On Thursday 15 Dec 2005 22:15, Wakko Warner wrote:
> Graeme Fowler wrote:
> > On Mon 12 Dec 2005 14:15:50 GMT , Philip Hazel <ph10@???>
wrote:
> > >This is more evidence that convinces me that the best thing is to
> > > insist on 4 components and give an error for anything else.
> >
> > That sounds the best way to me, since it's definitely not ambiguous
> > and greatly simplifies the code required for validation. It's also
> > completely platform (and therefore developer/vendor) independent,
> > and therefore portable.
> >
> > Gets my vote :)
>
> I've been reading over this. I've noticed that some places that list
> IPs (like by region) may list them as a.b/16 or a/8 or something
> like that.
>
> Is there any place the specifically describes CIDR notation?
RFC1519 "Classless Inter-Domain Routing (CIDR): an Address Assignment
and Aggregation Strategy",
<
ftp://ftp.rfc-editor.org/in-notes/rfc1519.txt> might be a good place
to start :-)
--
Rob Skedgell <rob+exim-users@???>
