Re: Extending PostgreSQL with a Domain-Specific Language (DSL) - Development
| От | David G. Johnston |
|---|---|
| Тема | Re: Extending PostgreSQL with a Domain-Specific Language (DSL) - Development |
| Дата | |
| Msg-id | CAKFQuwZTu=J3gqQkbb+9+yhUgc5m_y1WtJA6cHEh-PdY8i3bHA@mail.gmail.com обсуждение исходный текст |
| Ответ на | Re: Extending PostgreSQL with a Domain-Specific Language (DSL) -Development (Tom Mercha <mercha_t@hotmail.com>) |
| Ответы |
Re: Extending PostgreSQL with a Domain-Specific Language (DSL) -Development
|
| Список | pgsql-hackers |
On Tue, Jul 9, 2019 at 5:23 PM Tom Mercha <mercha_t@hotmail.com> wrote:
I understand that you never wrote any PL handler but was just thinking
about this functionality as a follow-up to our conversation. I was just
wondering whether anonymous DO blocks *must* return void or not?
The docs for DO say it is a function returning void -
https://www.postgresql.org/docs/current/sql-do.html
But the docs for CREATE LANGUAGE's INLINE HANDLER say 'typically return
void' - https://www.postgresql.org/docs/current/sql-createlanguage.html
No, the language cannot override the SQL execution environment's limitations.
"The code block is treated as though it were the body of a function with no parameters, returning void. It is parsed and executed a single time."
The above applies regardless of the language the code block is written in.
It can, however, affect permanent session state (so, use tables).
David J.
В списке pgsql-hackers по дате отправления: