Re: BUG #18780: Bindings types are lost for complex queries

On Sunday, January 19, 2025, PG Bug reporting form <noreply(at)postgresql(dot)org>
wrote:

> The following bug has been logged on the website:
>
> Bug reference: 18780
> Logged by: Viktr
> Email address: vik(at)notexi(dot)st
> PostgreSQL version: 17.2
> Operating system: Linux d2c635331de7 6.10.14-linuxkit #1 SMP PREEMPT
> Description:
>
> Hi there,
>
> Actual discussion is here: https://github.com/pgjdbc/pgjdbc/issues/3482
> In brief: I do understand that "select 1 as one, 2 as two, 3 as three"
> might
> have lack of type info. But I suppose, in case these values are used in the
> assignment, the type could be taken from the corresponding column type.
> Like
> for "insert into sometable(one, two, three) select 1 as one, 2 as two, 3 as
> three" it is obvious that types should match.
>
>
A bug report should be self-contained. Only pointing to a 40 message long
thread isn’t helping get the bug (well, feature) fixed.

The crux of that thread is your driver is sending along a text data typed
value because it cannot decide whether timestamp or timestamptz is needed.
Implicitly casting text to something else isn’t going to happen.

This falls into a feature request, one that comes up from time-to-time, and
doesn’t ever seem to meet anyone’s benefit/cost threshold for working on;
or at least get pushed over the edge.

The underlying feature, I think, is you want the parse to be able to say
“let the server decide the type” and the server reply with type info for
unspecified parameters. Or maybe accept the pseudo-type “unknown” at the
API level. In any case a thread to discuss a patch for such a change and
how JDBC would leverage it to solve this problem would be the next step.
Searching the mailing lists for existing discussions may yield fruit too
though I don;t know for certain.

David J.