Re: Question on creating keys on partitioned tables - Mailing list pgsql-general

From Siddharth Jain
Subject Re: Question on creating keys on partitioned tables
Date
Msg-id CAPqV3pTuTV6qPpvWT8f67yd8EPP79vk2Ww5ddJkwmEdWCCQdXA@mail.gmail.com
Whole thread Raw
In response to Re: Question on creating keys on partitioned tables  (Laurenz Albe <laurenz.albe@cybertec.at>)
Responses Re: Question on creating keys on partitioned tables  (Tom Lane <tgl@sss.pgh.pa.us>)
List pgsql-general
Thanks Laurenz.

I think the two are equivalent. If not, could you please explain why? 

On Fri, Mar 31, 2023 at 6:46 AM Laurenz Albe <laurenz.albe@cybertec.at> wrote:
On Thu, 2023-03-30 at 17:05 -0700, Siddharth Jain wrote:
> I have this question. Say I create a partitioned table on column X.
>
> Option 1:
>
> I add a primary key on (X,Y). Y is another column. Even though Y is a globally unique PK (global meaning it is unique across partitions, not just in one partition), Postgres does not allow me to
> create a PK on Y in a partitioned table.
>
> Option 2:
>
> I add PK on Y on each of the partitions
>
> Are these not equivalent? If not, which is better and why?

No, they are not equivalent.

Option 2 comes closer to guaranteeing uniqueness for column X, so use that.

> PS: This is what my best friend had to say:
>
> [...] If you are using the "table inheritance" approach [...]

Don't even consider that.  Declarative partitioning is so much better.

Yours,
Laurenz Albe

pgsql-general by date:

Previous
From: Arnaud Lesauvage
Date:
Subject: RE: Cannot terminate backend
Next
From: Paul Ramsey
Date:
Subject: Re: Cannot terminate backend