Received: from www.propertykey.com (propertykey.com [206.196.37.193]) by postgresql.org (8.11.3/8.11.1) with ESMTP id f3HKkiu38869 for ; Tue, 17 Apr 2001 16:46:45 -0400 (EDT) (envelope-from jeff@propertykey.com) Received: from propertykey.com (gotojail.propertykey.com [206.196.37.197]) by www.propertykey.com (8.9.3/8.9.3) with ESMTP id PAA21810; Tue, 17 Apr 2001 15:46:11 -0500 Message-ID: <3ADCAB5C.38E58603@propertykey.com> Date: Tue, 17 Apr 2001 15:45:16 -0500 From: Jeff Hoffmann Organization: PropertyKey.com X-Mailer: Mozilla 4.75 [en] (WinNT; U) X-Accept-Language: en MIME-Version: 1.0 To: Oleg Bartunov CC: Tom Lane , pgsql-sql@postgresql.org Subject: Re: RTREE on points References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Archive-Number: 200104/283 X-Sequence-Number: 2638 Oleg Bartunov wrote: > > GiST is great ! > > You may look at http://www.sai.msu.su/~megera/postgres/gist/ > for GiST implementation of RTree - it could be not compiled with 7.1 > release due to some api changes, but it's not difficult to do. it looks like i just wasted a good couple of hours trying to convert the gist_box_ops. it did help find the pointer problem i was having because i'm still not up to speed on the new function calling conventions, though... > If somebody want it I could contribute it to contrib area. i'm definitely interested. i'm going to play with it & if oleg's claim holds about index insertion time holds, i can definitely see myself moving to it over the built in rtree. anything that can cut down the hours of index creation time would be great. also, it seems that it'd be a good choice for inclusion in the standard distribution because it'd be easy to test -- you already have to run rtree tests anyway, you can just duplicate them with gist & gist_box_ops. -- Jeff Hoffmann PropertyKey.com