Tom,
I will try the current snapshot and I will report in the list.
Thanks to Tomasz Myrta too for the help.
Alexandre
> I said:
>> PG 7.4 will be better prepared to handle this sort of query, but I
>> don't think it will realize that the T1/T2 left join could be reduced
>> to a plain join given these conditions
>
> I take that back --- actually, the algorithm used in CVS tip *does*
> deduce that all these left joins can be plain joins.
>
> Don't suppose you'd like to experiment with a current snapshot to see
> how well it does for you?
>
> regards, tom lane