Shane Ambler <[EMAIL PROTECTED]> writes:
> Tom Lane wrote:
>> It seems clear to me that we ought not impose a bias unless the join
>> type is such that both directions of hashing are feasible.
> I think that the selected sort (or at least the merge join) is incorrect
> - the column sorted (or bot
Tom Lane wrote:
Shane Ambler <[EMAIL PROTECTED]> writes:
I am putting together searches on the catalog info and came up with a
select that was rather slow and I noticed that in the explain analyze
there is a sort step on one of the left joins which I don't think
belongs there.
Well, it's cer
Shane Ambler <[EMAIL PROTECTED]> writes:
> I am putting together searches on the catalog info and came up with a
> select that was rather slow and I noticed that in the explain analyze
> there is a sort step on one of the left joins which I don't think
> belongs there.
Well, it's certainly nece
I am putting together searches on the catalog info and came up with a
select that was rather slow and I noticed that in the explain analyze
there is a sort step on one of the left joins which I don't think
belongs there.
I found the small error in my query (using tl.oid instead of tr.oid and
