CODEWITHSUNDEEP
c++python-3.5vtkparaview
Tony
Tony

Reputation: 43

VTK Python Wrapper - Exporting Surface Plot

I'm using VTK 8.0.1 and python 3.5 and am brand new to VTK. I am trying to export a surface plot using vtkPlotSurface.

By referencing TestSurfacePlot.cxx I have successfully created a surface plot and have been able to render it in python (even though it doesn't really look like a surface plot).

import vtk
import math as m
import numpy as np

## Set things up
chart = vtk.vtkChartXYZ()
view = vtk.vtkContextView()
view.GetRenderWindow().SetSize(800,800)
view.GetScene().AddItem(chart)

## Create a surface
table = vtk.vtkTable()
numPoints = 70;
inc = 9.424778 / (numPoints - 1);
for i in range(0,numPoints):
    arr = vtk.vtkFloatArray()
    table.AddColumn(arr)

table.SetNumberOfRows(numPoints)
for i in range(0,numPoints):
    x = i * inc;
    for j in range(0,numPoints):
        y  = j * inc;
        table.SetValue(i, j, m.sin(m.sqrt(x*x + y*y)))

# Using table, create a surface plot
test = vtk.vtkPlotSurface()
test.SetXRange(0,9.424778)
test.SetYRange(0,9.424778)
test.SetInputData(table)

# Start visualizing the surface plot
chart.AddPlot(test)
view.GetRenderWindow().SetMultiSamples(0)
view.GetInteractor().Initialize()
view.GetRenderWindow().Render()

out = vtk.vtkOBJExporter()
out.SetFilePrefix("test")
out.SetInput(chart)
out.Write()


view.GetInteractor().Start()

In order to better visualize what I've made, I wanted to try and export it and then visualize using Paraview/Visit. However, I'm struggling to find any concrete examples where this type of vtk object is exported...

I have tried adding the following:

out = vtk.vtkOBJExporter()
out.SetFilePrefix("test")
out.SetInput(chart)
out.Write()

But end up with the following type error:

TypeError: SetInput argument 1: method requires a vtkRenderWindow, a vtkContextView was provided.

Can anyone provide assistance? Thanks in advance.

Upvotes: 1

Views: 1403

Answers (1)

Bane Sullivan
Bane Sullivan

Reputation: 241

You might benefit from using PyVista as it makes creating these types of spatially reference datasets and rendering much more user-friendly. I would avoid using a vtkTable like you have above and move towards VTK data objects that actually represent meshes/surfaces.

import pyvista as pv
import numpy as np

# Create a spatial reference
numPoints = 70
inc = 9.424778 / (numPoints - 1)
x = np.arange(0, numPoints) * inc
y = np.arange(0, numPoints) * inc
xx, yy, _ = np.meshgrid(x, y, [0])
zz = np.sin(np.sqrt(xx*xx + yy*yy))
# Make a PyVista/VTK mesh
surface = pv.StructuredGrid(xx, yy, zz)

# Plot it!
surface.plot(show_edges=True, show_grid=True, notebook=False)

# or save it out for opening in ParaView
surface.save("my_surface.vtk")

enter image description here

Upvotes: 1

Related Questions