Re: optimal way of comparing 2 arrays

Oh, I don't know about optimization but I would probably do something like

#!/usr/bin/perl -l

use strict;
use warnings;

my @file1=qw/foo bar baz/;
my @file2=qw/fred bar barney/;

my %saw;
$saw{$_}=1 for @file1;
$saw{$_}.=2 for @file2;
my %comm; push @{ $comm{$saw{$_}} }, $_ for keys %saw;

print <<"EOT"
\@file1 only: (@{ $comm{1} })
\@file2 only: (@{ $comm{2} })
common: (@{ $comm{12} })
EOT

__END__
[download]

But you didn't tell us whether order does matter and if your arrays contain duplicates and if you also want them in the output and so on, so YMMV...

Comment on Re: optimal way of comparing 2 arrays Download Code

Replies are listed 'Best First'.
Re^2: optimal way of comparing 2 arrays by narashima (Beadle) on Oct 20, 2005 at 13:51 UTC
Hi Blazar, No duplicates in these arrays. All values are unique. thanks narashima	[reply]
Re^3: optimal way of comparing 2 arrays by blazar (Canon) on Oct 20, 2005 at 15:00 UTC
Then the solution above may be simplified to: `#!/usr/bin/perl -l use strict; use warnings; my @file1=qw/foo bar baz/; my @file2=qw/fred bar barney/; my %saw; $saw{$_}++ for @file1, @file2; print <<"EOT" \@file1 only: (@{[ grep $saw{$_}==1, @file1 ]}) \@file2 only: (@{[ grep $saw{$_}==1, @file2 ]}) common: (@{[ grep $saw{$_}==2, @file2 ]}) EOT __END__` [download] or even, using modulo-2 math, but more obscurely: `#!/usr/bin/perl -l use strict; use warnings; my @file1=qw/foo bar baz/; my @file2=qw/fred bar barney/; my %saw; $saw{$_} ^= 1 for @file1, @file2; print <<"EOT" \@file1 only: (@{[ grep $saw{$_}, @file1 ]}) \@file2 only: (@{[ grep $saw{$_}, @file2 ]}) common: (@{[ grep !$saw{$_}, @file2 ]}) EOT __END__` [download]	[reply] [d/l] [select]