[Paraview] Memory leak with Python script (XDMF and HDF5)

[Nikolaos Beratlis]

I am using the following python script to extract images from a series of
files for an animation:

try: paraview.simple
except: from paraview.simple import *
paraview.simple._DisableFirstRenderCameraReset()

index = [None]*9999

m = 0
for i in range(0, 9):
  for j in range(0, 9):
    for k in range(0, 9):
      for l in range(0, 9):
        index[m] = chr(i+48)+chr(j+48)+chr(k+48)+chr(l+48)
        m = m+1


for i in range(1,2):

    cf_VP_anim_xmf = XDMFReader( FileName='XDMF/cf_VP'+index[i]+'.xmf' )

    cf_VP_anim_xmf.CellArrays = []
    cf_VP_anim_xmf.Sets = []
    cf_VP_anim_xmf.PointArrays = ['Cf']

    RenderView1 = GetRenderView()
    a1_Cf_PVLookupTable = GetLookupTableForArray( "Cf", 1 )

    Cfcontours = Show()
    Cfcontours.EdgeColor = [0.0, 0.0, 0.50000762951094835]
    Cfcontours.SelectionPointFieldDataArrayName = 'Cf'
    Cfcontours.ScalarOpacityFunction = []
    Cfcontours.ColorArrayName = 'Cf'
    Cfcontours.ScalarOpacityUnitDistance = 2.4071425938556859
    Cfcontours.LookupTable = a1_Cf_PVLookupTable
    Cfcontours.ScaleFactor = 15.0

    cf_VP_anim_xmf.Grids = ['STL Grid']

    RenderView1.CacheKey = 2.0
    RenderView1.CameraClippingRange = [288.47079837284241,
299.16435388730815]
    RenderView1.UseCache = 0

    a1_Cf_PVLookupTable.RGBPoints = [-0.0080000000000000002, 0.0, 0.0, 1.0,
0.0080000000000000002, 1.0, 0.0, 0.0]
    a1_Cf_PVLookupTable.LockScalarRange = 1

    separationline = Contour( PointMergeMethod="Uniform Binning" )

    separationline.ContourBy = ['POINTS', 'Cf']
    separationline.Isosurfaces = [0.0]
    DataSeparationLine = Show()
    DataSeparationLine.ScaleFactor = 12.321390008926393
    DataSeparationLine.SelectionPointFieldDataArrayName = 'Cf'
    DataSeparationLine.EdgeColor = [0.0, 0.0, 0.50000762951094835]
    DataSeparationLine.LineWidth = 2.0
    DataSeparationLine.DiffuseColor = [0.0, 0.0, 0.0]

    DataSeparationLine.ColorArrayName = ''

    Cfcontours.Visibility = 1

    Q_VP_xmf = XDMFReader( FileName='XDMF/Q_VP'+index[i]+'.xmf' )

    Q_VP_xmf.CellArrays = []
    Q_VP_xmf.Sets = []
    Q_VP_xmf.PointArrays = ['Q']

    DataQ = Show()
    DataQ.Representation = 'Outline'
    DataQ.ScaleFactor = 12.341455459594727
    DataQ.SelectionPointFieldDataArrayName = 'Q'
    DataQ.EdgeColor = [0.0, 0.0, 0.50000762951094835]

    Q_VP_xmf.Grids = ['Eulerian Grid']

    ContourQ = Contour( PointMergeMethod="Uniform Binning" )
    ContourQ.PointMergeMethod = "Uniform Binning"
    ContourQ.ContourBy = ['POINTS', 'Q']
    ContourQ.Isosurfaces = [1.0]
    ContourQ.ComputeNormals = 0

    DataQiso = Show()
    DataQiso.ScaleFactor = 7.4976861000061037
    DataQiso.SelectionPointFieldDataArrayName = 'Normals'
    DataQiso.EdgeColor = [0.0, 0.0, 0.50000762951094835]

    DataQ.Visibility = 0


    #--- Setup view
    RenderView1 = GetRenderView()
    RenderView1.Background = [1.0, 1.0, 1.0]
    RenderView1.CenterAxesVisibility = 0
    RenderView1.OrientationAxesVisibility = 0
    RenderView1.ViewSize = [2000, 350]

    RenderView1.CameraViewUp = [0.0,-1.0, -2.2204460492503131e-16]
    RenderView1.CameraPosition = [41.5, 0.0, 56.5]
    RenderView1.CameraClippingRange = [120.40716073726786,
126.85668499830999]
    RenderView1.InteractionMode = '3D'
    RenderView1.CameraFocalPoint = [0, 0, 56.5]
    RenderView1.CenterOfRotation = [0, 0, 56.5]

    #--- Export image
    Render()
    WriteImage('Q_cf_topview_'+index[i]+'.jpg')


    #--- Delete objects (order is important)
    Delete(DataSeparationLine)
    Delete(separationline)
    Delete(Cfcontours)
    Delete(cf_VP_anim_xmf)
    Delete(DataQiso)
    Delete(ContourQ)
    Delete(DataQ)
    Delete(Q_VP_xmf)


At the end of the for loop, I delete all the objects and readers. I run
this python script in batch mode (Paraview 3.98) and in parallel with 4
processors. However I am having a memory leak, by the 6th time in the for
loop it complains about insufficient memory. Can anyone understand what is
causing the memory to increase? Do I forget to delete sth?

Thank you,

Nikos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.paraview.org/pipermail/paraview/attachments/20130620/699c4284/attachment-0001.htm>