On Mon, Jan 25, 2016 at 1:21 PM, Corey Huinker <corey.huinker@gmail.com> wrote: >> - We could consider folding fetch_size into "Remote Execution >> Options", but maybe that's too clever. > > If you care to explain, I'm listening. Otherwise I'm going forward with the > other suggestions you've made.
It's just a little unfortunate to have multiple sections with only a single option in each. It would be nice to avoid that somehow.
Revised in patch v3:
* get_option() and get_fetch_size() removed, fetch_size searches added to existing loops.
* Move fetch_size <= 0 tests into postgres_fdw_validator() routine in option.c
* DEBUG1 message removed, never intended that to live beyond the proof of concept.
* Missing regression test mentioned in makefile de-mentioned, as there's nothing to see without the DEBUG1 message.
* Multi-line comment shrunk
(There's a v2 patch that is prior to the change to postgres_fdw_validator() in option.c, but in retrospect that's not interesting to you).
I'm not too keen on having *no* new regression tests, but defer to your judgement.
Still not sure what you mean by remote execution options. But it might be simpler now that the patch is closer to your expectations.