Re: Full GUID support - Mailing list pgsql-hackers

From ktm@rice.edu
Subject Re: Full GUID support
Date
Msg-id 20110712204028.GQ14305@staff-mud-56-27.rice.edu
Whole thread Raw
In response to Re: Full GUID support  (Andrew Dunstan <andrew@dunslane.net>)
Responses Re: Full GUID support
List pgsql-hackers
On Tue, Jul 12, 2011 at 04:29:33PM -0400, Andrew Dunstan wrote:
> 
> 
> On 07/12/2011 03:44 PM, Joshua D. Drake wrote:
> >>
> >>What about extensions makes them less usable?
> >
> >
> >It is an extra step, that is less usable. Does it matter? Shrug, I
> >know I hate having to type apt-get just to use xyz, does it mean
> >it is a big deal? Probably not.
> 
> 
> By that argument we wouldn't have any extensions at all, just a
> monolithic product. I don't think that would be an advance.
> 
> cheers
> 
> andrew
> 

For me, the criteria I like to use for core functionality are:

1. It is available with a common definition from a number of DB products.
With a UUID, it's size/structure is predefined and this allows a dump from
another SQL product to be loaded into a PostgreSQL DB.

2. It would benefit from the tighter integration with the core DB for
either performance or development use.

3. It is a feature where the "extra step" is an unexpected nuisance.

That is why I think having the UUID generators be a contrib module
is the correct place for them to be, but the UUID type is better as
a core function.

Regards,
Ken


pgsql-hackers by date:

Previous
From: Andrew Dunstan
Date:
Subject: Re: Full GUID support
Next
From: Pavan Deolasee
Date:
Subject: Single pass vacuum - take 1