At Thu, 09 Jan 2020 14:12:25 -0500, Tom Lane <tgl@xxxxxxxxxxxxx> wrote in > Stephen Frost <sfrost@xxxxxxxxxxx> writes: > > I agree with all of that- but it seems like we've got people asking for > > it, and I can understand why they are, and therefore I'd be inclined to > > provide a way for users to get that behavior and I'm even inclined to > > say that pushing down should be the 'default' if everything else is > > equal and it's possible to do so. > > [ shrug... ] I think that's a completely arbitrary value judgment > unsupported by any facts. Furthermore, it'd greatly expand our exposure > to problems with "we thought this function behaves the same on the remote > end, but it doesn't". That hazard makes me feel that the default should > *never* be to encourage pushing down, unless there's a pretty clear > potential win to justify the risk of breakage. > > If SQL/MED has a notion of a "foreign function" that is only executable on > the remote side, and that fails if you try to execute it locally, then > it'd make sense to implement that feature and then expect postgres_fdw to > try hard to push down such calls. But we don't have that (yet). I'm > unconvinced that ordinary functions ought to be treated as if remote > execution is preferable. Isn't ROUTING MAPPING [1] that? Definers should define one at their own risk as table constraints are. [1] https://www.postgresql.org/message-id/CAD21AoCSzZO%2Bd3EL-9w%3D97aC28%3DRb3%2BBh77wRLcP7HCH8-Bn9A%40mail.gmail.com regards. -- Kyotaro Horiguchi NTT Open Source Software Center
