# SPDX-FileCopyrightText: 2026 Oak Ridge National Laboratory and Contributors # # SPDX-License-Identifier: Apache-2.0 import argparse from paraview.simple import * from paraview import print_info # ------------------------------------------------------------------------------ # Catalyst options from paraview import catalyst options = catalyst.Options() options.GlobalTrigger = "TimeStep" options.EnableCatalystLive = 1 options.CatalystLiveTrigger = "TimeStep" options.ExtractsOutputDirectory = "/tmp" # options.ExtractsOutputDirectory = '.' DISABLE_EXTRACTOR = False # setup and returns the view def SetupRenderView(): # Create a new 'Render View' view = CreateView("RenderView") camera = GetActiveCamera() camera.Azimuth(45) camera.Elevation(45) SetActiveView(None) # create new layout object 'Layout #1' layout1 = CreateLayout(name="Layout #1") layout1.AssignView(0, view) layout1.SetSize(824, 656) # restore active view SetActiveView(view) return view # A different proxy type is used depending on whether you're using # catalyst, or doing post hoc visualization # this returns the catalyst producer def SetupCatalystProducer(): producer = TrivialProducer(registrationName="fides") return producer # this returns the fides proxy which should be used for post hoc vis # i.e., you're reading .bp files def SetupFidesReader(json, bp, sst): if json is None: # in this case the bp file must contain the Fides attributes fides = FidesReader(StreamSteps=1, FileName=bp) return fides fides = FidesJSONReader(StreamSteps=1, FileName=json) if sst: fides.DataSourceEngines = ["source", "SST"] # 'source' is the name of the ADIOS data source in the JSON data model fides.DataSourcePath = ["source", bp] # required to update the fides reader fides.UpdatePipelineInformation() return fides # takes in a producer and view and sets up the visualization pipeline def SetupVisPipeline(producer, view): Show(producer, view, "UniformGridRepresentation") view.ResetCamera() # get color transfer function/color map for 'U' uLUT = GetColorTransferFunction("V") uLUT.AutomaticRescaleRangeMode = "Clamp and update every timestep" uLUT.RescaleOnVisibilityChange = 1 contour1 = Contour(registrationName="Contour1", Input=producer) contour1.ContourBy = ["POINTS", "V"] contour1.Isosurfaces = [0.1, 0.3, 0.5, 0.7] contour1.PointMergeMethod = "Uniform Binning" contour1Display = Show(contour1, view, "GeometryRepresentation") contour1Display.Representation = "Surface" contour1Display.ColorArrayName = ["POINTS", "V"] Hide(producer, view) # trace defaults for the display properties. contour1Display.Representation = "Surface" contour1Display.ColorArrayName = ["POINTS", "V"] contour1Display.SetScaleArray = ["POINTS", "V"] contour1Display.ScaleTransferFunction = "PiecewiseFunction" contour1Display.LookupTable = uLUT uLUTColorBar = GetScalarBar(uLUT, view) uLUTColorBar.Title = "V" uLUTColorBar.ComponentTitle = "" # show color legend contour1Display.SetScalarBarVisibility(view, True) return contour1, contour1Display # sets up an extractor for writing out PNG images def SetupExtractor(view): # create extractor pNG1 = CreateExtractor("PNG", view, registrationName="PNG1") # trace defaults for the extractor. pNG1.Trigger = "TimeStep" # init the 'PNG' selected for 'Writer' pNG1.Writer.FileName = "output_{timestep:06d}.png" pNG1.Writer.ImageResolution = [800, 800] pNG1.Writer.Format = "PNG" # Catalyst uses this to update the pipeline and print out some info on each time step # you don't need to call this directly in your script; ParaView Catalyst will call it for you def catalyst_execute(info): print_info("in '%s::catalyst_execute'", __name__) global pipeline, display pipeline.UpdatePipeline() display.RescaleTransferFunctionToDataRange() print_info("executing (cycle={}, time={})".format(info.cycle, info.time)) print_info("U-range: {}".format(producer.PointData["U"].GetRange(0))) print_info("V-range: {}".format(producer.PointData["V"].GetRange(0))) def ParseArgs(): parser = argparse.ArgumentParser() parser.add_argument( "-j", "--json_filename", help="path to Fides JSON file", type=str, required=False, ) parser.add_argument("-b", "--bp_filename", help="path to bp file", type=str, required=True) parser.add_argument("--staging", help="use SST engine", action="store_true") args = parser.parse_args() return args def StreamingVis(args): # adios/fides step status NotReady = 1 EndOfStream = 2 # setup the reader, view, pipeline fides = SetupFidesReader(args.json_filename, args.bp_filename, args.staging) view = SetupRenderView() step = 0 while True: status = NotReady while status == NotReady: # must call PrepareNextStep to get Fides ready to read the # next step fides.PrepareNextStep() fides.UpdatePipelineInformation() status = fides.NextStepStatus if status == EndOfStream: # done reading the file return if step == 0: # set up the pipeline on the first step pipeline, display = SetupVisPipeline(fides, view) # need to update the pipeline and then save the output pipeline.UpdatePipeline() display.RescaleTransferFunctionToDataRange() output = f"output-{step:05d}.png" SaveScreenshot(output, view, ImageResolution=[800, 800]) step += 1 # ------------------------------------------------------------------------------ if __name__ == "__main__": print("in __main__()") args = ParseArgs() StreamingVis(args) else: # in this case we're running from Catalyst view = SetupRenderView() producer = SetupCatalystProducer() pipeline, display = SetupVisPipeline(producer, view) # normally not needed, but a bug fix is in progress to fix an issue # when using ParaView Live with extractors if not DISABLE_EXTRACTOR: SetupExtractor(view)