Re: Named Operators
| От | Tom Lane |
|---|---|
| Тема | Re: Named Operators |
| Дата | |
| Msg-id | 3497966.1675875539@sss.pgh.pa.us обсуждение |
| Ответ на | Re: Named Operators (Tom Lane <tgl@sss.pgh.pa.us>) |
| Список | pgsql-hackers |
I wrote:
> This approach does have a couple of shortcomings:
> * You still have to invent an operator name, even if you never
> plan to use it in queries. This is just cosmetic though.
> It's not going to matter if the operator name is long or looks like
> line noise, if you only need to use it a few times in setup DDL.
Oh, one other thought is that we could address that complaint
by allowing OPERATOR(identifier), so that your DDL could use
a meaningful name for the operator. I see that we don't
actually support OPERATOR() right now in CREATE OPERATOR or
ALTER OPERATOR:
regression=# create operator operator(+) (function = foo);
ERROR: syntax error at or near "("
LINE 1: create operator operator(+) (function = foo);
^
but I doubt that'd be hard to fix.
regards, tom lane
В списке pgsql-hackers по дате отправления: