[Qgis-user] 3D Vertical line as invalid geometries

[Nicolas Cadieux]

Hi,
I have been playing with the 3D functions in QGIS with the latest version of QGIS. Since drawing in 3D is normally something I do in a CAD, I have been asked to do this is QGIS and ArcMap for this particular project (let’s just say this reaffirmed my belief in the strength QGIS…).

Anyway, since using a 2D geometry with functions like extrusion is out of the question for this project, I have been typing my 3D geometries as WKT (just simple boxes), uploading the csv in QGIS then updating the geometries to 3D using the wkt field as the csv import does not seem to create 3D line.  (Something that should probably be updated).

When viewed in the QGIS 3D map or in ArcScene my 3D boxes are perfect. Problems occurs when trying to split the file using a field.  

It so happens that lines that are perfectly straight and vertical like LINESTRINGZ(0 0 0, 0 0 1) are flagged as invalid and I am told to fix the geometry before continuing.   If I remove these objects from the files, the split file by field works perfectly.  In the python Shapely library, this line would also receive a length of 0 even thought is 1 z unit high basically because the library is a 2D library even though it tolerates 3D geometries. This is fare enough.

So where is my question, are the problems I am having with invalid caused by the QGIS 2D object based model?  Or is this just a “split file using a field problem?” I imagine that building a true 3D GIS is a while away but should we suggest to the developers to be a bit more tolerant and not to flag this type of geometries as invalid?

Waiting for your thoughts…

Nicolas Cadieux
https://gitlab.com/njacadieux
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/qgis-user/attachments/20210610/32a730de/attachment.html>