On Tue, May 20, 2014 at 11:18:29AM -0700, Jeff Davis wrote:
> I think this can be done without breaking upgrade compatibility, because
> I think the structure already satisfies the invariants I mentioned in
> the other email (aside from the special case of a root tuple with two
> nodes and no prefi
On Tue, 2014-05-20 at 09:52 -0700, Jeff Davis wrote:
> 2. Why would any tuple have 2 nodes? If there are some non-empty ranges,
> why not make a centroid and have 4 or 5 nodes?
This is slightly more complicated than I thought, because we need to do
something about the root node if a bunch of empty
I am trying to understand the rangetypes spgist code and its interaction
with empty ranges. (Slightly embarrassing, because I reviewed the code.)
I see an old email here:
http://www.postgresql.org/message-id/50145a9c.7080...@enterprisedb.com
But still don't have a clear picture.
What I don't und
