Alvaro Herrera <alvherre@commandprompt.com> writes:
> Bruce Momjian wrote:
>> Oh, psql needs to know before the command is sent? How do we handle it
>> now with CLUSTER?
> We don't, which is exactly the problem. If I'm not mistaken, currently
> psql in autocommit off mode, CLUSTER doesn't start a transaction block,
> which is arguably wrong because some forms of CLUSTER (single-table) are
> able to work within a transaction.
psql could actually tell these apart if it worked just a bit harder.
CLUSTER with no arguments is the one case, CLUSTER with anything after
it is the other. Not sure why we couldn't be bothered to get that
right in psql the first time :-(.
But to get back to the point at hand, I think that there should be some
equally obvious syntactic clue about what CREATE INDEX does --- and
burying an ONLINE keyword near the end of the command doesn't qualify.
regards, tom lane