Re: Suggested "easy" TODO: pg_dump --from-list
От | Robert Haas |
---|---|
Тема | Re: Suggested "easy" TODO: pg_dump --from-list |
Дата | |
Msg-id | AANLkTin-8QGW-p5euFNFM80s1GyCTYWHJCi2Oayj+3jZ@mail.gmail.com обсуждение исходный текст |
Ответ на | Re: Suggested "easy" TODO: pg_dump --from-list (Andrew Dunstan <andrew@dunslane.net>) |
Список | pgsql-hackers |
On Wed, Nov 24, 2010 at 8:41 AM, Andrew Dunstan <andrew@dunslane.net> wrote: > On 11/24/2010 07:29 AM, Robert Haas wrote: >> As a first attempt at syntax, I might suggest something along the >> lines of "object type: object name", where the types and names might >> look to COMMENT ON for inspiration. > > pg_dump already uses a list of object types (e.g. as seen in the output from > pg_restore --list). Let's not invent something new if we don't need to. But > we'll need more than that. We'll need enough for disambiguation, especially > for functions which is your stated use case. So, something like this might > work: > > FUNCTION:myschema.myfunc:integer,text,timestamp with time zone Actually, that's pretty much exactly what I was proposing, except that I would've kept the existing convention for how to write a function's arguments: FUNCTION:myschema.myfunc(integer,text,timestamp with time zone) -- Robert Haas EnterpriseDB: http://www.enterprisedb.com The Enterprise PostgreSQL Company
В списке pgsql-hackers по дате отправления: