> > > AFAIR, none of the index access methods except btree
> > > handle NULLs at all --- they just ignore NULL values
> > > and don't store them in the index.
...
>
> and what does this error means ?
>
> create table rtree_test ( r box );
> copy rtree_test from stdin;
> \N
> ........ total 10,000 NULLS
> \.
>
> create index rtree_test_idx on rtree_test using rtree ( r );
> --ERROR: floating point exception! The last floating point
> operation either exceeded legal ranges or was a divide by zero
>
> seems rtree doesn't ignore NULL ?
No, it doesn't. As well as GiST. Only hash ignores them.
And there is no code in GiST & rtree that take care about NULL
keys. It's probably ok for GiST which is "meta-index" -
index/type methods implementator should decide how to handle NULLs.
As for rtree - seems it's better to ignore NULLs as we did before
for single key btree: rtree is just variation of it.
Vadim