Hello David,
thanks to you for your feedback.
> On occasion, I write scripts to
> process objects in firewall databases. In the past I've used Perl, but
> now I am using Ruby. In Perl, I've used Net::IP. It has a couple of
> things that would fit nicely in IPAddress.
>
> First, arbitrary range objects. Not defined by class or mask, just two
> addresses defining an arbitrary range.
This is a very interesting suggestion and I'm actually working on some nice
addition to include an IPAddress::Range class for IPAddress 0.9.0, to
be released
probably in September.
> Second, a great method called 'overlaps'. Here is description from
> Net::IP doc..
>
> "Check if two IP ranges/prefixes overlap each other. The value
> returned by the function should be one of: $IP_PARTIAL_OVERLAP (ranges
> overlap) $IP_NO_OVERLAP (no overlap) (IP_A_IN_B_OVERLAP (range2
> contains range1) $IP_B_IN_A_OVERLAP (range1 contains range2)
> $IP_IDENTICAL (ranges are identical) undef (problem)"
Would you consider this method for Range object types only or for canonical
IPAddress::IPv4 (or IPv6) objects? Because I'm very favorable in the
first case, it will
be a joy to implement in Ruby.
Thanks very much for your suggestions David. Stick around and we'll
see what 0.9.0 will bring! :)