On Tue, Jan 21, 2025 at 6:36 PM Jim Jones <jim.jones@uni-muenster.de> wrote:
> On 21.01.25 23:45, Robert Treat wrote:
> > Is there some concrete use case you have seen that this would help
> > with? Not objecting to adding it, but you've mentioned this migration
> > idea twice but it seems to me this doesn't conform with existing
> > implementations, and I don't see much benefit in migration use cases
> > specifically, so I'm just curious if I am overlooking something?
>
> I wouldn’t frame it as a "migration idea". My point is that this would
> be one less function to modify when migrating a script from another
> database system to PostgreSQL.
>
<snip>
>
> Which compatibility issues with existing implementations are you
> referring to?
>
I'm mostly referring back to Pavel's statements:
| 1. Oracle doesn't support this
| 2. DB2 has different implementations for z/OS (variadic) and for
unix (nonvariadic)
With an added #3 which is AFAIK sql server doesn't implement this
either (I believe they implement something similar with
sp_xml_preparedocument)
So even if we are following the spec (which I think technically we may
not be), if no other database implements it the way we are, then it
really isn't one less function people will have to modify. Which
again, I'm not trying to argue against having it, I'm just trying to
figure out if there is some specific group that would benefit from it.
Robert Treat
https://xzilla.net
