Oracle FAQ | Your Portal to the Oracle Knowledge Grid |
![]() |
![]() |
Home -> Community -> Usenet -> c.d.o.misc -> Re: Calling function from select statement is slower then using subselect in select
DA Morgan wrote:
> Martin T. wrote:
> > DA Morgan wrote:
> >> Martin T. wrote:
> >>
> >>> When will they give me templates for my packaged SQL?!? :)
> >>>
> >>> cheers,
> >>> Martin
> >> Please explain ... off-line if you prefer.
> >>
> >
> > Hu? Off-line?
> >
> > It's just a fancy idea of mine ...
> > So we have VIEWS ... which are stored queries to make things more
> > organized.
> > Why not have more flexible "views" where part of the "view" remains
> > variable until used ... so that the user of the view could push
> > predicates or whatever into the view ... the SQL engine wouldn't even
> > be aware of it, just the PL/SQL compiler.
> > Or at least give me packaged views ... cursors are alright but they're
> > so awful to use in PL/SQL code compared to inline queries ... (And I so
> > hate having to define everything on the schema level.)
> >
> > Well. Just my 002. I'm sure you disagree ;-)
> >
> > cheers,
> > Martin
>
> I don't disagree. But how is this different from putting a pipelined
> table function with NDS into a package?
> --
Yes, this would be a possible solution (or work-around).
However, NDS is evaluated at runtime, so we do not have the advantage
of compile-time check.
Maybe it would also be a bit less performant and it frankly sounds more
complicated :-)
Anyway .. thanks for pointing this out, I have actually never used pipelined functions yet so I'll think about it next time ...
cheers,
Martin
Received on Thu Nov 02 2006 - 03:36:58 CST
![]() |
![]() |