Posted by jhofmeyr on 11/28/07 11:40

On Nov 28, 5:46 am, Ed Murphy <emurph...@socal.rr.com> wrote:
> creedp...@gmail.com wrote:
> > Consider this alternative and let me know if I screw this one up,
> > after all, it's getting late, and I'm still at work :) .....
>
> > SELECT company.code, company.name, company.type,
> > contacts.fullname
> > FROM company Left JOIN
> > contacts ON company.code = contacts.code
> > AND (company.type in ('C', 'R')) and (contacts.fullname =
> > 'Accounting')
> > WHERE contacts.code IS NULL
> > order by company.code
>
> I think this would work (except in the unlikely case that
> contacts.code is nullable, in which case it might return
> some data that it shouldn't). But NOT EXISTS has the strong
> advantage of letting you say what you mean.

The last time I tried this (on a fairly complex query as well) on SQL
2005, the execution plan of using a NOT EXISTS and LEFT OUTER JOIN to
filter rows was identical. Could it be that the query optimiser
actually understands what we are trying to achieve and derives the
best method to do so regardless of syntax these days? More likely it
was simply a quirk of the query / tables / indexing I guess...

J

[Back to original message]