[Paraview] XY plotting via filter extending vtkRectilinearGridAlgorithm, resulting plot always empty
Christine Corbett Moran
corbett at physik.uzh.ch
Mon Oct 5 16:44:53 EDT 2009
Ah, I didn't know that the automatic plotting of vtkRectilinear data
output from a filter was done with the "Plot over line" filter. So do
you know of a way to fix the problem? I had also tried a sample filter
with both x and y coordinate values, one with higher dimensions of the
dummy coordinate arrays, and neither of these things worked.
There are a number of other messages of people trying to do this, and
some claiming to have succeeded with similar approaches
http://www.paraview.org/pipermail/paraview/2008-March/007491.html
http://www.paraview.org/pipermail/paraview/2008-April/007870.html
etc.
Christine
On Mon, Oct 5, 2009 at 10:28 PM, Utkarsh Ayachit
<utkarsh.ayachit at kitware.com> wrote:
> The "Plot Over Line" filter isn't working for your dataset since you
> are producing a data with 0 width and 0 height. So when it tries to
> probe with the line, it's failing since the points on the line never
> falls within any cells.
>
> Utkarsh
>
> On Mon, Oct 5, 2009 at 4:07 PM, Christine Corbett Moran
> <corbett at physik.uzh.ch> wrote:
>> By "plotting setup" I mean that I first want to make a filter which
>> plots X vs. Y for a very simple case, so that when I try to make the
>> filter do something more complicated I don't have to worry about basic
>> bugs/conceptual issues (e.g. if I have a wrong understanding about
>> structured extents or which algorithm to derive from to automatically
>> create a plot in the XY view of my desired variables after execution
>> of a filter).
>>
>> Thanks for the vtkTable idea, it does seem I may have the wrong idea
>> about which type of data to produce, I had read a number of posts on
>> the mailing list that said that the way to get ParaView to
>> automatically plot the results from a filter in the XY view was to
>> produce vtkRectilinearGrid. I will try vtkTable and see if it is
>> suitable. In the meantime if anyone can help with the
>> vtkRectilinearGrid simple XY plot example, it may come in handy if
>> vtkTable is not suitable, for others, or for the next time I work with
>> vtkRectilinearGrid.
>>
>> Cheers,
>> Christine
>>
>>
>> On Mon, Oct 5, 2009 at 9:53 PM, Utkarsh Ayachit
>> <utkarsh.ayachit at kitware.com> wrote:
>>> I am not sure what you mean by "plotting setup for the filter". Is
>>> this filter supposed to produce data that can plotted directly? In
>>> that case you might want to simply create a vtkTableAlgorithm
>>> subclassand produce vtkTable (instead of vtkRectilinearGrid) that has
>>> data rows/columns which can easily be plotted by ParaView. vtkTable is
>>> also easier to deal with since you don't have to worry about
>>> structured extents.
>>>
>>> Utkarsh
>>>
>>> On Mon, Oct 5, 2009 at 3:38 PM, Christine Corbett Moran
>>> <corbett at physik.uzh.ch> wrote:
>>>> Hi Utkarsh,
>>>>
>>>> Yes that works but I actually want to do more complicated things than
>>>> just plot the data values, so first am doing something simple to make
>>>> sure I have the plotting setup for the filter correct.
>>>>
>>>> Cheers,
>>>> Christine
>>>>
>>>> On Mon, Oct 5, 2009 at 9:34 PM, Utkarsh Ayachit
>>>> <utkarsh.ayachit at kitware.com> wrote:
>>>>> Since you simply want to plot the data_values from your rectilinear
>>>>> grid, try applying the "Plot Data" filter instead. Does that work?
>>>>>
>>>>> Utkarsh
>>>>>
>>>>> On Mon, Oct 5, 2009 at 3:30 PM, Christine Corbett Moran
>>>>> <corbett at physik.uzh.ch> wrote:
>>>>>> Hi,
>>>>>>
>>>>>> I want to begin adding some non-trivial 2d plotting capabilities to my
>>>>>> ParaView plugin; in that interest I've implemented a filter which is
>>>>>> derived from vtkRectilinearGridAlgorithm. For now, I am trying only a
>>>>>> simple example, where my X coordinates are the particle ids, my Y and
>>>>>> Z coordinates are dummy arrays with one element, and my data is the id
>>>>>> to the power 2. When I act my filter, the XY plot view and a new 3D
>>>>>> view appear. However the XY plot is empty, as is the 3D view. I can
>>>>>> get something to appear in the 3D view by adding a glyph to the
>>>>>> pipeline (and the result makes sense), but nothing I do seems to get
>>>>>> it the XY plot view to actually plot.
