Opened 2 years ago
Closed 2 years ago
#5234 closed defect (fixed)
REGRESSION: ring geometrically not-closed on 3D topology
Reported by: | amsilva | Owned by: | strk |
---|---|---|---|
Priority: | high | Milestone: | PostGIS 3.0.8 |
Component: | topology | Version: | 3.0.x |
Keywords: | Cc: |
Description
When adding an edge to the topology with totopogeom sometimes it gives this error "Corrupted topology: ring of edge x is geometrically not-closed".
I reduced the problem to an example with two lines. The first line is added with no errors to the topology but the second one throws this error. The second line intersects with the first one, but there's no vertex on the intersection. (example attached)
This bug results of a postgis-user post where Sandro Santilli confirmed the bug and complemented with this:
Gory details: the problem likely lays in _lwt_MakeRingShell which was recently changed to NOT use GEOS but rather do things internally, to reduce overhead. The internal implementation is NOT dropping the Z as the geos implementation did.
Attachments (1)
Change History (11)
by , 2 years ago
Attachment: | sample_script.sql added |
---|
comment:1 by , 2 years ago
Milestone: | PostGIS 3.3.1 → PostGIS 3.2.4 |
---|---|
Priority: | medium → high |
Summary: | ring geometrically not-closed on 3D topology → REGRESSION: ring geometrically not-closed on 3D topology |
comment:2 by , 2 years ago
Milestone: | PostGIS 3.2.4 → PostGIS 3.3.2 |
---|
comment:3 by , 2 years ago
Milestone: | PostGIS 3.3.2 → PostGIS 3.1.8 |
---|---|
Version: | 3.2.x → 3.1.x |
comment:4 by , 2 years ago
Milestone: | PostGIS 3.1.8 → PostGIS 3.0.8 |
---|---|
Version: | 3.1.x → 3.0.x |
3.0.8dev (current stable-3.0 branch) is also affected, while I confirm 3.0.1 works.
comment:5 by , 2 years ago
The regression was introduced in version 3.0.2 via commit [273b7eaad3e9a6dfae8eef9e29f78dd67582a4ca/git] which references ticket #4709
Version 3.1.7 (and 3.1 branch at time of writing) is also affected. Reduced testcase: