On 2018/06/06 2:08, Alvaro Herrera wrote:
> On 2018-Jun-05, Amit Langote wrote:
>
>> On 2018/06/05 16:41, Ashutosh Bapat wrote:
>>> On Tue, Jun 5, 2018 at 1:07 PM, Amit Langote
>>> <Langote_Amit_f8@lab.ntt.co.jp> wrote:
>>>> On 2018/06/05 1:25, Alvaro Herrera wrote:
>>>>> In the meantime, my inclination is to fix the documentation to point out
>>>>> that AFTER triggers are allowed but BEFORE triggers are not.
>>>>
>>>> Wasn't that already fixed by bcded2609ade6?
>
> Ah, yes, so it's already OK.
>
>>> Thought correct that suggestion is problematic for upgrades. Probably
>>> users will create triggers using same function on all the partitions.
>>> After the upgrade when we start supporting BEFORE TRIGGER on
>>> partitioned table, the user will have to drop these triggers and
>>> create one trigger on the partitioned table to have the trigger to be
>>> applicable to the new partitions created.
>>
>> Hmm yes, maybe there is scope for rewording then.
>
> Reading that subsection in its entirety, I'm not sure how to do better
> -- it's all about restrictions that currently exist and we think we
> could do better in the future, with the exception of the one about an
> UPDATE/DELETE not seeing the updated version after a row migrating to
> another partition. One idea would be to split it into its own list of
> issues; something like:
>
> "The following limitations apply, and might be lifted in the future:
> - no way to create exclusion constraint
> - foreign keys cannot refer to partitioned tables
> - BEFORE row triggers are not supported
>
> The following caveat applies to partitioned tables:
> - When an UPDATE causes a row to move ..."
>
> In other words, make a distinction of things we expect to change soon
> (which might be too optimistic), vs. others we don't expect to change.
>
> Or we could leave it alone; any behavior change would be called out in
> the future version's release notes anyway. This is what I propose.
That works for me. :)
Thanks,
Amit