Re: Cleaning up the INET/CIDR mess

* Greg Stark (gsstark@mit.edu) wrote:
> I have a question in a different direction. What is the meaning of the network
> mask in the inet data type anyways? Hosts don't have network masks, only
> networks.
>
> If we could store inet in four bytes it would be vastly more efficient both in
> disk space usage and in cpu at runtime.
>
> I think it would also clear up the perpetual user confusion between the two
> datatypes. I posit that the main source of the confusion is that currently
> Postgres lets you use inet for everything, even if what you're really storing
> is a network address range which is what the cidr datatype is really for.

I wholeheartedly agree with this.  It also makes the only cast option
from inet to cidr to be with a /32 and thus there's no zeroing of bits.
It would then be nice to have a function to which you pass in a cidr and
a netmask and say "give me the CIDR this CIDR is in with this mask".
With the inet-to-cidr implicit cast you could use this function to get
the CIDR you want without having to worry about the implicit cast
throwing data away.  The cidr-to-inet is then throwing away the mask and
so I'm not sure we should have an implicit cast in that direction but
honestly I wouldn't complain if we did.
Thanks,
    Stephen