Re: review: FDW API

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: Heikki Linnakangas <heikki(dot)linnakangas(at)enterprisedb(dot)com>
Cc: Postgres - Hackers <pgsql-hackers(at)postgresql(dot)org>, Shigeru HANADA <hanada(at)metrosystems(dot)co(dot)jp>, Robert Haas <robertmhaas(at)gmail(dot)com>, Jan Urbański <wulczer(at)wulczer(dot)org>
Subject: Re: review: FDW API
Date: 2011-02-20 05:28:20
Message-ID: 4759.1298179700@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Heikki Linnakangas <heikki(dot)linnakangas(at)enterprisedb(dot)com> writes:
> On 11.02.2011 22:50, Heikki Linnakangas wrote:
>> I spent some more time reviewing this, and working on the PostgreSQL FDW
>> in tandem. Here's an updated API patch, with a bunch of cosmetic
>> changes, and a bug fix for FOR SHARE/UPDATE.

> Another version, rebased against master branch and with a bunch of small
> cosmetic fixes.

I've applied this after a moderate amount of editorialization.

The question of avoiding extra relkind lookups is still open. We talked
about adding relkind to RangeTblEntry, but I wonder whether adding a new
RTEKind would be a better idea. Haven't researched it yet.

I have a hacked-up version of contrib/file_fdw that I've been using to
test it with. That needs some more cleanup before committing, but I
think it should not take too long. The one thing that is kind of
annoying is that the regression tests need generated files (to insert
absolute paths) and it seems like the PGXS infrastructure doesn't know
how to clean up the generated files afterwards. Anybody have any
thoughts about fixing that?

regards, tom lane

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message 页公倒龙 2011-02-20 06:03:33 How to make the user-defined function or type being accessed in the global database server?
Previous Message Bruce Momjian 2011-02-20 05:02:25 Re: Documentation, window functions