On Sun, Jun 5, 2011 at 2:59 PM, Jeff Davis <pgsql@j-davis.com> wrote:
> So, where on this spectrum should range types fall? I think the most
> minimalist would be to only support #1 (and the necessary type IO
> functions); and leave all other functions, operators, and opclasses to
> an extension. That has a lot of appeal, but I don't think we can ignore
> the challenges above.
>
> On the other hand, trying to make it a complete feature in core has
> challenges as well. For instance, even with Range Types, Exclusion
> Constraints aren't practical out-of-the-box unless we also have
> BTree-GiST in core. So there's a snowball effect.
>
> There might also be some middle ground, where its like the minimalist
> approach, but with a few very basic constructors and accessors. That
> would at least make it easier to test, but then to be actually useful
> (with index support, operators, fancy functions, etc.) you'd need the
> extension.
I don't have clear feeling on this question in general, but if we're
going to break this up into pieces, it's important that they be
logical pieces. Putting half the feature in core and half into an
extension just because we can will simplify complicate code
maintenance to no good end. The snowball effect is something to
avoid, and we need to watch out for that, but if the upshot of putting
part of it in core is that the core code can no longer be understood
or maintained because it depends heavily on a bunch of non-core code,
that's not helpful.
--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company