Horton

Re: altering a column's collation leaves an invalid foreign key

Peter Eisentraut <peter@eisentraut.org>

View thread View on postgr.es
From: Peter Eisentraut <peter@eisentraut.org>
To: jian he <jian.universality@gmail.com>
Cc: Paul Jungwirth <pj@illuminatedcomputing.com>, PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>
Date: 2024-11-15T14:46:15Z
Lists: pgsql-hackers 
On 14.11.24 12:35, jian he wrote:
> On Thu, Nov 14, 2024 at 4:04 PM Peter Eisentraut <peter@eisentraut.org> wrote:
> 
>> I propose that I go ahead with committing the v7 patch (with your typo
>> fixes) and then we continue discussing these other issues afterwards in
>> a separate thread.
>>
> looks good to me.

done

> but in create_table.sgml
>       <para>
>        In addition, when the data in the referenced columns is changed,
>        certain actions are performed on the data in this table's
>        columns.
> <para/>
> I actually want to add a sentence like:
> 
>        If the references columns collation is indeterministic, we use
> bitwise comparison to
>        check if the referenced columns data is being changed.

I think this is also part of the foreign key action behavior that is a 
separate topic.  We can discuss documentation improvements as part of that.

Commits

  1. Fix error code for referential action RESTRICT

  2. doc: Improve description of referential actions

  3. Add tests for foreign keys with case-insensitive collations

  4. Fix collation handling for foreign keys

  5. Clarify a foreign key error message