use List::MoreUtils 'uniq';
my @new = uniq ( @old );
[download]

my @set = qw(a a b c c);
my %seen;
$seen{$_}++ foreach @set;
my @unique = grep { $seen{$_} == 1 } @set;
[download]

Your code could be condensed to

my @set = qw(a a b c c);
my %seen;
my @unique = grep { not $seen{$_} ++ } @set;
[download]

That method preserves the order of the original array but if that's not important you could also do

my @set = qw(a a b c c);
my %seen;
@seen{@set} = ();
my @unique = keys %seen;
[download]

I hope this is of interest.

Cheers,

JohnGG

I would pump my array into a hash, then let it flow back out to an array if that was what I really wanted. I assume of course that your not banging on performance issues really hard. Or just do it the easy way and use the hash in the first place... :-)

my %unique;
my @ids = map
{ chop;
  my @fld = split /\t/;
  !defined(  $unique{ $fld[0] } ) && $unique{ $fld[0] }++ && $fld[0];
} <$fileHandle>;
[download]

Ouch. chop is almost always wrong (what if your EOL marker is more than one character)--and what if $fld[0] is a valid value that evaluates to false in boolean context?

The popular misuse of chomp() deserves a new topic. As for $fld[0] - defined( unique{ $fld[0] } ) will be true in the case you cite.

There might be functional issues or data quality issues in any real example but these are in general OT.

__________________________________________________________________________________

^M Free your mind!