Bjorn,
you'll get problem with your approach not to create additional columns
if index will not used ! In our approach sequential scan will use
txtidx column. I don't understand what error you're comment.
Probably I miss something. We are working on new version of tsearch
and alpha version will be available next week. Are you sure
'iscacheability' should be configureable somehow ? Teodor, what do
you think ?
Oleg
On Thu, 23 Jan 2003, Bjorn Metzdorf wrote:
> > Do index functions always have to be marked "iscachable" ?
> >
> > Btw, the error message was confusing. It said that the function has to be
> > marked "isImmutable", but there is no such attribute, instead "iscachable"
> > had to be used. This seems to be a bug.
>
> The manual says:
>
> "All functions and operators used in an index definition must be immutable,
> that is, their results must depend only on their input arguments and never
> on any outside influence (such as the contents of another table or the
> current time). This restriction ensures that the behavior of the index is
> well-defined. To use a user-defined function in an index, remember to mark
> the function immutable when you create it."
>
> Well, in the tsearch case the results don't depend on any influence outside
> of the function. The stemmer is integrated in the function and we don't use
> the current time nor another table. So it should be safe to mark that
> function "iscachable" or when it is fixed "isimmutable", am I right?
>
> Regards,
> Bjoern
>
Regards,
Oleg
_____________________________________________________________
Oleg Bartunov, sci.researcher, hostmaster of AstroNet,
Sternberg Astronomical Institute, Moscow University (Russia)
Internet: oleg@sai.msu.su, http://www.sai.msu.su/~megera/
phone: +007(095)939-16-83, +007(095)939-23-83