Re: not fully correct error message - Mailing list pgsql-hackers

From jian he
Subject Re: not fully correct error message
Date
Msg-id CACJufxEOQ+2=L3aLBzW6A4vu75gN7vKw07+QgU_398Avqb5KTg@mail.gmail.com
Whole thread Raw
In response to not fully correct error message  (Pavel Stehule <pavel.stehule@gmail.com>)
List pgsql-hackers
On Thu, Jan 1, 2026 at 3:10 PM Pavel Stehule <pavel.stehule@gmail.com> wrote:
>
> Hi
>
> I tested one use case, and maybe I found little bit possible error message
>
> create procedure test()
> as $$
> begin
>   vacuum;
> end;
> $$ language plpgsql;
>
> (2026-01-01 08:04:05) postgres=# call test();
> ERROR:  25001: VACUUM cannot be executed from a function
> CONTEXT:  SQL statement "vacuum"
> PL/pgSQL function test() line 3 at SQL statement
> LOCATION:  PreventInTransactionBlock, xact.c:3695
> (2026-01-01 08:09:18) postgres=#
>
> should be "VACUUM cannot be executed from a function or a procedure" instead ?
>

hi.
"VACUUM cannot be executed from a function or a procedure"
looks good to me.

similarly, in ExecWaitStmt we have:
        ereport(ERROR,
                errcode(ERRCODE_OBJECT_NOT_IN_PREREQUISITE_STATE),
                errmsg("WAIT FOR must be only called without an active
or registered snapshot"),
                errdetail("WAIT FOR cannot be executed from a function
or a procedure or within a transaction with an isolation level higher
than READ COMMITTED."));

PreventInTransactionBlock is used in so many places, but this error message:
``
(errmsg("%s cannot be executed from a function", stmtType)));
``
only appears once in the regress tests.
maybe we can add some dummy tests for it.



pgsql-hackers by date:

Previous
From: Pavel Stehule
Date:
Subject: Re: Planner : anti-join on left joins
Next
From: shveta malik
Date:
Subject: Re: Proposal: Conflict log history table for Logical Replication