Re: [HACKERS] New INET and CIDR types

Thus spake Bruce Momjian
> > For sure not the documentation. As for the regression testing, it should
> > be final by Friday. Unfortunately I'm having problems. The last sup
> > seems to have broken the inet type. I'll try to get you in #PostgreSQL
> > and talk about this.
>
> Fixed.

Mostly. There seems to still be a problem with the network function as
I mentioned in the messages to patches. Also, as I mentionwed there,
the output of the functions should be left justified, not right. Also,
I think the netmasklen function should be called masklen. Given the
polymorphic nature of the type system and the fact that it is being
applied to inet or cidr types, I don't think there is any confusion
using the shorter, more familiar term.

> Thomas will do all the markup. Just some info for the docs are needed.
> He should provide you with what he needs.

I just received that. I'll get to work. I also have most of the queries
for the regression tests which I'll send along shortly.

> > Internally it does. Paul said it doesn't matter and I assume it is
> > his function leaving it off though I haven't checked.
>
> I recommend displaying the /32 for cidr. Paul says it is not required,
> but would be 'nice', and I agree.

darcy=> select '1.2.3.4/32'::cidr;
?column?
----------
1.2.3.4/32
(1 row)

darcy=> select '1.2.3.4/32'::inet;
?column?
--------
1.2.3.4
(1 row)

> Sorry to be pressuring, but I want this off the Open Items list before
> Marc has a coronary. His head swells badly, and turns all red. :-)

Perhaps we should hand out plastic sheeting like at Gallagher concerts
just in case. :-)

--
D'Arcy J.M. Cain <darcy(at){druid|vex}.net> | Democracy is three wolves
http://www.druid.net/darcy/ | and a sheep voting on
+1 416 424 2871 (DoD#0082) (eNTP) | what's for dinner.