Re: plpgsql + named parameters

On Tue, May 19, 2009 at 5:59 PM, Steve Prentice <prentice(at)cisco(dot)com> wrote:
> I followed the past discussions regarding the syntax for named parameters
> and I am currently using Pavel Stehule's patch for named and mixed notation
> on top of the 8.4 beta.
>
> It seems the way plpgsql substitutes $1, $2, etc for the parameters is going
> to reduce the usefulness of this feature. Consider these two functions:
>
> CREATE FUNCTION fun1(a INT DEFAULT 1) RETURNS INT AS 'SELECT $1' LANGUAGE
> SQL;
> CREATE FUNCTION fun2(a INT) RETURNS INT AS $$
> DECLARE
>        t INT;
> BEGIN
>        t := fun1(1 as a);      -- syntax error: "SELECT  fun1(1 as  $1 )"
>        t := fun1(a as a);      -- syntax error: "SELECT  fun1( $1  as  $1 )"

you have a name conflict here...is it deliberate? I've learned the
hard way to always, always prefix arguments and locals to plpgsql
functions with '_'. Or are you trying to do something fancier?

merlin