On Sun, 2020-09-13 at 19:04 +0300, Nikolay Shaplov wrote:
> Moving reloptions to AM code is the goal I am slowly moving to. I've
> started
> some time ago with big patch
> https://commitfest.postgresql.org/14/992/ and
> have been told to split it into smaller parts. So I did, and this
> patch is
> last step that cleans options related things up, and then actual
> moving can be
> done.
Thank you for working on this.
Can you outline the plan for moving these options to the table AM to
make sure this patch is a step in the right direction?
I was trying to work through this problem as well[1], and there are a
few complications.
* Which options apply to any relation (of any table AM), and which
apply to only heaps? As far as I can tell, the only one that seems
heap-specific is "fillfactor".
* Toast tables can be any AM, as well, so if we accept new reloptions
for a custom AM, we also need to accept options for toast tables of
that AM.
* Implementation-wise, the bytea representation of the options is not
very easy to extend. Should we have a new text field in the catalog to
hold the custom options?
Regards,
Jeff Davis
[1]
https://www.postgresql.org/message-id/43c6ec161f930e385dbc3169a065a917cfc60714.camel%40j-davis.com