So sad...
Tks.

At least I know now I must avoid as many as i can...

--- In firebird-support@yahoogroups.com, fabianoaspro@... wrote:
>
> Subselects never get optimized in the way you posted.
> Bad but truth.
> Em 16/04/2013 11:09, "skander_sp" <skander_sp@...> escreveu:
>
> > **
> >
> >
> >
> > May be obvious, but til now, i don't see how much unoptimized works the
> > "field in subselect" syntax.
> >
> > I'm using it in some procedure, just right til now, after check they work
> > in the worse unoptimized mode...
> >
> > example:
> >
> > select * from Table1 T1
> > where T1.id in (select T2.id
> > from Table2 T2
> > where T2.Customer=:customer)
> >
> > of course T2.id is a Primary key and T1.id is a Foreing key
> >
> > I think they resolve getting the value(s) in the subselect, and then
> > applying to the main where, but the performace analisys show he run all the
> > T1 table from begin to end and then check with the subselect
> >
> > of course I can solve (in procedure) with a FOR subselect to get every
> > value and check after.... not the nicest way, and pretty ugly and less
> > "human readable", but by far much more quick and optimized.
> >
> > for select T2.id
> > from Table2 T2
> > where T2.Customer=:customer
> > into :id do
> > select * from Table1 T1
> > where T1.id=:id
> >
> > There is some way to get a good result using "field in subselect"? some
> > trick to get an optimized result using this syntax?
> >
> >
> >
>
>
> [Non-text portions of this message have been removed]
>