如何基于paraview在python脚本中显示vtkUnstructuredGrid?

时间:2019-02-09 04:45:22

标签: python paraview pvpython

我在Ubuntu 18.04系统上安装了paraview 5.6,我想编写一个python脚本来显示vtkUnstructuredGrid。

import numpy as np
from paraview.simple import *
import paraview.vtk as vtk
from paraview.vtk.numpy_interface import dataset_adapter as dsa
import paraview.vtk.util.numpy_support as vnp

node = np.array(
        [[0.0, 0.0, 0.0],
         [1.0, 0.0, 0.0],
         [1.0, 1.0, 0.0],
         [0.0, 1.0, 0.0]], dtype=np.float)
cell = np.array([[3, 1, 2, 0], [3, 3, 0, 2]], dtype=np.int)
NC = cell.shape[0]

points = vtk.vtkPoints()
points.SetData(vnp.numpy_to_vtk(node))
cells = vtk.vtkCellArray()
cells.SetCells(NC, vnp.numpy_to_vtkIdTypeArray(cell))

uGrid =vtk.vtkUnstructuredGrid() 
uGrid.SetPoints(points)
uGrid.SetCells(vtk.VTK_TRIANGLE, cells)
# how to put uGrid into the following codes
view = GetActiveViewOrCreate('RenderView') 
dispaly = Show()
render = Render()
Interact()

我在互联网上找不到任何示例可以在python脚本中进行此类操作。因此,我需要您的帮助,非常感谢。

更新

我尝试编写如下的源类:

import numpy as np
from paraview.simple import *
import vtk 
import vtk.util.numpy_support as vnp
from vtkmodules.util.vtkAlgorithm import VTKPythonAlgorithmBase
from vtkmodules.numpy_interface import dataset_adapter as dsa
from paraview.util.vtkAlgorithm import smproxy, smproperty, smdomain

@smproxy.source(name="MeshSource", label="triangle mesh!")
class MeshSource(VTKPythonAlgorithmBase):
    def __init__(self):
        print("Initialize the source!")
        VTKPythonAlgorithmBase.__init__(self,
                nInputPorts=0,
                nOutputPorts=1,
                outputType='vtkUnstructuredGrid')
        node = np.array(
                [[0.0, 0.0, 0.0],
                 [1.0, 0.0, 0.0],
                 [1.0, 1.0, 0.0],
                 [0.0, 1.0, 0.0]], dtype=np.float)
        cell = np.array([[3, 1, 2, 0], [3, 3, 0, 2]], dtype=np.int)
        NN = node.shape[0]
        NC = cell.shape[0]

        points = vtk.vtkPoints()
        points.SetData(vnp.numpy_to_vtk(node))
        cells = vtk.vtkCellArray()
        cells.SetCells(NC, vnp.numpy_to_vtkIdTypeArray(cell))

        self.mesh = vtk.vtkUnstructuredGrid() 
        self.mesh.SetPoints(points)
        self.mesh.SetCells(vtk.VTK_TRIANGLE, cells)
        rho = vnp.numpy_to_vtk(np.zeros(NN))
        rho.SetName('rho_A')
        self.mesh.GetPointData().AddArray(rho)
        self.Port = 0

    def RequestData(self, request, inInfo, outInfo):
        print("Request the data!")
        output = vtk.vtkUnstructuredGrid.GetData(outInfo)
        optput.ShallowCopy(self.mesh)
        return 1

    def UpdatePointData(self, rho):
        print("Update the point data!")
        rho = vnp.numpy_to_vtk(rho)
        rho.SetName('rho_A')
        self.mesh.GetPointData().AddArray(rho)
        self.Modified()

source = MeshSource()
view = GetActiveViewOrCreate('RenderView') 
display = Show(source, view)
Interact()

但是我遇到了一些错误:

Traceback (most recent call last):
  File "test_triangle.py", line 55, in <module>
    dispaly = Show(source, view)
  File "/home/why/local/lib/python3.6/site-packages/paraview/simple.py", line 482, in Show
    rep = controller.Show(proxy, proxy.Port, view)
  File "/home/why/local/lib/python3.6/site-packages/paraview/servermanager.py", line 158, in __ConvertArgumentsAndCall
    retVal = func(*newArgs)
TypeError: Show argument 1: method requires a vtkSMSourceProxy, a vtkPythonAlgorithm was provided.

我必须错过一些东西。

1 个答案:

答案 0 :(得分:1)

要理解的关键是,ParaView中有两个级别的Python脚本可用。在较低的级别,您可以使用VTK创建或过滤数据。较高的级别可让您控制ParaView的操作,例如显示数据,设置显示属性等。所缺少的是脚本中两个级别之间的桥梁。

在您的原始示例中,您将在VTK中创建一个非结构化网格。要将其放置到ParaView可以使用的地方,请添加以下内容:

# how to put uGrid into the following codes
view = GetActiveViewOrCreate('RenderView')

# create a trivial producer to bridge between the VTK object and ParaView
tp = TrivialProducer()
tp.GetClientSideObject().SetOutput(uGrid)

dispaly = Show(tp)

这将创建一个ParaView代理(TrivialProducer),该代理是一个简单的VTK数据源vtkTrivialProducer的代理。它所做的只是获取数据集并将其传递给请求它的任何下游过滤器。 (注意:由于GetClientSideObject(),这仅在以内置服务器模式运行时才有效,这很常见。)

如何定义单元格也存在问题。单元格定义中的第一个条目必须是定义单元格的点数。因此,将该行更改为

cell = np.array([[3, 1, 2, 0], [3, 3, 0, 2]], dtype=np.int)