>>>>>>
>>>>>> I have included the RequestData portion of my filter at the end of the
>>>>>> message as well as the information tab of the object inspector of the
>>>>>> filter. Everything looks normal to me (except I am not sure why the
>>>>>> number of cells is less than the number of points?). Can anyone see
>>>>>> the problem? Is there something more I need to do? I've already done a
>>>>>> lot of searching of the archives, playing around with the parameters,
>>>>>> and reading the vtk documentation but haven't found the bug or the
>>>>>> missing call.
>>>>>>
>>>>>> The information tab of the object inspector displays:
>>>>>> Type: Rectilinear Grid
>>>>>> Number of Cells: 1000
>>>>>> Number of Points: 1001
>>>>>> Data Arrays:
>>>>>> Name | Data Type | Data Ranges
>>>>>> data values | int | [0,1e+06]
>>>>>> Extents
>>>>>> X Extent: 0 to 1000 (dimension: 1001)
>>>>>> Y Extent: 0 to 0 (dimension: 1)
>>>>>> Z Extent: 0 to 0 (dimension: 1)
>>>>>> Bounds
>>>>>> X range: 0 to 1e+03
>>>>>> Y range: 0 to 0
>>>>>> Z range 0 to 0
>>>>>>
>>>>>> And my request data method is:
>>>>>> //----------------------------------------------------------------------------
>>>>>> int vtkMassFunctionFilter::RequestData(vtkInformation*,
>>>>>> vtkInformationVector** inputVector,
>>>>>> vtkInformationVector* outputVector)
>>>>>> {
>>>>>> // Get input and output data.
>>>>>> vtkPointSet* input = vtkPointSet::GetData(inputVector[0]);
>>>>>> vtkRectilinearGrid* output = vtkRectilinearGrid::GetData(outputVector);
>>>>>> // Setting the dimensions of this to be equal to our number of points,
>>>>>> // in X, and equal to 1 in the Y and Z directions,
>>>>>> // as these are dummy arrays.
>>>>>> output->SetDimensions(input->GetPoints()->GetNumberOfPoints(),1,1);
>>>>>> output->SetWholeExtent(0,input->GetPoints()->GetNumberOfPoints(),\
>>>>>> 0,0,\
>>>>>> 0,0);
>>>>>> // Allocate the arrays for the X,Y, and Z coordinates, and inserts a
>>>>>> // single value for the dummy arrays, as well as allocate the array
>>>>>> // for the scalar data
>>>>>> vtkSmartPointer<vtkDoubleArray> XArray=vtkSmartPointer<vtkDoubleArray>::New();
>>>>>> XArray->SetNumberOfComponents(1);
>>>>>> XArray->SetNumberOfTuples(input->GetPoints()->GetNumberOfPoints());
>>>>>> vtkSmartPointer<vtkDoubleArray>
>>>>>> DummyYArray=vtkSmartPointer<vtkDoubleArray>::New();
>>>>>> DummyYArray->SetNumberOfComponents(1);
>>>>>> DummyYArray->SetNumberOfTuples(1);
>>>>>> DummyYArray->InsertValue(0,0);
>>>>>> vtkSmartPointer<vtkDoubleArray>
>>>>>> DummyZArray=vtkSmartPointer<vtkDoubleArray>::New();
>>>>>> DummyZArray->SetNumberOfComponents(1);
>>>>>> DummyZArray->SetNumberOfTuples(1);
>>>>>> DummyZArray->InsertValue(0,0);
>>>>>> vtkSmartPointer<vtkIntArray> dataValues=vtkSmartPointer<vtkIntArray>::New();
>>>>>> dataValues->SetNumberOfComponents(1);
>>>>>> dataValues->SetNumberOfTuples(input->GetPoints()->GetNumberOfPoints());
>>>>>> dataValues->SetName("data values");
>>>>>> for(int nextPointId = 0;\
>>>>>> nextPointId < input->GetPoints()->GetNumberOfPoints();\
>>>>>> ++nextPointId)
>>>>>> {
>>>>>> XArray->InsertValue(nextPointId,nextPointId);
>>>>>> dataValues->InsertValue(nextPointId,pow(nextPointId,2));
>>>>>> }
>>>>>> // Updating the output
>>>>>> output->SetXCoordinates(XArray);
>>>>>> output->SetYCoordinates(DummyYArray);
>>>>>> output->SetZCoordinates(DummyZArray);
>>>>>> output->GetPointData()->SetScalars(dataValues);
>>>>>> return 1;
>>>>>> }
>>>>>> _______________________________________________
>>>>>> Powered by www.kitware.com
>>>>>>
>>>>>> Visit other Kitware open-source projects at http://www.kitware.com/opensource/opensource.html
>>>>>>
>>>>>> Please keep messages on-topic and check the ParaView Wiki at: http://paraview.org/Wiki/ParaView
>>>>>>
>>>>>> Follow this link to subscribe/unsubscribe:
>>>>>> http://www.paraview.org/mailman/listinfo/paraview
>>>>>>
>>>>>
>>>>
>>>
>>
>
More information about the ParaView
mailing list