From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
---|---|
To: | Wells Oliver <wells(dot)oliver(at)gmail(dot)com> |
Cc: | pgsql-admin <pgsql-admin(at)postgresql(dot)org> |
Subject: | Re: Struggling with 13->14 and anyarray -> anycompatiblearray |
Date: | 2022-11-04 21:31:29 |
Message-ID: | 2491000.1667597489@sss.pgh.pa.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-admin |
Wells Oliver <wells(dot)oliver(at)gmail(dot)com> writes:
> Thanks. This seems to return aggregates where array_append is used with
> anyarray, but I now see aggregates where e.g. array_cat is used with
> anyarray. Is there some way to generally query aggregates where any array
> function might be looking for anyarray and need to be dropped/re-created
> with anycompatiblearray?
Looking for aggtranstype = 'anyarray'::regtype would give you a hit list,
but beware that not every one of those was changed. Looking at HEAD,
I see
postgres=# select aggfnoid::regprocedure from pg_aggregate where aggtranstype = 'anyarray'::regtype;
aggfnoid
---------------
max(anyarray)
min(anyarray)
(2 rows)
so you'd want to exclude those. (I guess just skipping ones with oid <
9999 would be enough to skip built-ins.)
regards, tom lane
From | Date | Subject | |
---|---|---|---|
Next Message | Erik Serrano | 2022-11-05 17:46:25 | Postgresql - Modelo backup full + incremental |
Previous Message | Wells Oliver | 2022-11-04 20:51:59 | Re: Struggling with 13->14 and anyarray -> anycompatiblearray |