in reply to storing domain info in hashes of hashes

Something like this should work (untested):
sub insert_domain { my $cursor = shift; $cursor = $cursor->{$_} ||= {} for @_; $cursor = {}; } insert_domain(\%domain, qw(org perlmonks www));

The only "trick" here is using ||=, which only assigns to a variable if it's false. Since we only add hash references in this hash structure and references are always true, it means we only assign an empty anonymous hash {} to nonexistant values; otherwise, we move the cursor down one level to the already existing subhash.

Update: pg's post compelled me to test mine, and it does work:

insert_domain \%domain, qw(com perlmonks www); insert_domain \%domain, qw(org perlmonks www); insert_domain \%domain, qw(org perl www); insert_domain \%domain, qw(org perl use); insert_domain \%domain, qw(org perl jobs); insert_domain \%domain, qw(com perl www); insert_domain \%domain, qw(org cpan www); use Data::Dumper; print Dumper \%domain; __END__ $VAR1 = { 'org' => { 'cpan' => { 'www' => {} }, 'perl' => { 'www' => {}, 'jobs' => {}, 'use' => {} }, 'perlmonks' => { 'www' => {} } }, 'com' => { 'perl' => { 'www' => {} }, 'perlmonks' => { 'www' => {} } } };

Makeshifts last the longest.

Replies are listed 'Best First'.
Re: Re: storing domain info in hashes of hashes
by ajefferies (Initiate) on Nov 20, 2002 at 14:38 UTC
    That is one fine solution. Completely different than I had been attempting. Thanks so much.