Hi,
Am 21.11.2003 um 06:54 schrieb Tom Lane:
> Martin Hampl <Martin(dot)Hampl(at)gmx(dot)de> writes:
>> Index Scan using word_lower_idx on token (cost=0.00..98814.08
>> rows=25382 width=16)
>> Index Cond: (lower((word)::text) = 'saxophone'::text)
>
> The rows estimate (and therefore also the cost estimate) is a complete
> guess in this situation, because the system keeps no statistics about
> the values of lower(word). Improving this situation is on the TODO
> list.
Any ideas when this will work? Is it difficult to implement?
(For those who don't recall the context: I asked about indexing lower
values of a varchar-coloumn ("create index xy_idx on
table(lower(coloumn));") and how the query planner uses this index).
Regards,
Martin.