class TestConvergencePlot(unittest.TestCase):
def setUp(self):
self.analysis_model = AnalysisModel()
self.data_view = SamplingDataView(analysis_model=self.analysis_model)
self.plot = self.data_view.sampling_plot
self.analysis_model.notify(("x", "y", "E"))
def check_update_is_requested_and_apply(self):
# check
self.assertTrue(self.plot.update_required)
self.assertTrue(self.plot.plot_updater.active)
# update
self.plot._update_displayable_value_names()
self.plot._update_plot()
self.plot.update_required = False
def add_data_points(self):
# non monotonic convergence
for datum in [
(1, 3, 8), (2, 5, 7.7), (3, 2.6, 7.6), (4, 2.5, 7.62),
(5, 2.47, 7.543), (6, 2.465, 7.54)]:
self.analysis_model.notify(datum)
self.check_update_is_requested_and_apply()
def test_init(self):
self.analysis_model.notify((1.0, 1.0, 1.0))
self.plot._update_displayable_value_names()
self.assertIsInstance(self.plot._axis, LinePlot)
self.assertEqual('iteration', self.plot._plot.x_axis.title)
def test_resize_plot(self):
ranges = self.plot.recenter_plot()
self.assertEqual(ranges, (0.0, 1.0, -0.1, 1))
self.add_data_points()
self.plot._update_plot()
ranges = self.plot._get_plot_range()
# the second value for y (5) shouldn't contribute to the running min.
self.assertEqual(3.1, ranges[3])
self.assertEqual('iteration', self.plot._plot.x_axis.title)
def test_change_variable(self):
self.add_data_points()
self.plot._update_plot()
self.assertListEqual(
[3, 3, 2.6, 2.5, 2.47, 2.465],
self.plot._custom_data_array)
self.plot.y = "E"
self.assertListEqual(
[8, 7.7, 7.6, 7.6, 7.543, 7.54],
self.plot._custom_data_array)
def test_notify(self):
with mock.patch.object(AnalysisModel, "_add_header") as mock_header:
model = AnalysisModel()
model.notify(None)
mock_header.assert_called_once()
with mock.patch.object(
AnalysisModel,
"_add_header") as mock_header, mock.patch.object(
AnalysisModel,
"_add_metadata") as mock_metadata, mock.patch.object(
AnalysisModel, "_add_data") as mock_data:
model = AnalysisModel()
model.notify(None)
# This line is necessary because the header must be set
# in order to add data to the model.
model.header = ("", )
model.notify(None, metadata=True)
model.notify(None)
mock_header.assert_called_once()
mock_metadata.assert_called_once()
mock_data.assert_called_once()
class TestResultsTable(unittest.TestCase):
def setUp(self):
self.analysis_model = AnalysisModel()
self.analysis_model.header = ["x", "y", "compound"]
self.analysis_model.notify((2.1, 56, "CO"))
self.analysis_model.notify((1.23, 51.2, "CO2"))
self.results_table = ResultsTable(analysis_model=self.analysis_model)
def test_columns(self):
self.assertEqual(len(self.results_table.columns), 3)
self.assertEqual(self.results_table.columns[2].label, "compound")
def test_value(self):
column_2 = self.results_table.columns[2]
row_1 = self.analysis_model.evaluation_steps[1]
self.assertEqual(column_2.get_value(row_1), "CO2")
def test_append_evaluation_steps(self):
self.analysis_model.notify((1.5, 50, "CO"))
self.assertEqual(self.results_table.rows[2], (1.5, 50, "CO"))
def test_selection(self):
# From table to the model
self.assertIsNone(self.analysis_model.selected_step_indices)
self.results_table._selected_rows = [(2.1, 56, "CO")]
self.assertEqual(self.analysis_model.selected_step_indices, [0])
self.results_table._selected_rows = [(1.23, 51.2, "CO2")]
self.assertEqual(self.analysis_model.selected_step_indices, [1])
self.results_table._selected_rows = []
self.assertIsNone(self.analysis_model.selected_step_indices)
# From model to the table
self.analysis_model.selected_step_indices = [1]
self.assertEqual(self.results_table._selected_rows,
[(1.23, 51.2, "CO2")])
self.analysis_model.selected_step_indices = [0, 1]
self.assertEqual(
self.results_table._selected_rows,
[(2.1, 56, "CO"), (1.23, 51.2, "CO2")],
)
self.analysis_model.selected_step_indices = None
self.assertEqual(self.results_table._selected_rows, [])
class TestAnalysisModel(TestCase):
def setUp(self):
self.model = AnalysisModel()
self.header = ("a", "b", "c")
self.data = ((1, 2, 3), (4, 5, 6))
self.metadata = ({}, {'d': 10})
self.state_dict = {
"header": self.header,
"1": {
'data': self.data[0],
'metadata': self.metadata[0]
},
"2": {
'data': self.data[1],
'metadata': self.metadata[1]
}
}
def test_initialize(self):
self.assertEqual(tuple(), self.model.header)
self.assertFalse(self.model.export_enabled)
self.assertFalse(self.model._export_enabled)
self.assertEqual([], self.model.evaluation_steps)
self.assertEqual([], self.model._evaluation_steps)
self.assertDictEqual(self.model._row_data, {})
def test__add_header(self):
self.model._add_header(self.header)
self.assertTupleEqual(self.header, self.model.header)
self.assertDictEqual(dict.fromkeys(self.header), self.model._row_data)
wrong_header = 1
log_error = (
f"The Header of the AnalysisModel can't be defined by "
f"the {wrong_header}. A list or tuple of strings is required.")
