print "Content-type: text/html\n";
print "Set-Cookie: MyCookie=Something\n\n";
[download]

I can understand why you might need this, since some webhosters don't even provide CGI.pm. You can retrieve most cookies with a function like this.

sub GetCookies
{
  %decode = ('\+'=>' ','\%3A\%3A'=>'::','\%26'=>'&','\%3D'=>'=',
             '\%2C'=>',','\%3B'=>';','\%2B'=>'+','\%25'=>'%');

  my %Cookies = ();
  foreach (split(/; /,$ENV{'HTTP_COOKIE'})) {
    my ($cookie,$value) = split(/=/);
    foreach $ch ('\+','\%3A\%3A','\%26','\%3D','\%2C','\%3B','\%2B','\
+%25') {
      $cookie =~ s/$ch/$decode{$ch}/g;
      $value =~ s/$ch/$decode{$ch}/g;
    }
    $Cookies{$cookie} = $value;
  }
  return %Cookies;
}
[download]

where the Cookies hash returned contains the cookies
----
FightLiteracy.com
http://www.fightliteracy.com

When some hosts explicitly remove CGI.pm from the core package, they're plain stupid...
Parsing cookies is pretty similar to parsing forms, as it can be broken quite easily if you're not carefull. I'm confident your code is flawded in one way or another (altho I'm not gonna chase the hole). Check merlyn's No excuses about not using CGI.pm. I confess that most of my older CGI code didn't use CGI.pm for that kinda stuff, but now I've payed my sin.

Maybe I should do a use CGI; #damnit! bat too...

Needless to say I know how to parse cookies manually, since I set it manually in my original node :))

Greetz
Beatnik
... Quidquid perl dictum sit, altum viditur.

I'm pretty sure that the second book's name is "CGI programming with perl". There's not very much, but a little Sample-Header: Page 32 (2nd edition). Some more detailed information about cookies with cgi.pm can be found at page 286. Have fun :)
octopus
--
GED/CC d-- s:- a--- C++(+++) UL+++ P++++$ L++>++++ E--- W+++@ N o? K? w-- O- M-(+) V? !PS !PE !Y PGP+(++) t-- 5 X+ R+(+++) tv+(++) b++@ DI+() D+ G++ e->+++ h!++ r+(++) y+