[vtkusers] XY graph in VTK

[Siddartha Krishnan]

Paul,

Thanks for your prompt reply. THe code did work in VTK5 (with some  
minor modifications) and I am now able to plot multiple data sets on  
the same set of axes!

Thanks,
Siddartha


On 9-Jun-06, at 5:49 AM, Paul Cochrane wrote:

> Siddartha,
>
>> I'm new to VTK and am trying to plot a simple 2D graph. I am using
>> vtkXYPlotActor() and am trying to plot fielddata. I have managed to
>> plot a single line, but am unsure of how to add another line to the
>> plot (i.e. plot another set of data on the same set of axes). I would
>> appreciate it If someone could help me with this.
>
> There was a great post by Sander Niemeijer ages ago about how to plot
> simple 2D graphs in vtk, and I've used that advice in a project of
> mine (PyVisi; pyvisi.sf.net).  Here is an example from the PyVisi
> distribution which plots multiple curves on the one set of axes, the
> code is Python, but it shouldn't be too hard to convert to C++ or
> other.  BTW: this code works with vtk 4.0, but probably not with vtk
> 5.0+ since the pipline model has changed.
>
> I hope this helps!
>
> Regards,
>
> Paul
>
> code follows:
>
> """
> Example of plotting multiple curves with pyvisi
> """
>
> # set up some data to plot
> from Numeric import *
>
> x = arange(0, 2*pi, 0.1, typecode=Float)
> y1 = sin(x)
> y2 = cos(x)
> y3 = cos(x)**2
>
> #### original vtk code
>
> import vtk
>
> # set up the renderer and the render window
> _ren = vtk.vtkRenderer()
> _renWin = vtk.vtkRenderWindow()
> _renWin.AddRenderer(_ren)
> _renWin.SetSize(640, 480)
>
> # do a quick check to make sure x and y are same length
> if len(x) != len(y1):
>    raise ValueError, "x and y vectors must be same length"
>
> if len(x) != len(y2):
>    raise ValueError, "x and y vectors must be same length"
>
> # set up the x and y data arrays to be able to accept the data (code
> # here adapted from the C++ of a vtk-users mailing list reply by  
> Sander
> # Niemeijer)
> _xData = vtk.vtkDataArray.CreateDataArray(vtk.VTK_FLOAT)
> _xData.SetNumberOfTuples(len(x))
>
> _yData1 = vtk.vtkDataArray.CreateDataArray(vtk.VTK_FLOAT)
> _yData1.SetNumberOfTuples(len(y1))
>
> _yData2 = vtk.vtkDataArray.CreateDataArray(vtk.VTK_FLOAT)
> _yData2.SetNumberOfTuples(len(y2))
>
> _yData3 = vtk.vtkDataArray.CreateDataArray(vtk.VTK_FLOAT)
> _yData3.SetNumberOfTuples(len(y3))
>
> # put the data into the data arrays
> for i in range(len(x)):
>    _xData.SetTuple1(i,x[i])
>    _yData1.SetTuple1(i,y1[i])
>    _yData2.SetTuple1(i,y2[i])
>    _yData3.SetTuple1(i,y3[i])
>
> # create a field data object
> # (I think this is as a containter to hold the data arrays)
> _fieldData1 = vtk.vtkFieldData()
> _fieldData1.AllocateArrays(2)
> _fieldData1.AddArray(_xData)
> _fieldData1.AddArray(_yData1)
>
> _fieldData2 = vtk.vtkFieldData()
> _fieldData2.AllocateArrays(2)
> _fieldData2.AddArray(_xData)
> _fieldData2.AddArray(_yData2)
>
> _fieldData3 = vtk.vtkFieldData()
> _fieldData3.AllocateArrays(2)
> _fieldData3.AddArray(_xData)
> _fieldData3.AddArray(_yData3)
>
> # now put the field data object into a data object so that can add  
> it as
> # input to the xyPlotActor
> _dataObject1 = vtk.vtkDataObject()
> _dataObject1.SetFieldData(_fieldData1)
>
> _dataObject2 = vtk.vtkDataObject()
> _dataObject2.SetFieldData(_fieldData2)
>
> _dataObject3 = vtk.vtkDataObject()
> _dataObject3.SetFieldData(_fieldData3)
>
> # set up the actor
> _plot = vtk.vtkXYPlotActor()
> _plot.AddDataObjectInput(_dataObject1)
> _plot.AddDataObjectInput(_dataObject2)
> _plot.AddDataObjectInput(_dataObject3)
>
> # the size of the actor should be 80% of the render window
> _plot.SetPosition(0.1, 0.1)  # (0.1 = (1.0 - 0.8)/2)
> _plot.SetWidth(0.8)
> _plot.SetHeight(0.8)
>
> # set the title and stuff
> _plot.SetTitle("Example 2D plot")
> _plot.SetXTitle("x")
> _plot.SetYTitle("y")
> _plot.SetXValuesToValue()
> # set which parts of the data object are to be used for which axis
> _plot.SetDataObjectXComponent(0, 0)
> _plot.SetDataObjectYComponent(0, 1)
> _plot.SetDataObjectYComponent(1, 1)
> _plot.SetDataObjectYComponent(2, 1)
>
> # set up the lookup table for the appropriate range of colours
> _lut = vtk.vtkLookupTable()
> _lut.Build()
> _colours = []
> for i in range(3):
>    _colours.append(_lut.GetColor(float(i)/float(3-1)))
>
> # change the colour of the separate lines
> _plot.SetPlotColor(0, _colours[0][0], _colours[0][1], _colours[0][2])
> _plot.SetPlotColor(1, _colours[1][0], _colours[1][1], _colours[1][2])
> _plot.SetPlotColor(2, _colours[2][0], _colours[2][1], _colours[2][2])
>
> _plot.GetXAxisActor2D().GetProperty().SetColor(0,0,0)
> _plot.GetYAxisActor2D().GetProperty().SetColor(0,0,0)
> _ren.SetBackground(1,1,1)
>
> # add the actor
> _ren.AddActor2D(_plot)
>
> # render the scene
> _iren = vtk.vtkRenderWindowInteractor()
> _iren.SetRenderWindow(_renWin)
> _iren.Initialize()
> _renWin.Render()
> _iren.Start()
>
> # convert the render window to an image
> _win2imgFilter = vtk.vtkWindowToImageFilter()
> _win2imgFilter.SetInput(_renWin)
>
> # save the image to file
> _outWriter = vtk.vtkPNGWriter()
> _outWriter.SetInput(_win2imgFilter.GetOutput())
> _outWriter.SetFileName("multiCurveLinePlot.png")
> _outWriter.Write()
>
> # pause for input
> raw_input('Press enter to continue...\n')