search

qgis / QGIS

QGIS is a free, open source, cross platform (lin/win/mac) geographical information system (GIS)
https://qgis.org
GNU General Public License v2.0
10.59k stars 3.01k forks source link

Validity check: Consider z-values in geometry evaluation #41963

Open tschmetzer opened 3 years ago

tschmetzer commented 3 years ago

Currently the validity check/invalid features filtering algorithm classifies certain geometries with errors although they don't seem to be from a 3D perspective.

grafik

Example data attached: https://github.com/qgis/QGIS/files/6061362/richtungsbestimmung.zip (EPSG:25832)

The attribute A_TYPE means surface type i.e. 1=ground, 2=wall, 3=roof The only surface type that the QGIS validity algorithm characterizes as invalid are the walls. Walls possess Z-values which from a 2D perspective might seem wrong. I cannot see any error there.

I personally highly suspect the validity check/invalid features filtering algorithm to be "incorrect" for certain types of 3D data like the one used in templayer2. I examined several features point by point that have been marked as "invalid output"/"error output" by the validity check and all were totally fine for me.

Can anybody confirm my suspicion? I suggest to check each polygon geometry more in depth if z-values are present and if it consists of more then 3 points. In the case of this example data one check could for example be that it's closed and that there's no intersections between the segments but maybe other people could have different data use cases. Just to be clear: I'm not asking for a validity check of my data. I just want to seed the idea that certain data should not be erroneously classified as invalid.

roya0045 commented 3 years ago

geos is used most of the time. Nothing to do qgis in those cases.