On Thu, Apr 24, 2014 at 11:30:15AM -0400, Tom Lane wrote:
> Hannu Krosing <hannu@2ndQuadrant.com> writes:
> > On 04/24/2014 04:57 PM, Tom Lane wrote:
> >> The reason why there's no generation function in core is that
> >> there is no standardized,
> >> guaranteed-to-produce-a-universally-unique-value generation
> >> algorithm. That was the reason for not putting something in core
> >> when the type was first created, and I do not see that the
> >> technology has advanced.
>
> > Why can't we implement all 5 variants from
> > http://en.wikipedia.org/wiki/Universally_unique_identifier and
> > just warn about the dangers in documentation ?
>
> Essentially, that would mean carrying around our own implementation
> of libuuid --- which includes a bunch of not-terribly-portable
> stuff, such as discovering the machine's MAC address(es). That's
> not really something I want to see us putting project manpower into.
We don't need to do the not-terribly-portable stuff in the first
round. For that, there could still be a bundled extension.
The point is that UUIDs are nowhere near as usable as users have the
right to expect, and we should fix that.
> I wonder what it would take to adapt contrib/uuid-ossp to work on
> top of some other popular implementation of that code. We pretty
> much bet on the wrong horse when we picked the OSSP library to
> depend on, but otherwise I think the principle of using an external
> library was good.
So long as we can pick another horse later, sure.
Cheers,
David.
--
David Fetter <david@fetter.org> http://fetter.org/
Phone: +1 415 235 3778 AIM: dfetter666 Yahoo!: dfetter
Skype: davidfetter XMPP: david.fetter@gmail.com
iCal: webcal://www.tripit.com/feed/ical/people/david74/tripit.ics
Remember to vote!
Consider donating to Postgres: http://www.postgresql.org/about/donate
