Re: ALTER EXTENSION ... UPGRADE;

Tom Lane <tgl@sss.pgh.pa.us>

From: Tom Lane <tgl@sss.pgh.pa.us>
To: jd@commandprompt.com
Cc: Andrew Dunstan <andrew@dunslane.net>, Josh Berkus <josh@agliodbs.com>, Dimitri Fontaine <dimitri@2ndQuadrant.fr>, "David E. Wheeler" <david@kineticode.com>, PostgreSQL Hackers <pgsql-hackers@postgresql.org>
Date: 2010-12-10T20:54:08Z
Lists: pgsql-hackers
"Joshua D. Drake" <jd@commandprompt.com> writes:
> On Fri, 2010-12-10 at 15:42 -0500, Andrew Dunstan wrote:
>> On 12/10/2010 03:24 PM, Josh Berkus wrote:
>>> Also, once extensions and pgxn are operating full swing, I see contrib
>>> going away anyway ...

>> We've heard this before, but I'm still quite skeptical about it. Quite 
>> apart from anything else we should keep enough extensions in core to 
>> test the extension mechanism, as well as to provide examples as part of 
>> the base distribution. Some (e.g. hstore and citext) should probably 
>> move into core. Others like pgcrypto are probably in just the right 
>> place as they are.

> I hope that contrib goes away. I agree with your assertion that things
> like hstore and citext shoudl be in core but it is my hope that with
> extensions and pgxn, there will be no reason for contrib to exist at
> all.

I agree with Andrew --- we're going to need a collection of "standard
extensions" if only for testing purposes.  It may someday not be called
contrib, but it'll still be there.

			regards, tom lane