Tom Lane wrote:
> Tom Raney <raneyt@cecs.pdx.edu> writes:
>> Why does the index scan for tenk1 include a path key from
>> onek.unique2? Is it implying an equivalence there?
>
>> bench=# explain select * from tenk1 JOIN onek ON
>> tenk1.unique2=onek.unique2;
>
> Yes, for an example like that the planner knows that tenk1.unique2 and
> onek.unique2 will have equal values in any valid join row, so it's okay
> to suppose that a sort by one is the same as a sort by the other. So
> the pathkey items actually reference sets of variables
> {tenk1.unique2, onek.unique2}
> not just individual variables.
Thanks.
>
>> RELOPTINFO (tenk1): rows=10000 width=244
>> path list:
>> SeqScan(tenk1) rows=10000 cost=0.00..434.00
>> IdxScan(tenk1) rows=10000 cost=0.00..583.25
>> pathkeys: ((tenk1.unique2, onek.unique2)) <---
>
>> cheapest startup path:
>> SeqScan(tenk1) rows=10000 cost=0.00..434.00
>
>> cheapest total path:
>> SeqScan(tenk1) rows=10000 cost=0.00..434.00
>
> Hm, I don't recognize this output format, is it coming from some custom
> code?
Yes, it is. I thought it was easier to read the OPTIMIZER_DEBUG
output with the relation names instead of the relation indexes. I
will post a patch against CVS HEAD if you think it will help others.
-Tom