Re: Cross-column statistics revisited

On Thu, Oct 16, 2008 at 01:34:59PM -0400, Robert Haas wrote:
> I suspect that a lot of the correlations people care about are
> extreme.  For example, it's fairly common for me to have a table where
> column B is only used at all for certain values of column A.  Like,
> atm_machine_id is usually or always NULL unless transaction_type is
> ATM, or something.  So a clause of the form transaction_type = 'ATM'
> and atm_machine_id < 10000 looks more selective than it really is
> (because the first half is redundant).

That case is easily done by simply considering the indexed values of a
column with a partial index. This should be fairly easy to do I think.

It might be worthwhile someone trawling through the archives looking
for examples where we estimate the correlation wrong.

Have a nice day,
--
Martijn van Oosterhout   <kleptog@svana.org>   http://svana.org/kleptog/
> Please line up in a tree and maintain the heap invariant while
> boarding. Thank you for flying nlogn airlines.