Re: SQL/MED - core functionality

On 25.11.2010 18:28, Tom Lane wrote:
> Heikki Linnakangas<heikki(dot)linnakangas(at)enterprisedb(dot)com> writes:
>> Hmm, I see, cached plans are planned in a shorter-lived context first,
>> and copied to permanent storage afterwards. Needs more thought then.
>> Maybe the FDW needs to provide a copyFdwPlan() function to copy FdwPlans
>> returned by that FDW.
>
> Or just specify a format for the extra information. Perhaps it could be
> thought of as being a value of type bytea? Obviously we can't just have
> a fixed amount of info, but maybe a blob with a length word is enough.

That seems quite awkward to work with. Let's at least make it a Node *,
so that you can store a Value or List there, or anything else that
already has copyObject support.

I think the PostgreSQL FDW would want to store the remote query there.
But it's not a stretch that you want to use parameter markers in the
remote query, with the parameter values determined at runtime. In that
case you'd also store a list of Exprs for the parameter values (Hmm,
BeginScan needs an ExprContext for that..). This is very hand-wavy, but
I think we'll hit the wall with a single blob pretty quickly.

--
Heikki Linnakangas
EnterpriseDB http://www.enterprisedb.com