Hi,
On 2019-05-15 13:53:26 -0400, Tom Lane wrote:
> FWIW, given the conflict against "EXPLAIN EXECUTE prepared_stmt_name",
> we should probably just drop the whole idea. It seemed like a great
> idea at the time, but it's going to confuse people not just Bison.
I'm not particularly invested in the idea of renaming ANALYZE - but I
think we might be able to come up with something less ambiguous than
EXECUTE. Even EXECUTION might be better.
> So ... never mind that suggestion. Can we get anywhere with the
> rest of it?
Yes, please. I still think getting rid of
if (es->buffers && !es->analyze)
ereport(ERROR,
(errcode(ERRCODE_INVALID_PARAMETER_VALUE),
errmsg("EXPLAIN option BUFFERS requires ANALYZE")));
and
/* check that timing is used with EXPLAIN ANALYZE */
if (es->timing && !es->analyze)
ereport(ERROR,
(errcode(ERRCODE_INVALID_PARAMETER_VALUE),
errmsg("EXPLAIN option TIMING requires ANALYZE")));
and then changing the default for BUFFERs would be good. I assume they'd
still only apply to query execution.
Althouh, in the case of BUFFERS, I more than once wished we'd track the
plan-time stats for buffers as well. But that's a significantly more
complicated change.
Greetings,
Andres Freund