All,
I need a couple homebrew versions of Pt-to-Cell Data because:
1. I don't want all pts variables done - just 1 at a time
2. I need special handling for angles that go from -180 to +180 (when 1
end of element has angle of -178 and other end has angle of 180, a straight
average will give an element angle of 1, when the correct answer is -179)
I have attached the kludge versions which I am using. They are terribly slow.
The Pt-to-Cell data filter is much faster processing the entire set of
variables than these are to process a single one. Is there any VTK routines I
can tap to make these faster?
If I'm doing something stupid which is making them slow, please feel free to
correct me.
Thanks for looking
Dennis
#
#
def AveragePointToCellData(block,point_variable):
#
# Simple averaging of value at all nodes in element
# like a simple PointDataToCellData filter
numCells = block.GetNumberOfCells()
elem_variable=zeros( (numCells,1) )
#
for element in range (numCells):
# Initialize element variables
elem_variable[element]=0.0
cell=block.GetCell(element)
numElemPts=cell.GetNumberOfPoints()
# loop over pts and accumulate - divide by numPts for average
for pt in range(numElemPts):
pointID=cell.GetPointId(pt)
#print pointID
elem_variable[element]=elem_variable[element]+point_variable[pointID]
elem_variable[element]=elem_variable[element]/numElemPts
#print elem_variable[element]
return elem_variable
#
def calc_element_angles(block):
#
# Must be done separately because of unique angle averaging problem at
180degrees
# If average angles of 180 and -178, will get angle of 1 and we desire -179
#
# check if angles already exist
if check_elem_variable_exists(block,'Sector_Angle_undef'):
el_angles=block.CellData['Sector_Angle_undef']
return el_angles
#
# Didn't find angles so:
# Get Node angles and average at each cell
# Undeformed Sector Angles (degrees about Y-axis, 0=FP)
node_angles=calc_node_angles(block)
#
#All nodes now have angles - average and assign to elements
# declare new variable arrays
numCells = block.GetNumberOfCells()
el_angles=zeros( (numCells,1) )
#
for element in range (numCells):
cell=block.GetCell(element)
numElemPts=cell.GetNumberOfPoints()
# flags to detect border sectors with both positive and negative angles
sector_angle_pos=0
sector_angle_neg=0
# Initialize element variables
el_angles[element]=0.0
# loop over pts and accumulate - divide by numPts for average
for pt in range(numElemPts):
pointID=cell.GetPointId(pt)
#print pointID
el_angles[element]=el_angles[element]+node_angles[pointID]
if node_angles[pointID]>=0:
sector_angle_pos=1
else:
sector_angle_neg=1
#print el_angles[element]
# detect "boundary" sectors with + and - angles and treat specially
# sector between 180 and -178 deg would get +1deg from averaging
# boundary sectors are always negative
if (sector_angle_pos and sector_angle_neg):
el_angles[element]=0.0
for pt in range(numElemPts):
pointID=cell.GetPointId(pt)
el_angles[element]=el_angles[element]-abs(node_angles[pointID])
el_angles[element]=el_angles[element]/numElemPts
#print el_angles[element]
return el_angles
_______________________________________________
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
Search the list archives at: http://markmail.org/search/?q=ParaView
Follow this link to subscribe/unsubscribe:
http://public.kitware.com/mailman/listinfo/paraview
