Re: unique across two tables

From: Merlin Moncure <mmoncure(at)gmail(dot)com>
To: David Johnston <polobo(at)yahoo(dot)com>
Cc: Misa Simic <misa(dot)simic(at)gmail(dot)com>, Tarlika Elisabeth Schmitz <postgresql4(at)numerixtechnology(dot)de>, pgsql-general(at)postgresql(dot)org
Subject: Re: unique across two tables
Date: 2011-06-22 16:06:46
Message-ID: BANLkTikEi2Rxkck=zawZQ8Zz24qdZY3aOA@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

On Wed, Jun 22, 2011 at 8:34 AM, David Johnston <polobo(at)yahoo(dot)com> wrote:
> Your statement is utterly false simply by virtue of the documentation.
>
>
>
> Inserts never propagate to other tables in a hierarchy
>
> Indexes/Constraints only apply to individual tables
>
>
>
> Since inserts do not propagate the base table never sees the records and
> thus the index on the base table cannot enforce uniqueness.
>
>
>
> Other caveats and restrictions to the current inheritance implementation can
> be found in section 5.8 (Inheritance) of the documentation.  It is in the
> “Data Definition” chapter.

yeah -- postgresql table inheritance is completely useless for doing
what most people instinctively want it to do...it's a half implemented
kludge that dates back forever. however, it does have a few quirky
neat uses and also underpins the table inheritance feature so we're
stuck with the current behavior.

merlin

In response to

Responses

Browse pgsql-general by date

  From Date Subject
Next Message Rob Sargent 2011-06-22 16:27:43 Re: building 9.1 on suse-11.4 (64bit)
Previous Message Merlin Moncure 2011-06-22 15:58:50 Re: data compression in protocol?