arbitrary size integer/float math package

 Math::BigInt and Math::BigFloat allow one to declare big integers / floating
 point numbers as
 .
  $i = new Math::BigInt '123_456_789_123_456_789';
  $i = new Math::BigFloat '12_3.456_789_123_456_789E-2';
 .
 All operators are overloaded; operations with overloaded operators preserve
 the arguments.
 .
 Math::BigInt is also in perl core, this package provides newer versions
 separately.