> Hannu Krosing wrote:
>
> John Ridout wrote:
> >
> > Hi,
> >
> > I know of course that the pgsql optimizer is never wrong
>
> :)
>
> OTOH, it is never completely right either :)
>
> > but for the sake of argument do you think we could have
> > optimizer hints the way say MSSQL has them?
>
> I think it has been discussed (a little) but more on the
> lines of 'the way Oracle has them' ;)
>
Oh, you mean properly.
> It seems that this is not currently very high priority for
> any of the current real developers, but if you would
> contribute a nice clean implementation I'm sure it would
> be accepted.
>
> ----------------
> Hannu
>
Come the end of September I will have enough time
to play with the internals of pgsql.
I suppose there are more useful things that
can be done before optimizer hints such as nearly
everything on the TODO list.
John.
