On Tue, Feb 8, 2011 at 6:54 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
> Dimitri Fontaine <dimitri@2ndQuadrant.fr> writes:
>> Tom Lane <tgl@sss.pgh.pa.us> writes:
>>> (I was vaguely imagining that it could share most of the COMMENT
>>> infrastructure --- but haven't looked yet).
>
>> Well the code footprint is quite small already.
>
> Having now looked at it a bit closer, I think the syntax choice is a
> complete wash from an implementation standpoint: either way, we'll have
> a list of bison productions that build AlterObjectExtensionStmt nodes,
> and it goes through the same way after that. I do think that the
> implementation will be a lot more compact if it relies on the COMMENT
> infrastructure (ie, get_object_address), but that's an independent
> choice.
>
> So really it boils down to which syntax seems more natural and/or easier
> to document. As I said, I think a centralized ALTER EXTENSION syntax
> has some advantages from the documentation standpoint; but that's not a
> terribly strong argument, especially given that Dimitri has already done
> a patch to document things the other way.
>
> Preferences anyone?
The closest exstant parallel is probably:
ALTER SEQUENCE foo OWNED BY bar;
I think paralleling that would probably be the most SQL-ish thing to
do, but I can't get excited about it. The ALTER EXTENSION syntax will
be a lot more self-contained, with all of it one part of the grammar
and one part of the documentation. And you could even allow multiple
objects:
ALTER EXTENSION extension_name ADD object-description [, ...];
Which might be handy.
--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company