Robert Haas <robertmhaas@gmail.com> writes:
> On Fri, Feb 11, 2011 at 12:35 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
>> OK, let me see if I can summarize what I think we've agreed to:
>>
>> CREATE syntax is extended to
>>
>> CREATE EXTENSION extname [WITH] [SCHEMA s] [VERSION v] [FROM oldv]
> It strikes me that if you used the same options syntax here that we're
> already using for EXPLAIN and VACUUM and COPY, you wouldn't have to
> worry about adding keywords for current or future options.
Hmm. You have a point, and there's some precedent for this in our other
non-standard CREATE commands such as CREATE OPERATOR and CREATE
AGGREGATE. On the other hand, we have no precedent for handling ALTER
syntaxes that way. Also, I think most people feel that the CREATE
OPERATOR and CREATE AGGREGATE syntaxes are ugly, not-very-SQL-ish beasts
carried over from PostQUEL days.
On the whole I have a weak preference for leaving it as above, but would
readily yield to a consensus to do the other.
One minor point is that I was planning to drop the opt_equals from the
syntax --- it doesn't fit at all with the FROM case.
regards, tom lane