Tom Lane <tgl@sss.pgh.pa.us> writes:
> Greg Stark <gsstark@mit.edu> writes:
> > So should I set up a nested scan, essentially implementing a nested loop? or
> > should I gather together all the children in a list?
>
> I'd use the predigested form of the constraints attached to the Relation
> tupledescs, cf. RelationBuildTupleDesc, equalTupleDescs. It might be
> worth refactoring equalTupleDescs so you could share code --- ISTM what
> you're trying to implement is something like a "subsetTupleDesc".
Unless I'm missing something that predigested form only has the conbin field.
It doesn't have the name of the constraint nor the other fields like
deferrable and deferred by default. It also doesn't have foreign key
constraints which I'm ignoring now but suggesting that we will want to be
copying to children and checking for in new children in the future.
And subsetTupleDesc seems to be checking that the attributes are in the same
specific order, not that they match by name. That seems like a very different
kind of quality/subset nature than needed here.
--
greg