Help wanted: unknown ZeroDivisionError about nurbs-python

[orbingol]

I don't have a valid ACIS license anymore, so I cannot test it. ZeroDivisionError may happen, because the algorithms are using the user input directly. There is a frange implementation, but I doubt that's the problem. There were several other problems like that and converting values like -59.99999999999998579 to -60 worked fine at that point. You could try that.

I was planning to implement a more precise floating point control to eliminate such issues, but never had time to do it without adding any additional dependencies.

[chenboshuo]

I don't have a valid ACIS license anymore, so I cannot test it. ZeroDivisionError may happen, because the algorithms are using the user input directly. There is a frange implementation, but I doubt that's the problem. There were several other problems like that and converting values like -59.99999999999998579 to -60 worked fine at that point. You could try that.

I was planning to implement a more precise floating point control to eliminate such issues, but never had time to do it without adding any additional dependencies.

I changed this snippets to

from geomdl import BSpline

# Create a BSpline surface instance (Bezier surface)
surf = BSpline.Surface()

surf.degree_u=2
surf.degree_v=2
surf.ctrlpts_size_u=5
surf.ctrlpts_size_v=9
surf.ctrlpts=[
                (52.5, 60, 0),
                (52.5, 60, 60),
                (52.5, 0, 60),
                (52.5, -60, 60.0),
                (52.5, -60, 0),
                (52.5, -60.0, -60),
                (52.5, 0, -60),
                (52.5, 60, -60.0),
                (52.5, 60, 0),
                (17.5, 60, 0),
                (17.5, 60, 60),
                (17.5, 0, 60),
                (17.5, -60, 60),
                (17.5, -60, 0),
                (17.5, -60, -60),
                (17.5, 0, -60),
                (17.5, 60, -60),
                (17.5, 60, 0),
                (17.5, 0, 0),
                (17.5, 0, 0),
                (17.5, 0, 0),
                (17.5, 0, 0),
                (17.5, 0, 0),
                (17.5, 0, 0),
                (17.5, 0, 0),
                (17.5, 0, 0),
                (17.5, 0, 0),
                (0, 0, 0),
                (0, 0, 0),
                (0, 0, 0),
                (0, 0, 0),
                (0, 0, 0),
                (0, 0, 0),
                (0, 0, 0),
                (0, 0, 0),
                (0, 0, 0),
                (0, 0, 0),
                (0, 0, 0),
                (0, 0, 0),
                (0, 0, 0),
                (0, 0, 0),
                (0, 0, 0),
                (0, 0, 0),
                (0, 0, 0),
                (0, 0, 0),
]
surf.weights=[
1, 0.707107, 1, 0.707107, 1, 0.707107, 1, 0.707107, 1, 0.707107, 0.5, 0.707107, 0.5, 0.707107, 0.5, 0.707107, 0.5, 0.707107, 1, 0.707107, 1, 0.707107, 1, 0.707107, 1, 0.707107, 1, 0, 52.5, 60, 0, 52.5, 60, 60, 52.5, 0, 60, 52.5, -60, 60, 52.5, -60, 0, 52.5, -60, ]
surf.knotvector_u=[0, 0, 0, 1, 1, 1, 1, 1, ]
surf.knotvector_v=[0, 0, 0, 0.25, 0.25, 0.5, 0.5, 0.75, 0.75, 1, 1, 1, ]
# Evaluate surface points
surf.evaluate()

# Plot the control points grid and the evaluated surface
surf.vis = VisMPL.VisSurface(vis_config)

I am still encountering the same error.

[orbingol]

Could you also update the weights like you did for the control points? I'd try converting 0.707107 to 0.70711 and remove one significant if it doesn't work.