[vtkusers] Selection with vtkCompositePolyDataMapper2 mapper problem

Alex Malyushytskyy alexmalvtk at gmail.com
Tue Jul 14 22:55:23 EDT 2015


I was unable to test selection with VTK built with OpenGL2 option.
Application was built without errors, but displayed multiple problems.
It was failing inside vtk during rendering multiblock ( negative index in
the stl container )
dataset until I commented line after adding actor to renderer.

renderer_->AddActor(actor_ );

// renderer_->RemoveCuller(renderer_->GetCullers()->GetLastItem());
And even after this model was stripped off half components.
I guess I am stuck again.


On Tue, Jul 14, 2015 at 5:43 PM, Alex Malyushytskyy <alexmalvtk at gmail.com>
wrote:

> I attached data set written with vtkXMLMultiBlockDataWriter
> I failed to reproduce problem with ParaView myself.
> But its pipeline is different as well as mapper code seems different from
> vtk 6.1.0 and 6.2.0
>
> I am trying to re-build vtk 6.2.0 with Opengl2 module enabled. Will update
> when done.
>
> On Tue, Jul 14, 2015 at 2:32 AM, Utkarsh Ayachit <
> utkarsh.ayachit at kitware.com> wrote:
>
>> Alex,
>>
>> Can you share the dataset (as a vtm perhaps)? I see if I can reproduce
>> this with ParaView.
>>
>> Utkarsh
>>
>> On Tue, Jul 14, 2015 at 12:56 AM Alex Malyushytskyy <alexmalvtk at gmail.com>
>> wrote:
>>
>>> I tried to debug selection and found that in :
>>>
>>> vtkSelection* vtkHardwareSelector::GenerateSelection( unsigned int x1,
>>> unsigned int y1,
>>> unsigned int x2, unsigned int y2)
>>>
>>> GetPixelInformation(pos, 0); already
>>> returns PixelInformation.CompositeID for block which is outside selection
>>> frame as a valid
>>>
>>> This means that problem is either within CaptureBuffers() or
>>> GetPixelInformation()
>>> and possibly involves painter or mapper.
>>> I afraid I will have to give up on this unless I get any suggestion
>>> which may help me in finding bug location.
>>>
>>>
>>> On Mon, Jul 13, 2015 at 5:50 PM, Alex Malyushytskyy <
>>> alexmalvtk at gmail.com> wrote:
>>>
>>>> Attempt to upgrade to vtk-6-2-0 did not fix the problem
>>>>
>>>> On Fri, Jul 10, 2015 at 10:26 PM, Alex Malyushytskyy <
>>>> alexmalvtk at gmail.com> wrote:
>>>>
>>>>> Using vtkHardwareSelector with vtkMultiBlockDataSet and vtkCompositePolyDataMapper2
>>>>> sometimes is producing wrong results.
>>>>> I have a vtkMultiBlockDataSet with polydata (sometimes NULL) as
>>>>> blocks.
>>>>> On the screenshot you can see the simplest  geometry which I can
>>>>> reproduce the problem on. Each rectangle there is a separate block.
>>>>> It also displays window frame which supposed to define selection.
>>>>> As you can see results (in red like colors) There is an extra block
>>>>> selected which is totally outside of the frame.
>>>>>
>>>>> Relative  functions which do selection and extraction are available in
>>>>> the attachment.
>>>>> For some reason if I select a single component - basically by clicking
>>>>> without dragging selection frame, result is always correct.
>>>>> Result seems also depends on view angle.
>>>>> I tracked  that wrong results appear after vtkConvertSelection::ToSelectionType
>>>>> in converted to vtkSelectionNode::BLOCKS
>>>>> vtkSelection.
>>>>>
>>>>>
>>>>> I would appreciate any help or ideas.
>>>>> I am using vtk-6-1-0.
>>>>>
>>>>> Regards,
>>>>>       Alex
>>>>>
>>>>>
>>>>>
>>>>>
>>>>
>>> _______________________________________________
>>> 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 VTK FAQ at:
>>> http://www.vtk.org/Wiki/VTK_FAQ
>>>
>>> Search the list archives at: http://markmail.org/search/?q=vtkusers
>>>
>>> Follow this link to subscribe/unsubscribe:
>>> http://public.kitware.com/mailman/listinfo/vtkusers
>>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/vtkusers/attachments/20150714/1954e8af/attachment.html>


More information about the vtkusers mailing list