A solution using a Guttman Rosler Transform which might be more efficient if there are a large number of rectangles.

```use strict;
use warnings;

use feature qw{ say };

my @x     = ( 3, 2, 4, 0, 3, 0, 4, 0, 3, 2, 0, 4 );
my @y     = ( 0, 0, 0, 1, 1, 0, 3, 3, 3, 3, 2, 2 );

my @rects =
map { [ split m{:}, substr \$_, 8 ] }
sort
map { my \$packed
= ~ ( pack q{N}, \$x[ \$_ ] )
. pack q{Na*}, \$y[ \$_ ], join( q{:}, \$x[ \$_ ], \$y[ \$_ ] )
}
0 .. \$#x;

my \$n = 0;
say qq{@{[ sprintf q{%2d}, \$n ++ ] }: @\$_} for @rects;

The output.

``` 0: 4 0
1: 4 2
2: 4 3
3: 3 0
4: 3 1
5: 3 3
6: 2 0
7: 2 3
8: 0 0
9: 0 1
10: 0 2
11: 0 3

I hope this is of interest.

Update: Simplified decoration/de-decoration removing need for join and split.

```use strict;
use warnings;

use feature qw{ say };

my @x     = ( 3, 2, 4, 0, 3, 0, 4, 0, 3, 2, 0, 4 );
my @y     = ( 0, 0, 0, 1, 1, 0, 3, 3, 3, 3, 2, 2 );

my @rects =
map { [ unpack( q{N}, ~ \$_ ), unpack( q{x4N}, \$_ ) ] }
sort
map { my \$packed
= ~ ( pack q{N}, \$x[ \$_ ] )
. pack q{N}, \$y[ \$_ ]
}
0 .. \$#x;

my \$n = 0;
say qq{@{[ sprintf q{%2d}, \$n ++ ] }: @\$_} for @rects;