On Mon, May 15, 2000 at 09:53:34AM -0300, The Hermit Hacker wrote:
> On Mon, 15 May 2000, Hannu Krosing wrote:
> > > Everythingn up to here sounds great ... but this part here totally throws
> > > me off ... this would mean that, unlike now where we rely on *zero*
> > > external code, we essentially gut our internals and create API-links to
> > > someone else's code ...
> >
> > And what will happen to R-trees and Gist access methods ?
> >
> > Will we implement them on top of Berkeley DB :) ?
> >
> > OTOH, it may be good to have a clearly defined API to our own storage
> > manager that could possibly be used separately from the rest of PostgreSQL.
> >
> > So people who can't accept/afford BSDDB licence could use ours instead.
>
> Hrmmm, some sort of --with-berkeley-db configure switch, so by default, it
> uses ours, but if someone wants to do the db code, it could plug-n-play?
>
> That one has possibilities ...
I have not looked at what kind of API would be required, but this
would be my favorite. It sould be nice to be able to choose different
storage methods, optimized for different data patterns or even for
different media.
--
Adam Haberlach |"You have to understand that the
adam@newsnipple.com | entire 'Net is based on people with
http://www.newsnipple.com/ | too much free time on their hands."
