Proposal [Trac #1212]: add IPv6 support to network library

[Bryan O'Sullivan]

Sven Panne wrote:

> Another option would be to export all constants on all platforms. If a 
> constant is not supported by the underlying platform, use a value which is 
> guaranteed to result in e.g. a usual I/O error when used with a network 
> function.

That's a possibility, but I think that the best place to start from is 
the portable constants from RFC 2553, which every platform supports. 
That way, the functionality becomes available reasonably soon in a 
useful form.  Trying to figure out how to deal acceptably with constants 
that are not portable looks like a rathole, one I'd rather descend into 
later :-)

	<b