Re: pgsql_fdw in contrib

On Thu, Jun 14, 2012 at 7:29 AM, Shigeru HANADA
<shigeru(dot)hanada(at)gmail(dot)com> wrote:
> I'd like to propose pgsql_fdw, FDW for PostgreSQL, as a contrib module
> in core, again.  This patch is basically rebased version of the patches
> proposed in 9.2 development cycle, and contains some additional changes
> such as concern about volatile variables for PG_CATCH blocks.  In
> addition, I combined old three patches (pgsql_fdw core functionality,
> push-down support, and analyze support) into one patch for ease of
> review. (I think these features are necessary for pgsql_fdw use.)
>
> After the development for 9.2 cycle, pgsql_fdw can gather statistics
> from remote side by providing sampling function to analyze module in
> backend core, use them to estimate selectivity of WHERE clauses which
> will be pushed-down, and retrieve query results from remote side with
> custom row processor which prevents memory exhaustion from huge result set.
>
> It would be possible to add some more features, such as ORDER BY
> push-down with index information support, without changing existing
> APIs, but at first add relatively simple pgsql_fdw and enhance it seems
> better.  In addition, once pgsql_fdw has been merged, it would help
> implementing proof-of-concept of SQL/MED-related features.

I can't help but wonder (having been down the contrib/core/extension
road myself) if it isn't better to improve the facilities to register
and search for qualified extensions (like Perl CPAN) so that people
looking for code to improve their backends can find it. That way,
you're free to release/do xyz/abandon your project as you see fit
without having to go through -hackers. This should also remove a lot
of the stigma with not being in core since if stock postgres
installations can access the necessary modules via CREATE EXTENSION, I
think it will make it easier for projects like this to get used with
the additional benefit of decentralizing project management.

merlin