On Tue, Oct 26, 2010 at 1:42 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
> Robert Haas <robertmhaas@gmail.com> writes:
>> On Tue, Oct 26, 2010 at 1:26 PM, Jeff Davis <pgsql@j-davis.com> wrote:
>>> However, this is orthogonal, I think. I can always ask the user to
>>> specify everything when creating a Range Type, and then we can make them
>>> default to use the interface functions later. Some, like "plus" might be
>>> constant, but people certainly might want to specify alternate
>>> comparators.
>
>> If it were me, I would go design and implement the type interface part
>> first. But it's not.
>
> I agree with Jeff's plan: seems like taking a first cut at the higher
> level is worthwhile, to make sure you know what you need from the
> type-system interfaces.
>
> FWIW, I don't agree with the proposed syntax. We already have a
> perfectly extensible CREATE TYPE syntax, so there is no reason to
> implement this as an ALTER TYPE operation. What's more, altering
> existing datatype declarations is fraught with all kinds of fun
> risks, as we were reminded with the recent enum patch.
Fair enough. I'm not wedded to the syntax (or the order of development).
--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company