Re: Problem (bug?) with like

El lunes 03 de diciembre, Tom Lane escribió:
> > Here are my queries:
>
> You sure you didn't paste in the same result for #2 and #3? They're
> the same plan with the same rows estimates --- but I'd expect the rows
> estimates, at least, to change given the more-selective LIKE pattern.

I don't know for sure if I have send wrong data, but I think not.
Tomorrow I'll get more info.

> Also, how many rows are there really that match '%DA%' and '%DAVID%'?

Very few for that incredible difference in time:

'%DA%' -> 2 sec.
'%DAVID%' -> 3 min.

> I suspect the planner is being overoptimistic about the selectivity of
> '%DAVID%', and is choosing a plan that doesn't work well when there are
> lots of DAVIDs :-(

I have thought that it only occurs when item to find is present
completely in any number of registers, but this is only a burde
hypothesis :?

Thanks for all.

David