[postgis-users] Bad plan without && operator

[Paul Ramsey]

I don't think so, the comment says 10000*ndims is the max number of
cells, and the code underneath does seem to enforce that, so max cells
would be 20,000 for a 2d index.

P

On Thu, Dec 7, 2017 at 8:02 AM, Sandro Santilli <strk at kbt.io> wrote:
> On Thu, Dec 07, 2017 at 06:48:05AM -0800, Paul Ramsey wrote:
>> Are you sure that's what SET STATISTICS means? I thought it was the
>> size of the sample drawn to build the statistics. The histogram size
>> is somewhat hardcoded,
>> https://github.com/postgis/postgis/blob/svn-trunk/postgis/gserialized_estimate.c#L1446-L1453
>
> As you can read in that comment, each side of the grid
> is composed by stats->attr->attstattarget cells, so 100x100
> by default and can be raised by SET STATISTICS.
>
> You both make the grid bigger *and* the sample bigger.
>
> --strk;
> _______________________________________________
> postgis-users mailing list
> postgis-users at lists.osgeo.org
> https://lists.osgeo.org/mailman/listinfo/postgis-users