RSS Parrot

#6130: Geography distance is 0 for two lines where an end point of one is the antipode of an end point of the other

https://trac.osgeo.org/postgis/ticket/6130

Published: September 11, 2026 21:14

ST_Distance on geography answers 0, and ST_Intersects and ST_DWithin answer true, for two lines that are 6654 km apart, whenever an end point of one line is the antipode of an end point of the other. Moving that end point by 0.001 degree gives the expec

#6129: Missing area with topology.ST_GetFaceGeometry('arstat_2026_overlay_02_348',101383), but alle edges seems to be prensent

https://trac.osgeo.org/postgis/ticket/6129

Published: September 11, 2026 15:07

In the file when have a set of edges for one face from topology where SELECT * FROM topology.ValidateTopology('arstat_2026_overlay_02_348'); return no errors. The problem is that this cmd only return SELECT topology.ST_GetFace

#6117: Expose GEOS revision level version to SQL

https://trac.osgeo.org/postgis/ticket/6117

Published: September 11, 2026 14:14

Proposed change: ​https://gitea.osgeo.org/postgis/postgis/pulls/800

#6128: ST_intersects Gives false Positives related to meridian / antimeridian

https://trac.osgeo.org/postgis/ticket/6128

Published: September 9, 2026 18:07

I am seeing some squirrely behavior with ST_intersects when determining if a point intersects a polygon when a polygon touches (but does not span) the antimeridian and the lookup point lies on the meridian. This is easily reproducible in the latest docker

#6127: PostgreSQL 20 doesn't work against 3.7.0

https://trac.osgeo.org/postgis/ticket/6127

Published: September 7, 2026 17:41

Both debbie and gha are registering errors because we haven't flipped their following of PostgreSQL master and amster recently flipped to PostgreSQL 20dev. That said recent changes have deviated enough from 19 to make it break with this error.

#6126: Document requirement for ci status update

https://trac.osgeo.org/postgis/ticket/6126

Published: September 4, 2026 07:21

Oh never mind, I see that README documents make publish which in turn ends up calling the ci target, so we can say it IS documented.

#6125: Document hugo version requirements for website building

https://trac.osgeo.org/postgis/ticket/6125

Published: September 4, 2026 07:19

Hugo v0.123.7 is unable to build the website: $ hugo version hugo v0.123.7+extended linux/amd64 BuildDate=2025-07-18T03:41:49Z VendorInfo=ubuntu:0.123.7-1ubuntu0.3 $ hugo serve hugo serve Watching for changes in /usr/local/src/postgis/postgis.

#6034: Missing node when adding overlapping line

https://trac.osgeo.org/postgis/ticket/6034

Published: September 4, 2026 06:22

I'm closing this as a specific case of the more general #6122 issue

#6124: 3.7.0rc1 compile failure on NetBSD

https://trac.osgeo.org/postgis/ticket/6124

Published: September 2, 2026 20:57

​https://lists.osgeo.org/pipermail/postgis-users/2026-August/047074.html gmake[3]: Leaving directory '/tmp/work/databases/postgresql-postgis2/work/postgis-3.7.0rc1/extensions/postgis_topology' gmake[2]: Leaving directory '/tmp/work/databases/postg

#6123: ST_MakeSimple

https://trac.osgeo.org/postgis/ticket/6123

Published: September 2, 2026 17:33

​here is the topology code relying on lwgeom_make_valid to handle line collapses. Evidently at some point that function did handle line collapses (suggesting there's been a regression in that function, which would be worth its own ticket).

#6122: Incoming linework vertices do not form nodes when below tolerance from existing edges

https://trac.osgeo.org/postgis/ticket/6122

Published: September 2, 2026 17:03

As mentioned in ​https://trac.osgeo.org/postgis/ticket/6034#comment:16 the current topology code doesn't make any effort at finding cases in which incoming vertices would end up being below tolerance from existing edge segments. Such support exists for *no

#6121: Adding a node below distance of multiple edge segments snaps only one of them

https://trac.osgeo.org/postgis/ticket/6121

Published: September 2, 2026 16:49

Starting in PostGIS-3.5.0 we added experimental code to snap more than one existing edges to incoming nodes when those edges are below tolerance. Previously, only the closest edge was snapped, to avoid having to deal with edge merging. This code is

#6120: XX000: SQL/MM Spatial exception - curve not simple

https://trac.osgeo.org/postgis/ticket/6120

Published: September 2, 2026 06:52

When adding first line it works ok SELECT ARRAY(SELECT abs(add_case05) FROM topology.TopoGeo_addLinestring('test_53_a_1_8','0102000020A2100000050000004313671D53DE34404E64FC3B10545140169A989329DE3440E00C65580F54514010526CDB36DE34406BEE28A10F545140E