Re: OWNER TO on all objects

Поиск
Список
Период
Сортировка
От Tom Lane
Тема Re: OWNER TO on all objects
Дата
Msg-id 25176.1087271701@sss.pgh.pa.us
обсуждение исходный текст
Ответ на OWNER TO on all objects  (Christopher Kings-Lynne <chriskl@familyhealth.com.au>)
Ответы Re: OWNER TO on all objects  (Christopher Kings-Lynne <chriskl@familyhealth.com.au>)
Список pgsql-hackers
Christopher Kings-Lynne <chriskl@familyhealth.com.au> writes:
> Then I made it so that pg_dump will output an OWNER TO statement after 
> every object creation.

Perhaps better to put these out towards the end of the dump, not right
after the creation of the object?  Or is that what you're doing?

I would envision the safest procedure as creating all objects, loading
all data, etc, then all ALTER OWNERs, then all GRANT/REVOKEs.

> * Do we need the set session auth for COPY commands still?

Not if you still own the table while loading into it (see above point).

However, this all assumes a complete dump/restore.  Consider data-only
restores.  Consider partial restores using pg_restore's options for
that.  What happens then?  It'd likely be appropriate to issue set
session auth during scenarios involving pre-existing objects.

> * Is there any reason there is no RENAME TO command for operators?

Lack of round tuits, no doubt.
        regards, tom lane


В списке pgsql-hackers по дате отправления:

Предыдущее
От: Greg Stark
Дата:
Сообщение: Re: building rpms from source rpm's
Следующее
От: Tom Lane
Дата:
Сообщение: Re: Spinlock assembly cleanup