Hi!
I have some 3D faces that have been saved out to a FFS.
Now I'm trying to use a Search envelope on a FFS reader to load the model above, I want to clip the model according to the bound. It works but many(all) of the clipped faces have faulty values at Z=0. I would expect to get a nice clean cut on the 3D-faces, with correct z-height at all their indices. See this image:
I have attached the soure files for debugging.
Is this a known bug? Any workaround available?
I think only 2D-clipping is supported via the featurereader. I think you must read "intersect" and then use this method to do the actual clipping afterwards: https://hub.safe.com/publishers/safe-sandbox/transformers/surfaceclipper
I think only 2D-clipping is supported via the featurereader. I think you must read "intersect" and then use this method to do the actual clipping afterwards: https://hub.safe.com/publishers/safe-sandbox/transformers/surfaceclipper
Thanks! I'm not using the FeatureReader but instead a FFS reader. It actually does clip the 3D faces which is great, if only all the indices z-values would be correct:) I assume it's a bug,
Did you file this with our support team? I don't know if it's a known bug, but I'm fairly certain that it's a bug.
What I'm noticing is that the zero z-values appear to be extra vertices. If they aren't truly necessary, then I think you might be able to simply remove them and the data would be OK - like this:
The CoordinateExtractor extracts a list of coordinates and the ListSearcher searches that list for zero values in the Z coordinate. The VertexRemover then removes that vertex (Coordinate Index = List Index). The LineCloser returns the feature back to a polygon.
I have the same sequence twice because some features had two zero-value coordinates, and it was easier to deal with them this way.
Anyway, I will raise this as an issue with the development team, but in the meantime I'm hopeful that the above will be a valid workaround for you.
Did you file this with our support team? I don't know if it's a known bug, but I'm fairly certain that it's a bug.
What I'm noticing is that the zero z-values appear to be extra vertices. If they aren't truly necessary, then I think you might be able to simply remove them and the data would be OK - like this:
The CoordinateExtractor extracts a list of coordinates and the ListSearcher searches that list for zero values in the Z coordinate. The VertexRemover then removes that vertex (Coordinate Index = List Index). The LineCloser returns the feature back to a polygon.
I have the same sequence twice because some features had two zero-value coordinates, and it was easier to deal with them this way.
Anyway, I will raise this as an issue with the development team, but in the meantime I'm hopeful that the above will be a valid workaround for you.
I confirmed the problem and filed it with the developers under the reference number: FMEENGINE-62169
I confirmed the problem and filed it with the developers under the reference number: FMEENGINE-62169
Thanks! Looking forward to the fix.
I'm clipping using gdal now instead but it would be great to have it working in FME.
Hi all,
Just to let you know that this is reported fixed in FME 2020, build 20136 or greater. I'm expecting all to be well now, but if you give it a try and there are still problems, then please do let me know.
Regards
Mark