Re: writable FDWs / update targets confusion

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: Tomas Vondra <tv(at)fuzzy(dot)cz>
Cc: PostgreSQL-development <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: writable FDWs / update targets confusion
Date: 2013-12-06 23:25:05
Message-ID: 22268.1386372305@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Tomas Vondra <tv(at)fuzzy(dot)cz> writes:
> I think that we should make the documentation more explicit about this
> limitation, because the current wording in fdw-callbacks documentation
> seems to suggest it's possible to add such hidden columns. At least
> that's how I read it before running into this.

You can add hidden columns if you've got 'em ;-). What's missing
is the ability to create any hidden columns other than the ones in
standard PG tables. What we most likely need is the ability for
an FDW to override the type assigned to the CTID column at foreign
table creation. (We'd then also need to think about where such a
column could be shoehorned into a tuple, but the catalog support
has to come first.) Alternatively, it might work to append "junk" columns
in the user column numbering domain, which would only exist in runtime
tuple descriptors and not in the catalogs.

regards, tom lane

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Andrew Gierth 2013-12-06 23:32:41 Re: WITHIN GROUP patch
Previous Message Tom Lane 2013-12-06 23:16:29 Re: WITHIN GROUP patch