[Paraview] How does one rescale CSVReader data via pvpython?

[Greg Schussman]

Hi, Jean.

Thank you for helping.   On your suggestion, I tried removing the Fetch, 
but then I was getting empty objects.  From googling around a bit, I 
realized that I needed to call wake_out.UpdatePipeline() before moving on.

Because this is pretty ad-hoc, I'd prefer to avoid creating a filter, 
and just keep the macro self-contained.  I'm having difficulty finding 
how to connect inputs and outputs.  I've spent a lot of time with 
google, and with the dir() command, but still don't see how to get at 
the data.

Thanks for your contiuned help.   Below is what I currently have.

Greg

##########################
# this part seems good

filename = '/raid/schussma/work2/ace3p/paraview/macros/wakeplot/wake.out'
wake_out = CSVReader(FileName=filename,
                      FieldDelimiterCharacters = ' ',
                      HaveHeaders = 0)
wake_out.UpdatePipeline()  # this is necessary!  blank objects otherwise...

##########################
# here is the mystery

di = wake_out.GetDataInformation()
ri = di.GetRowDataInformation()

scale = 1.0/ri.GetArrayInformation(2).GetComponentRange(0)[1]

input = ???  # what goes here?

import vtk
newcol0 = vtk.vtkDoubleArray()
newcol0.DeepCopy(input.GetColumn(1))

for i in range(newcol0.GetNumberOfTuples()):
   newcol0.SetValue(i, scale * newcol0.GetValue(i))  # rescale

   output = ???.GetOutputDataObject(0)
   output.AddColumn(newcol0)

##########################
# this was good without the scaling

pd = PlotData(???)   # is no longer wake_out, correct?
dr = Show(pd,
           XArrayName = 'Field 0',
           AttributeType = 'Row Data',
           SeriesVisibility = ['vtkOriginalIndices', '0'])

Render()



On 09/15/2011 02:05 AM, Favre Jean wrote:
> Greg,
>
> I would not use Fetch to do what you need. Here is something that should work:
>
> # reuse your wake_out proxy
>
> di = wake_out.GetDataInformation()
> rowInfo = di.GetRowDataInformation()
>
> # get largest magnitude value from column 2
>
> rowInfo.GetArrayInformation(1).GetComponentRange(0)[1]
>
> #You then use a programmable filter with the following copy op.
> #one deepcopy to get your initial array, and then you scale it the way you want
> # you may discard the old array later
>
>
> import vtk
>
> input = self.GetInput()
>
> newcol0 = vtk.vtkDoubleArray()
>
> newcol0.DeepCopy(input.GetColumn(0))
>
> for i in range(newcol0.GetNumberOfTuples()):
>
> newcol0.SetValue(i, 2.0 * newcol0.GetValue(i)) # rescale by a factor of 2
>
>
> output = self.GetOutputDataObject(0)
>
> output.AddColumn(newcol0)
>
> print output.GetNumberOfColumns()
>
> -----------------
> Jean M. Favre
> Swiss National Supercomputing Center
>
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.paraview.org/pipermail/paraview/attachments/20110915/9a08265d/attachment.htm>