I have two definitions, each resulting in a 2Dvoronoi with the cells being returned as polyline curves. One definition uses the 2Dvoronoi done my dmitrie, the other uses the geometric edition by SAC3. In both cases the polylines are returned with identical datatrees, but for some reason only the one by dmitrie works, and i need to use the other one.
im using 19... so far no problems... are you sure it works... i know that it draws the cells as i should, but the last component the vb.component returns null in the sac3(manual) edition and returns the index of the points as it should for the dmitrie(scripting) edition...
The vornoi solution you're trying to fix is having problems computing the boundary position. You should be able to fix the input parameters where I highlighted in order to fix the 'manual' solution. The definition is also starting off computing a 3d solution but then only uses the 2d part of the bounding box parameters therefore not accounting for the height of the points (which I know might all be at 0 but I think this is also causing an error).
I opened your file, but i was still not working. It rather strange because i get the 13 polyline curves, so the definition works up until when the vb.net component reads the curves. It just returns null, and it should return a list of the corner vertices as a text string. And for the dmitrie edition, it works, it returns the corner points as strings.
The problem is the manual solution isn't a clean solution because it's using a boolean region difference (pink line) whereas the VB script gives you a clean solution which is why you can extract the correct vertices from them. If you zoom into your two solutions you'll see the problem.