Re: BUG #16698: Create extension and search path
| От | David G. Johnston |
|---|---|
| Тема | Re: BUG #16698: Create extension and search path |
| Дата | |
| Msg-id | CAKFQuwYN2nsivt_LUhQ1zjugTZdQ2QiaX7DCZnRmxjSoKTvkNg@mail.gmail.com обсуждение исходный текст |
| Ответ на | Re: BUG #16698: Create extension and search path (Tom Lane <tgl@sss.pgh.pa.us>) |
| Список | pgsql-bugs |
On Tue, Nov 3, 2020 at 10:50 AM Tom Lane <tgl@sss.pgh.pa.us> wrote:
Miha Vrhovnik <miha.vrhovnik@gmail.com> writes:
> What are then extensions.. They live in some sort of limbo.. as they don't
> behave like a natural extension of the database... As they behave
> differently (because they "require" public shema in path if extensions are
> created while in public schema ).
You've got the wrong mental model. Extensions are containers, which
may *contain* objects that live within schemas, but the extension itself
does not. The key reason for doing it that way is that an extension might
itself contain a schema, or more than one schema.
Put a bit differently, extensions are global objects like roles. With roles you grant permissions on individual databases for each role to access it. For extensions you "create extension" within individual databases to enable the extension in each one - potentially making its components visible in a user-defined schema, but also the extension can use schemas in its own non-adjustable ways.
David J.
В списке pgsql-bugs по дате отправления: