class test_units(unittest.TestCase):
def setUp(self):
self.engine = Coordinator()
pass
def test_basic_conversion(self):
# add models to engine
mdl1 = '../../app_data/models/randomizer/randomizer.mdl'
params = parse_json(mdl1)
params.update({'id': 'm1', 'mdl': mdl1, 'model_type': 'mdl'})
m1 = self.engine.add_model(**params)
mdl2 = '../../app_data/models/multiplier/multiplier.mdl'
params = parse_json(mdl2)
params.update({'id': 'm2', 'mdl': mdl2, 'model_type': 'mdl'})
m2 = self.engine.add_model(**params)
self.assertTrue(m1)
self.assertTrue(m2)
self.assertTrue(len(self.engine.Models()) == 2)
mi1 = m1.instance()
mi2 = m2.instance()
# run the model mi1
mi1.run(inputs={})
# get the oei and iei that will be mapped
oei = mi1.outputs()['random POINT 1-10']
outvals = numpy.array(oei.getValues2())
iei = mi2.inputs()['some_value']
converted_vals = units.convert_units(oei=oei, iei=iei, vals=outvals)
print 'done'
class test_link_order(unittest.TestCase):
def setUp(self):
# initialize environment variables
environment.getEnvironmentVars()
if sys.gettrace():
print 'Detected Debug Mode'
# initialize debug listener (reroute messages to console)
self.d = sprint.DebugListener()
self.engine = Coordinator()
sprint.PrintTarget.CONSOLE = 1134
self.g = n.DiGraph()
def tearDown(self):
del self.g
def test_determine_execution_order(self):
# add models
mdl1 = '../data/randomizer.mdl'
mdl2 = '../data/multiplier.mdl'
# add both models. ids can be anything
id1 = 'm1'
id2 = 'm2'
m1 = self.engine.add_model(id=id1, attrib={'mdl': mdl1})
m2 = self.engine.add_model(id=id2, attrib={'mdl': mdl2})
self.assertTrue(m1)
self.assertTrue(m2)
self.assertTrue(len(self.engine.Models()) == 2)
# create link
rand_oei = self.engine.get_exchange_item_info(
modelid=id1, eitype=stdlib.ExchangeItemType.OUTPUT)
mult_iei = self.engine.get_exchange_item_info(
modelid=id2, eitype=stdlib.ExchangeItemType.INPUT)
self.engine.add_link(from_id=id1,
from_item_id=rand_oei[0]['name'],
to_id=id2,
to_item_id=mult_iei[0]['name'],
spatial_interp=None,
temporal_interp=None,
uid=None)
self.assertTrue(len(self.engine.get_all_links()) == 1)
# get execution order
order = self.engine.determine_execution_order()
self.assertTrue(''.join(order) == 'm1m2')
def test_basic(self):
"""
m1 -> m2 -> m3 -> m4 -> m5 -> m6
"""
# add some edges to simulate links
self.g.add_edge('m1', 'm2')
self.g.add_edge('m2', 'm3')
self.g.add_edge('m3', 'm4')
self.g.add_edge('m4', 'm5')
self.g.add_edge('m5', 'm6')
order = n.topological_sort(self.g)
self.assertTrue(''.join(order) == 'm1m2m3m4m5m6')
def test_simple_tree(self):
"""
m1 -> m2
-> m3
m6 -> m5 -> m4
"""
self.g.add_edge('m1', 'm2')
self.g.add_edge('m2', 'm3')
self.g.add_edge('m6', 'm5')
self.g.add_edge('m5', 'm4')
self.g.add_edge('m4', 'm3')
order = n.topological_sort(self.g)
self.assertTrue(''.join(order) == 'm1m2m6m5m4m3')
def test_loop(self):
"""
m6 -> m5 -> m4 \
^ \
| -> m3
<- |m1| -> m2 /
"""
self.g.add_edge('m1', 'm2')
self.g.add_edge('m1', 'm6')
self.g.add_edge('m2', 'm3')
self.g.add_edge('m6', 'm5')
self.g.add_edge('m5', 'm4')
self.g.add_edge('m4', 'm3')
order = n.topological_sort(self.g)
self.assertTrue(''.join(order) == 'm1m2m6m5m4m3')
def test_bidirectional(self):
"""
m1 <-> m2 -> m3
"""
self.g.add_edge('m1', 'm2')
self.g.add_edge('m2', 'm3')
self.g.add_edge('m3', 'm2')
self.g.add_edge('m3', 'm4')
# remove any models that done have links
#for
# determine cycles
cycles = n.recursive_simple_cycles(self.g)
for cycle in cycles:
# remove edges that form cycles
self.g.remove_edge(cycle[0], cycle[1])
# perform toposort
order = n.topological_sort(self.g)
# re-add bidirectional dependencies (i.e. cycles)
for cycle in cycles:
# find index of inverse link
for i in xrange(0, len(order) - 1):
if order[i] == cycle[1] and order[i + 1] == cycle[0]:
order.insert(i + 2, cycle[1])
order.insert(i + 3, cycle[0])
break
self.assertTrue(''.join(order) == 'm1m2m3m2m3m4')
