Re: Another thought about search_path semantics

Andres Freund <andres@2ndquadrant.com> writes:
> On 2014-04-04 17:24:00 -0400, Tom Lane wrote:
>> Maybe not many, but pg_dump itself certainly can try to do that.
>> (Most of the time, pg_dump won't dump things in pg_catalog, but there
>> are exceptions, eg --binary-upgrade dump of an extension containing
>> objects in pg_catalog.)

> If we're not backpatching, fixing that seems easy enough?

Not especially.  As I said, pg_dump believes that setting search_path is
an appropriate way to control where things get created, and that's wired
into its structure pretty deeply.  I would have exactly zero faith in a
hack that tried to change that just for objects in pg_catalog.  In any
case, it's not clear that the case can't arise in pre-existing dump files,
even if you discount --binary-upgrade cases.

> I don't like my own suggestion, which isn't a good sign, but I haven't
> heard anything I like more :(.

Don't see why you don't find what I suggested to be just a small variant
on it.
        regards, tom lane