Re: Difference arrays.

Since you're concerned about memory, you could do something like a Merge Sort.
Memory: O(1) (Not counting @a, @b and @c)
Speed: O(A+B) (Assuming @a and @b already sorted. As good as the other solutions)

my @a = sort { $a <=> $b } (43,43,44);
my @b = sort { $a <=> $b } (43,43);

my @c;
while (@a && @b) {
   if ($a[0] < $b[0]) {
      push @c, shift @a;
   }
   elsif ($a[0] > $b[0]) {
      die "Bad data";
   }
   else {
      shift @a;
      shift @b;
   }
}

push @c, $_ for @a;
die "Bad data" if @b;
[download]

A trivial change makes it non-destructive.

Update: Fixed bug mentioned in replies. Tested.

Comment on Re: Difference arrays. Download Code

Replies are listed 'Best First'.
Re^2: Difference arrays. by GrandFather (Saint) on Sep 04, 2008 at 21:39 UTC
That works better if the pops are shifts. .oO(How many times have I been caught by that!) Perl reduces RSI - it saves typing	[reply]
Re^2: Difference arrays. by kyle (Abbot) on Sep 04, 2008 at 20:23 UTC
When I run this, I end up with... `@a = ( 43 ); @b = (); @c = ( 43 );` [download] None of those is the OP's desired result, `( 44 )`. Am I missing something?	[reply] [d/l] [select]