"""Automated testing for simulation To be compatible with testcase_generator, a simulus class must have: setup(self): to initialize the test-case getRandomStimulus(self): to get a random stimulus name stimuli that inherit from the Stimulus class @copyright: University of Tennessee, 2007 @license: This software is provided as part of the DANSE project. """ import testcase_generator as generator from testcase_generator import Stimulus try: import VolumeCanvas print "Testing local VolumeCanvas\n" except: import sans.realspace.VolumeCanvas as VolumeCanvas def randomModel(): """ Return a random model name """ from random import random from math import floor model_list = ['sphere', 'cylinder', 'ellipsoid', 'singlehelix'] rnd_id = int(floor(random()*len(model_list))) return model_list[rnd_id] class Simulation: """Simulation testing class""" # Probability of stopping a test case without # adding additional stimuli end_frequency = 0.1 def __init__(self): self.end_flag = False def reset(self): self.end_flag = False def setup(self): """Setup the canvas for stimuli application""" return VolumeCanvas.VolumeCanvas() def getRandomStimulus(self): """Return the name of a random stimulus""" from random import random # If the end flag is up, compute I(q) one last time if self.end_flag == True: return None # Get the list of stimuli attrs = dir(self) stim_list = [] # Get the list of simuli and # compute the total normalization sum = 0 for item in attrs: if item.endswith('Stimulus'): obj = getattr(self, item) if hasattr(obj, 'frequency'): stim_list.append(item) sum += getattr(self, item).frequency # Choose a stimulus rnd = random() # Check if we need to stop if rnd < self.end_frequency: self.end_flag = True return "GetIq" run_sum = 0 for item in stim_list: run_sum += getattr(self, item).frequency/sum if run_sum >= rnd: pos = item.index('Stimulus') return item[0:pos] class AddStimulus(Stimulus): """Add an object to the canvas""" frequency = 1.0 def __call__(self, canvas): """Apply stimulus""" report = generator.StimulusReport(tag=self.name) # Select random model add_model = randomModel() # Add the model handle = canvas.add(add_model) # Check that it is in the list of objects if handle in canvas.getShapeList(): report.passed = 1 else: report.log = "Add: tried to add %s" % add_model report.trace = "Added %s" % add_model return canvas, report class GetParamStimulus(Stimulus): """Get the value of a canvas or shape parameter""" frequency = 1.0 def __call__(self, canvas): """Apply stimulus""" from random import random import math report = generator.StimulusReport(tag=self.name) # Read a parameter #TODO: access shape parameters rnd = random() i_rnd = int(math.floor(rnd*len(canvas.params))) par_name = canvas.params.keys()[i_rnd] value = canvas.getParam(par_name) # Check that it is in the list of objects try: float(value) report.passed = 1 except: report.log = "get: bad value for [%s]" % par_name report.trace = "Read %s" % par_name return canvas, report class GetIqStimulus(Stimulus): """Calculate I(q)""" frequency = 1.0 def __call__(self, canvas): report = generator.StimulusReport(tag=self.name) # Check that a float is returned # Validation testing will be done elsewhere value = canvas.getIq(0.1) # Check that it is in the list of objects try: float(value) report.passed = 1 except: report.log = "GetIq: bad value for Iq "+str(value) report.trace = "I(q) = %g" % value return canvas, report class GetIq2DStimulus(Stimulus): """Calculate I(q)""" frequency = 1.0 def __call__(self, canvas): report = generator.StimulusReport(tag=self.name) # Check that a float is returned # Validation testing will be done elsewhere value = canvas.getIq2D(0.01,0.01) # Check that it is in the list of objects try: float(value) report.passed = 1 except: report.log = "GetIq2D: bad value for Iq "+str(value) report.trace = "I(q) = %g" % value return canvas, report class SetCanvasParamStimulus(Stimulus): """Set the value of a canvas parameter""" frequency = 1.0 def __call__(self, canvas): """Apply stimulus""" from random import random import math report = generator.StimulusReport(tag=self.name) # Read a parameter rnd = random() i_rnd = int(math.floor(rnd*len(canvas.params))) par_name = canvas.params.keys()[i_rnd] # Get current value current = canvas.getParam(par_name) # Set new value rnd2 = random() new_value = current*(1.0+rnd2) canvas.setParam(par_name, new_value) # Read it back current = canvas.getParam(par_name) # Check that the new value is correct if current == new_value: report.passed = 1 else: report.log = "get: bad value for [%s]" % par_name report.trace = "Read %s" % par_name return canvas, report class SetShapeParamStimulus(Stimulus): """Set the value of a canvas parameter""" frequency = 1.0 def __call__(self, canvas): """Apply stimulus""" from random import random import math report = generator.StimulusReport(tag=self.name) # Read a parameter rnd = random() shape_list = canvas.getShapeList() if len(shape_list)==0: # No object available, let's test an error condition try: canvas.setParam("shape0.test-radius", 1.0) report.log = "SetShapeParam: set didn't throw exception" except: report.passed = 1 report.trace = "SetShapeParam: testing failure behavior" else: i_rnd = int(math.floor(rnd*len(shape_list))) shape_name = shape_list[i_rnd] found = False while found==False: rnd2 = random() par_keys = canvas.shapes[shape_name].params.keys() i_rnd2 = int(math.floor(rnd2*(len(par_keys)))) short_name = par_keys[i_rnd2] par_name = "%s.%s" % (shape_name, short_name) if not short_name.lower() == "type": found= True # Get current value current = canvas.getParam(par_name) # Set new value if short_name in ['orientation', 'center']: new_value = [random(), random(), random()] else: new_value = current*(1.0+random()) canvas.setParam(par_name, new_value) # Read it back current = canvas.getParam(par_name) # Check that the new value is correct if current == new_value: report.passed = 1 else: report.log = "get: bad value for [%s]" % par_name report.trace = "Read %s" % par_name return canvas, report class RemoveStimulus(Stimulus): """Calculate I(q)""" frequency = 1.0 def __call__(self, canvas): from random import random import math report = generator.StimulusReport(tag=self.name) # Get list of shapes if len(canvas.shapes)>0: rnd = random() i_rnd = int(math.floor(rnd*len(canvas.shapes))) shape_name = canvas.shapes.keys()[i_rnd] canvas.delete(shape_name) if shape_name in canvas.shapes.keys() \ or shape_name in canvas.getShapeList(): report.log = "Remove: object was not removed" else: report.passed = 1 else: # No shape to remove: test bad remove call try: canvas.delete("test-shape") report.log = "Remove: delete didn't throw exception" except: report.passed = 1 report.trace = "Remove: testing failure behavior" return canvas, report if __name__ == '__main__': import sys stimuli = Simulation() if len(sys.argv)>1: #t = generator.TestCase(stimuli, filename = "test-case.xml") t = generator.TestCase(stimuli, filename = sys.argv[1]) print "Test passed =", t.run() else: g = generator.TestCaseGenerator(stimuli) g.generateAndRun(200)