This is a one liner my friend Michael Anthon (MrChook) made back in 2006 for sorting the output of fdupes by file size.
It's been handy over the years so I though I would share it here.
cat fdupesoutput.txt | perl -00 -le 'print map $->[1], sort{$a->[0]<=>$b->[0]} map [/(\d+) bytes/, $], <>'