first subvalue-> bob
subkey-> second subvalue-> bill
subkey-> third subvalue-> bo

This is a snippet I came up with as a response for
someone's weird post. Its pretty useless, but I have
a feeling someone, somewhere, is going to use it.
Perhaps its good practice for me ( or you ) to make
a module for it.
Best regards, bl0rf

My Site

Comment on Multi keyed hash Download Code

Replies are listed 'Best First'.
Re: Multi keyed hash by jeffa (Bishop) on Sep 21, 2003 at 15:11 UTC
"This is a snippet I came up with as a response for someone's weird post..." That would be How to create multiple keys in a hash, please don't be afraid to link around here, it gives context to your snippet. My usually complaints follow: please use strict and warnings when you post code to the Monastery. Not because we think you don't know what you are doing, but because you are posting this code for others to use and review. You should at the very least declare your variables first. Here is a rewrite just to show you a more idiomatic way of doing what you have done. The main differences are declaring and populating the hash in one line and getting rid of the unnecessary C-style for loop: `use strict; use warnings; # two sample entries, notice i moved this comment my %modhash = ( 'first:second:third' => [ qw(bob bill bo) ], 'Canada:USA:China' => [ qw(beaver eagle Mao) ], ); for my $key ( keys %modhash ) { my @subkeys = split( /:/, $key ); for my $i (0..$#subkeys) { print "subkey-> $subkeys[$i]\t", "subvalue-> $modhash{$key}->[$i]\n"; } print "\n"; }` [download] But, now that we have this, how are we supposed to find out the value for say, Canada ... by itself? Ughhh ... suddenly this code seems unecessary. What advantage is there to grouping keys together if they don't point to the same data? In this case, they do point to the same data, but that data itself is to be split to find the right element. Total overkill, we could just use this instead: `my %hash = ( first => 'bob', second => 'bill', third => 'bo', Canada => 'beaver', USA => 'eagle', China => 'Mao', );` [download] Grouping multiple keys and multiple values in the manner you have really makes no sense. (But that's OK -- the original question made no sense either!) If we need to group keys together so that they give the same value, we can use another hash that is tied to something like Tie::RangeHash: `use Tie::RangeHash; tie my %rhash, 'Tie::RangeHash'; %rhash = ( 'first,second,third' => 1, 'Canada,USA,China' => 2, ); print $rhash{'second'},$/; print $rhash{'China'},$/;` [download] Now we can use the tied hash and the regular hash together to achieve our goal ... which i still haven't figured out. ;) jeffa L-LL-L--L-LL-L--L-LL-L-- -R--R-RR-R--R-RR-R--R-RR B--B--B--B--B--B--B--B-- H---H---H---H---H---H--- (the triplet paradiddle with high-hat)	[reply] [d/l] [select]

Replies are listed 'Best First'.

Re: Multi keyed hash
by jeffa (Bishop) on Sep 21, 2003 at 15:11 UTC

"This is a snippet I came up with as a response for someone's weird post..."

That would be How to create multiple keys in a hash, please don't be afraid to link around here, it gives context to your snippet.

My usually complaints follow: please use strict and warnings when you post code to the Monastery. Not because we think you don't know what you are doing, but because you are posting this code for others to use and review. You should at the very least declare your variables first. Here is a rewrite just to show you a more idiomatic way of doing what you have done. The main differences are declaring and populating the hash in one line and getting rid of the unnecessary C-style for loop:

use strict;
use warnings;

# two sample entries, notice i moved this comment
my %modhash =
(
   'first:second:third' => [ qw(bob bill bo) ],
   'Canada:USA:China'   => [ qw(beaver eagle Mao) ],
);

for my $key ( keys %modhash )
{
  my @subkeys = split( /:/, $key );
  for my $i (0..$#subkeys)
  {
    print "subkey-> $subkeys[$i]\t",
          "subvalue-> $modhash{$key}->[$i]\n";
  }
  print "\n";
}
[download]

my %hash =
(
   first  => 'bob',
   second => 'bill',
   third  => 'bo',
   Canada => 'beaver',
   USA    => 'eagle',
   China  => 'Mao',
);
[download]

Tie::RangeHash

use Tie::RangeHash;
tie my %rhash, 'Tie::RangeHash';

%rhash =
(
   'first,second,third' => 1,
   'Canada,USA,China'   => 2,
);

print $rhash{'second'},$/;
print $rhash{'China'},$/;
[download]

jeffa

L-LL-L--L-LL-L--L-LL-L--
-R--R-RR-R--R-RR-R--R-RR
B--B--B--B--B--B--B--B--
H---H---H---H---H---H---
(the triplet paradiddle with high-hat)

[reply]
[d/l]
[select]