Adam Zegelin writes:
> On 19/03/2013, at 12:40 PM, Tom Lane wrote:
>> In this particular case, it won't generate a cross-product join of l1
>> and l2 because there's a heuristic that says that's unlikely to be a
>> good idea.
> Is this heuristic a tuneable parameter, or something that would requ
Thanks for your assistance Tom.
On 19/03/2013, at 12:40 PM, Tom Lane wrote:
> I think you missed my point: you should not be insisting on a maximal
> set of required outer rels.
I’ve started generating multiple paths for the web services that support it,
which works great.
> In this particula
Adam Zegelin writes:
> My path generation logic seems to work:
> baserel->cheapest_parameterized_paths = (
>{FOREIGNPATH
>:pathtype 120
>:parent_relids (b 3)
>:required_outer (b 1 2)
>:rows 500
>:startup_cost 0.00
>:total_cost 0.00
>:pathkeys <>
>:fdw_private <
Tom,
Thank you for your prompt reply. Your advice has pointed me in the right
direction.
I now have the wrapper identifying columns that are inputs to the web service,
and thus parameterisable. The ec_classes, left_join_clauses and
right_join_clauses trees are scanned for Var exprs that match
Adam Zegelin writes:
> Some service endpoints have no concept of unqualified queries. In the example
> above, a âsequence scanâ of Bing is a not possible.
In that case, you shouldn't be generating such a path. But keep in mind
that this may lead to failure to produce any plan at all for som
