Re: [postgis-users] ST_Distance provides unexpected results when using Geography type

2020-03-05 Thread Wayne Rowcliffe
Wow. That is not at all what I expected. Thanks for the handy tip. *Wayne Rowcliffe* Software Engineer, Priority5 Holdings E-mail: wa...@priority5.com On Thu, Mar 5, 2020 at 5:06 PM Paul Ramsey wrote: > If you would like to see what your “boxes” actually look like, try > wrapping them in ST

Re: [postgis-users] JOIN performance for ST_Point()

2020-03-05 Thread Ryo NEYAMA
Hello, Thank you for the information. I'll try it with PostGIS 3.0. Ryo 2020年3月5日(木) 19:01 Darafei "Komяpa" Praliaskouski : > Hello, > > Equality has been significantly reworked in 3.0. Can you reproduce > your issue in PostGIS 3.0? > > On Thu, Mar 5, 2020 at 9:46 AM Ryo NEYAMA > wrote: > > >

Re: [postgis-users] ST_Distance provides unexpected results when using Geography type

2020-03-05 Thread Paul Ramsey
If you would like to see what your “boxes” actually look like, try wrapping them in ST_Segmentize() before viewing them on a flat map. select st_asgeojson(st_segmentize('SRID=4326;POLYGON ((-167.51953125 8.841651120809145, 1.23046875 8.841651120809145, 1.23046875 63.35212928507874, -167.5195312

[postgis-users] ST_Distance provides unexpected results when using Geography type

2020-03-05 Thread Wayne Rowcliffe
Hi, I've attached a file containing sample queries that reproduce what I am seeing. Basically, we have an application that shows data on the globe. We are filtering the results to those within the current window viewport. What I am seeing is that depending on where the map is, I'll either get res

Re: [postgis-users] Changed intersection test speed after PG 11 -> 12 and PostGIS 2.5 -> 3 upgrade

2020-03-05 Thread Tom Kazimiers
Sorry for sending the information in multiple pieces. What I meant to write as well is that I ran REINDEX on the relevant 12/3 table (treenode_edge) and reran the queries. It didn't make a measurable difference. The same for VACUUM, no difference. I ran VACUUM FULL too on the table and if anyth

Re: [postgis-users] Changed intersection test speed after PG 11 -> 12 and PostGIS 2.5 -> 3 upgrade

2020-03-05 Thread Tom Kazimiers
A small correction to my last mail: I only noticed after I have sent it that the 11.7/2.5 query didn't load everything from memory and had to hit the disk (Buffers: shared hit=302 read=1349 dirtied=2). This is the first query, with everything in cache: 11.7/2.5.2 (query 1 with &&&): Index Sc

Re: [postgis-users] JOIN performance for ST_Point()

2020-03-05 Thread Komяpa
Hello, Equality has been significantly reworked in 3.0. Can you reproduce your issue in PostGIS 3.0? On Thu, Mar 5, 2020 at 9:46 AM Ryo NEYAMA wrote: > > Hello, > > I had a performance issue when I JOIN two tables by using the equivalence > of two ST_Point()s. > I'm wondering if it is a kind of

[postgis-users] PostgreSQL/PostGIS Support tender

2020-03-05 Thread Marco Lechner - FOSSGIS e.V.
The German Federal Office for Radiation Protection published a tender for PostgreSQL/PostGIS Support. Offers welcome: https://www.evergabe-online.de/tenderdetails.html?1&id=316092 Dr. Marco Lechner FOSSGIS e.V. www.fossgis.de ___ postgis-users mailing l