I`m sorry, but why work around?

The overhead of "99%" of cases is so bigger? What about an "option" like constraint exclusion...

I've worked around also, but would like to see this topic in a TODO or further implementation.

Thanks by patience.



Kevin Grittner escreveu:
On Sat, Oct 6, 2007 at 11:28 AM, in message <24828.1191688097@sss.pgh.pa.us>,
        
Tom Lane <tgl@sss.pgh.pa.us> wrote: 
  
I've been wondering
if it would be worth the trouble to introduce scalarlesel and
scalargesel estimators ...
    
 
FWIW, this is an issue we ran into:
 
http://archives.postgresql.org/pgsql-performance/2006-03/msg00211.php
 
We worked around it in our framework by checking for a range with equal
values and converting to an equality test before presenting it to the
database.
 
-Kevin