The current implementation is not very clean or accurate. It is not accurate
because the uncompressed size is what is being used. It should instead be the
compressed sizes used.
Original issue reported on code.google.com by shouldab...@gmail.com on 10 Oct 2010 at 8:38
Original issue reported on code.google.com by
shouldab...@gmail.com
on 10 Oct 2010 at 8:38