#!/usr/bin/perl -w
use strict;
my @dup_array = qw/this this array has has duplicates/;
my @unique_array;
my %uniques;

foreach my $elem (@dup_array)
{
      push @unique_array, $elem unless  $unique{$elem}++;
}

print "Unique elements are @unique_array";
[download]

my @lines = <>;
[download]

Mind you, what a line of input is depends on whether you've messed around with $/;

I have an array that has multiple entries of the same element in it. I want to make a new array with just the unique elements in it. How can this be done?

Pretty much every time you see the word "unique" in a problem, you want a hash. For this, I'd take a two-pass approach: tally the elements of the array, then pick out the unique ones. So:

my %elem_count = ();

for my $elem (@array) {
  $elem_count{$elem}++;
}

my @uniques = ();

for my $elem (keys %elem_count) {
  push @uniques, $elem if $elem_count{$elem} == 1;
}
[download]

So (to be brutally obvious) we're storing the number of times we see each element in the hash %elem_count, then going through %elem_count key by key, picking out the ones that have a count of 1. (Of course, you can do the first bit at the same time as you're inserting elements in the array, which may or may not be practical.)

As for your second question, I confess I don't really see the problem. Unless you've been mucking around with $/, $_ will contain the line you just read in at every pass through the while.

You have a strange definition of unique :) ... your code removes any entries which don't occur exactly once, try for example this:

my @array = qw/one one two/;

# your algorithm

print @uniques;
# prints 'two'
# and I would expect 'onetwo'
[download]

So changing that and putting it a bit shorter:

my %count = ();
my @uniques = grep {! $count{$_}++} @array;
[download]

... and btw this should be in the FAQs somewhere.

-- Hofmator

You have a strange definition of unique :) ... your code removes any entries which don't occur exactly once

Oh, I don't know, I like my definition. :-) I read (misread?) the question as "pick out the elements that occur just once" rather than "make a copy of the list with duplicate elements removed".

... and btw this should be in the FAQs somewhere.

I believe it is, as mentioned in an earlier response.

--
:wq

use strict;

# Set up
my @dups_ary = qw(a b c b d e f c b g h);

# Do it
my %temp_hash;
undef @temp_hash{@dups_ary};
my @unique_ary = keys %temp_hash;

# Prove it
print "@unique_ary\n";
[download]

_____________________________________________________
Jeff[japhy]Pinyan: Perl, regex, and perl hacker.
s++=END;++y(;-P)}y js++=;shajsj<++y(p-q)}?print:??;

@array_thing = sort @array_thing;
foreach (@array_thing) {
   push @unique if $_ ne $prev;
   $prev = $_;
}
[download]

more simple - well.... yes and no...

#!/usr/bin/perl

my @a = ('a', 'b', 'c', 'c' ,'f', 'a', 'd');

my @b = grep (!$_{$_}++, @a);
[download]

the 'my @b' line is the solution - it comes up with the correct info, but does involve doing a little work to get to the answer.

It's definitely more simple to write :) but as for processing times etc - I really don't know if it's going to be better or not.

It's a bit nasty to put this in your code without commenting it also, as I've got a few perl friends (none uber perl ppls, but 'know' perl), and none of them understood the intent :)