Postgres Pro
Downloads
Home   >   mailing lists

Thread: symbol ':'

symbol ':'

From
Michael Meskes
Date:
Where is this symbol used? I see now way to check for correct syntax when
":a" can mean something else that the C variable a.

Michael
--
Dr. Michael Meskes, Project-Manager    | topsystem Systemhaus GmbH
meskes@topsystem.de                    | Europark A2, Adenauerstr. 20
meskes@debian.org                      | 52146 Wuerselen
Go SF49ers! Go Rhein Fire!             | Tel: (+49) 2405/4670-44
Use Debian GNU/Linux!                  | Fax: (+49) 2405/4670-10

Re: [HACKERS] symbol ':'

From
"Thomas G. Lockhart"
Date:
> Where is this symbol used? I see now way to check for correct syntax
> when ":a" can mean something else that the C variable a.

It's an allowed operator symbol assigned to a little-used math
operation. You can disallow it for embedded sql if you need to. Another
possibility though is to disallow it as an operator _unless_ it is
inside parentheses. I did a similar thing in gram.y in the BETWEEN/AND
parsing to allow the boolean "AND" operator to be used on the right side
of BETWEEN/AND. But maybe indicator variables are allowed to show up
inside expressions too, so this wouldn't work...

btw, I've been thinking about changing another operator symbol, ";",
because of the obvious parsing ambiguities with the SQL end of statement
symbol. We could change both ":" and ";" operators for v6.4??

                       - Tom

Re: [HACKERS] symbol ':'

From
Michael Meskes
Date:
Thomas G. Lockhart writes:
> It's an allowed operator symbol assigned to a little-used math
> operation. You can disallow it for embedded sql if you need to. Another

Yes, I have to.

> possibility though is to disallow it as an operator _unless_ it is
> inside parentheses. I did a similar thing in gram.y in the BETWEEN/AND
> parsing to allow the boolean "AND" operator to be used on the right side
> of BETWEEN/AND. But maybe indicator variables are allowed to show up
> inside expressions too, so this wouldn't work...

They are. And it's not only indicators. The variables begin with ':', too.

> btw, I've been thinking about changing another operator symbol, ";",
> because of the obvious parsing ambiguities with the SQL end of statement
> symbol. We could change both ":" and ";" operators for v6.4??

Good idea.

Michael

--
Dr. Michael Meskes, Project-Manager    | topsystem Systemhaus GmbH
meskes@topsystem.de                    | Europark A2, Adenauerstr. 20
meskes@debian.org                      | 52146 Wuerselen
Go SF49ers! Go Rhein Fire!             | Tel: (+49) 2405/4670-44
Use Debian GNU/Linux!                  | Fax: (+49) 2405/4670-10