Re: pg_dump --binary-upgrade vs. ALTER TYPE ... DROP ATTRIBUTE

Robert Haas <robertmhaas(at)gmail(dot)com> writes:
> On Wed, Apr 20, 2011 at 10:51 PM, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
>> How about "ALTER TABLE tabname [NOT] OF TYPE typename"? It's at least a
>> smidgeon less ambiguous.

> I thought of that, but I hate to make CREATE TABLE and ALTER TABLE
> almost-but-not-quite symmetrical.

Oh, good point.

> But one might well wonder why we didn't decide on:
> CREATE TABLE n OF TYPE t;
> ...rather than the actual syntax:
> CREATE TABLE n OF t;
> ...which has brevity to recommend it, but likewise isn't terribly clear.

> I presume someone will now refer to a standard of some kind....

SQL:2008 11.3 <table definition>, the bits around <typed table clause>
to be specific.

The SQL committee's taste in syntax is, uh, not mine. They are
amazingly long-winded in places and then they go and do something
like this ...

regards, tom lane