Re: security_definer_search_path GUC

Поиск
Список
Период
Сортировка
От David G. Johnston
Тема Re: security_definer_search_path GUC
Дата
Msg-id CAKFQuwaNFQcC5RafB=6NOB2KYn9n5BrFcrkBadgVbzoJe-Gg7Q@mail.gmail.com
обсуждение исходный текст
Ответ на Re: security_definer_search_path GUC  ("Joel Jacobson" <joel@compiler.org>)
Ответы Re: security_definer_search_path GUC  ("Joel Jacobson" <joel@compiler.org>)
Список pgsql-hackers
On Mon, Jun 7, 2021 at 1:55 PM Joel Jacobson <joel@compiler.org> wrote:
If we don't like "UNQUALIFIED" as a keyword, maybe we could reuse "PUBLIC"?
Or will that be confusing since "PUBLIC" is also a role_specification?


For me the concept resembles explicitly denoting certain schemas as being simple tags, while the actual "namespace" is the GLOBAL namespace.  Today there is no global namespace, all schemas generate their own individual namespace in addition to "tagging" their objects with a textual label.

Avoiding "public" is highly desirable.

To access a global object you should be able to still specify its schema tag.  Unqualified means "use search_path"; and "use search_path" includes global.  But there is a truth table waiting to be created to detail what combinations result in errors (including where those errors occur - runtime or creation time).

David J.
 

В списке pgsql-hackers по дате отправления:

Предыдущее
От: "David G. Johnston"
Дата:
Сообщение: Re: security_definer_search_path GUC
Следующее
От: "David G. Johnston"
Дата:
Сообщение: Re: A modest proposal vis hierarchical queries: MINUS in the column list