As mentioned previously, using sprintf will allow you to format your data correctly. As for the sort, the following would work, if I understood your requirements correctly, but I seriously doubt that it's the most efficient way to do things.

use strict;
use Data::Dumper;
my @expos;
chomp( @expos = <DATA> );

@expos = map  { "$_->[0]|$_->[1]/$_->[2]|$_->[3]/".(join "|", @{$_}[4.
+.$#$_]) }
         sort { $a->[1] <=> $b->[1] or $a->[2] <=> $b->[2] }
         map  { [ split '[|/]', $_, 8 ] } @expos;

print Dumper \@expos;

__DATA__
Macworld Expo|2/7|1/12|San Francisco|CA|http://www.macworldexpo.com 
Another Expo|1/27|2/5|San Francisco|CA|http://www.anotherexpo.com 
Some Expo|1/6|1/12|San Francisco|CA|http://www.macworldexpo.com
[download]

Generally, one should avoid a Schwartzian transform if the you do not have expensive data munging, but it's the first thing that occurred to me and I think they look rather cool :)

Cheers,
Ovid

Vote for paco!

Join the Perlmonks Setiathome Group or just click on the the link and check out our stats.