sub change_punch_process
   {
      my @numbers;
      @q = ("1a", "2b", "3c", "2k");

      foreach ( @q )
      {
         if (/^([0-9]+)/)
         {
            my $exists = 0;
            foreach (@numbers)
            {
               $exists = 1 if ($_ == $1)
            }
            if (!$exists) {push @numbers, $1;}
         }
      }
      print join(",",@numbers);
}

change_punch_process;
[download]

Like pg said, it won't work.

Perhaps a hash is in order? Here's an example:

use strict;

my %numbers;

while (<DATA>) {
    $numbers{$1}++ if /^(\d+)/;
}

print "$_\n" foreach keys %numbers;

__DATA__
123askdjha
1ad
asldhasd
asdklsd123
10
2fer
[download]

1
2
10
123
[download]

Update (fixed code): Alternatively, using an array:

use strict;

my @numbers;

while (<DATA>) {
    /^(\d+)/;
    push(@numbers, $1) unless grep(/^$1$/, @numbers) || $1 eq '';
}

print "$_\n" foreach @numbers;

__DATA__
123askdjha
1ad
asldhasd
asdklsd123
10
2fer
[download]

123
1
10
2
[download]

This bit looks fishy:

foreach (@numbers)
{
  unless ($_ == $1)
  { push @numbers,$1; }
}
[download]

This would push $1 onto @numbers multiple times (once for each existing element that didn't match) except that it never runs at all since @numbers starts out empty. You could use the grep() function to check whether a value occurs in the array, but a better approach would be to use a hash instead and check with exists().

Also, your subroutine doesn't explicitly return a value.

Perhaps try it with a hash, like this: (untested)

sub f
{
    my %numbers ;
    foreach ( $q->param )
    {
        /^([0-9]+)/ && $numbers{$1}++ ;
    }
    return keys %numbers;
}
[download]

Those who know that they are profound strive for clarity. Those who
would like to seem profound to the crowd strive for obscurity.
            --Friedrich Nietzsche

Whoop!

thanks for the replies, people

I didn't know there was a grep function in Perl!

sub change_punch_process
{
   my %transactions;
   my @numbers;


   foreach (sort ($q->param))
   {
      if (/^([0-9]+)/)
      {
        push (@numbers,$1) unless grep /\b$1\b/, @numbers;
      }
   }
print "<br><br>@numbers";
}
[download]

As, grantm said, I kept putting in the same numbers over and over.

IMO, a hash would be overkill. I just need the numbers to append my CGI variables.

That grep statement should work right?

At least it is so far.

Thanks again!

IMO, a hash would be overkill.

Far less than repeatedly grepping through an array with regular expressions. Either the number exists in the hash or it doesn't. The check takes the same amount of time regardless of the number of elements in the hash (modulo hash bucketing, a subject I really don't want to discuss). With a grep, every new number you add to the array will be checked, every time. The regex is worse, because it has to compile the regex for each parameter and fire up the regex engine for every number checked.

On the other hand, if you mean "it would be overkill for me to use a hash because I don't understand them and I do understand grep," that makes sense. Except now you do know why to use a hash. :)

I'm afraid your seond implicitly assigend $_ overwrites the first one. You could try this:

sub change_punch_process
   {
      my %numbers;
      foreach (my $par = $q->param )
      {
         $numbers{$1}++ if /^([0-9]+)/;
      {
   }
# now you can get all your numbers with keys %numbers
[download]

--
Cheers, Joe

sub change_punch_process {
      my @numbers;
      foreach ( $q->param ){
         if (/^([0-9]+)/){
            $number=$1;         
            push @numbers,$number unless grep{$_==$number}@numbers; 
         }
      }
}
[download]