Re: Function scan push-down using SQL/MED syntax

Hitoshi Harada <umi(dot)tanuki(at)gmail(dot)com> wrote:

> I've not read any part of code yet but have comments:
>
> * I'm not very happy with "Getting tuples from the foreign server"
> section. Present tuplestore isn't quite efficient and putting all
> tuples into TSS adds devastating overhead. In principle, storing
> tuples doesn't match SQL exectuor model. So something like cursor is
> needed here.

Sure, but your optimization requires some extensions in libpq protocol.
We could send HeapTuple in a binary form if the remote and the local
server uses the same format, but the present libpq can return tuples
only as text or libpq-specific binary forms (that is not a HeapTuple).

> * In FDW routines section there seems only methods to read out data,
> but I'd suggest there should be coverage of all CRUD operation.
> Sometimes dropping foreign data is so specific that only FdwRoutines
> know the way.

There is only SELECT in the SQL standard, where CREATE FOREIGN TABLE
means a symbolic link for the external data. But we also could support
CREATE and DROP in the routines, and it means CREATE/DROP FOREIGN TABLE
will actually create or drop external data.

Also, we could add methods for INSERT, UPDATE and DELETE -- but it
requires additional consideration that which key we should use to
determine the modified tuples; PostgreSQL uses CTID as a physical key,
but some of FDW might have completely different keys.

Regards,
---
Takahiro Itagaki
NTT Open Source Software Center