Re: WIP: CREATE TABLE AS / WITH DATA

From: Gavin Sherry <swm(at)linuxworld(dot)com(dot)au>
To: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
Cc: Neil Conway <neilc(at)samurai(dot)com>, Alvaro Herrera <alvherre(at)dcc(dot)uchile(dot)cl>, pgsql-patches <pgsql-patches(at)postgresql(dot)org>, Gavin Sherry <swm(at)alcove(dot)com(dot)au>
Subject: Re: WIP: CREATE TABLE AS / WITH DATA
Date: 2004-09-23 04:37:31
Message-ID: Pine.LNX.4.58.0409231434530.10839@linuxworld.com.au
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-patches

On Thu, 23 Sep 2004, Tom Lane wrote:

> Neil Conway <neilc(at)samurai(dot)com> writes:
> > On Wed, 2004-09-22 at 23:00, Alvaro Herrera wrote:
> >> Could that include supporting SELECT INTO as well as both types of
> >> CREATE TABLE AS?
>
> > Right; my thinking is to have the parser construct SELECT INTO as a
> > CreateTableAsStmt. That way all the code for creating the "into"
> > relation is centralized in one place.
>
> Another thing that would be really nice would be to get rid of all the
> warty special cases for SELECT INTO in executor/execMain.c. I have
> thought about handling this stuff in a new tuple receiver type (cf
> tcop/dest.h, dest.c) but haven't really pursued it. I also have some

That's a pretty good idea. I was also thinking about this and wondering
about the optimisations we could make. Skip WAL and just fsync() the file
at the end (or dump whole blocks into WAL if we're archiving). Its a bit
of a niche case though.

Gavin

In response to

Browse pgsql-patches by date

  From Date Subject
Next Message Neil Conway 2004-09-23 07:42:33 trivial cleanup: ExecProcAppend
Previous Message Tom Lane 2004-09-23 04:15:44 Re: WIP: CREATE TABLE AS / WITH DATA