with LogCapture() as capture:
with self.assertRaises(TraitError):
self.model._add_header(wrong_header)
capture.check(
("force_wfmanager.model.analysis_model", "ERROR", log_error))
def test_header_update(self):
self.model.from_json(self.state_dict)
self.assertTupleEqual(self.model.header, self.header)
self.model.header = ("new", )
self.assertTupleEqual(self.model.header, ("new", ))
self.assertEqual([], self.model.evaluation_steps)
self.assertEqual([], self.model.step_metadata)
def test__add_cell(self):
self.model._add_header(self.header)
self.model._add_cell(label="a", value="1")
self.assertEqual("1", self.model._row_data["a"])
self.assertIsNone(self.model._row_data["b"])
self.assertIsNone(self.model._row_data["c"])
label, value = "d", "2"
log_warning = (f"The AnalysisModel does not have the {label} column."
f"The value {value} has not been added to the table.")
with LogCapture() as capture:
self.model._add_cell(label=label, value=value)
capture.check((
"force_wfmanager.model.analysis_model",
"WARNING",
log_warning,
))
self.assertEqual("1", self.model._row_data["a"])
self.assertIsNone(self.model._row_data["b"])
self.assertIsNone(self.model._row_data["c"])
def test__add_cells(self):
self.model._add_header(self.header)
data = []
self.model._add_cells(data)
self.assertIsNone(self.model._row_data["a"])
self.assertIsNone(self.model._row_data["b"])
self.assertIsNone(self.model._row_data["c"])
data = (1, 2)
self.model._add_cells(data)
self.assertEqual(1, self.model._row_data["a"])
self.assertEqual(2, self.model._row_data["b"])
self.assertIsNone(self.model._row_data["c"])
data = (1, 2, 3, 4)
self.model._add_cells(data)
self.assertEqual(1, self.model._row_data["a"])
self.assertEqual(2, self.model._row_data["b"])
self.assertEqual(3, self.model._row_data["c"])
def test_finalize_row(self):
self.model._add_header(self.header)
data = (1, 2, 3)
self.model._add_cells(data)
self.model._finalize_row()
self.assertEqual(1, len(self.model.evaluation_steps))
self.assertEqual(1, len(self.model.step_metadata))
self.assertTupleEqual(data, self.model.evaluation_steps[0])
self.assertDictEqual({}, self.model.step_metadata[0])
self.assertDictEqual(self.model._row_data,
self.model._row_data_default())
self.assertDictEqual(self.model._row_metadata,
self.model._row_metadata_default())
data = (1, 2)
self.model._add_cells(data)
self.model._add_metadata({'extra': 7})
self.model._finalize_row()
self.assertEqual(2, len(self.model.evaluation_steps))
self.assertEqual(2, len(self.model.step_metadata))
self.assertTupleEqual((1, 2, None), self.model.evaluation_steps[1])
self.assertDictEqual({'extra': 7}, self.model.step_metadata[1])
self.assertDictEqual(self.model._row_data,
self.model._row_data_default())
self.assertDictEqual(self.model._row_metadata,
self.model._row_metadata_default())
def test__add_data(self):
with mock.patch.object(AnalysisModel, "_add_cell") as mock_cell:
model = AnalysisModel()
model._add_data({"name": "value"})
mock_cell.assert_called_with("name", "value")
with mock.patch.object(AnalysisModel,
"_add_cells") as mock_cells, mock.patch.object(
AnalysisModel, "_finalize_row"):
model = AnalysisModel()
model._add_data(["data", "entries"])
mock_cells.assert_called_with(["data", "entries"])
self.model._add_header(self.header)
self.model._add_data({"a": 1})
self.model._add_data({"c": 2})
self.model._add_data([3, 4])
self.assertEqual(1, len(self.model.evaluation_steps))
self.assertTupleEqual((3, 4, 2), self.model.evaluation_steps[0])
self.assertDictEqual(self.model._row_data,
self.model._row_data_default())
def test__add_metadata(self):
self.model._add_metadata({"a": 1})
self.assertDictEqual({"a": 1}, self.model._row_metadata)
self.model._add_metadata({"a": 1, "c": 2})
self.assertDictEqual({"a": 1, "c": 2}, self.model._row_metadata)
self.model._add_metadata([(3, 2, 5)])
self.assertDictEqual({"a": 1, "c": 2}, self.model._row_metadata)
def test__add_evaluation_step(self):
with self.assertRaisesRegex(
ValueError,
"Cannot add evaluation step to an empty Analysis model"):
self.model._add_evaluation_step(None)
self.model.notify(["column"])
step = (1, 2)
error = ("Size of evaluation step is incompatible with the length of "
"the header.")
with self.assertRaisesRegex(ValueError, error):
self.model._add_evaluation_step(step)
step = (1, )
self.model._add_evaluation_step(step)
self.assertEqual(1, len(self.model.evaluation_steps))
self.assertTupleEqual(self.model.evaluation_steps[0], step)
self.assertTrue(self.model.export_enabled)
def test_notify(self):
with mock.patch.object(AnalysisModel, "_add_header") as mock_header:
model = AnalysisModel()
model.notify(None)
mock_header.assert_called_once()
with mock.patch.object(
AnalysisModel,
"_add_header") as mock_header, mock.patch.object(
AnalysisModel,
"_add_metadata") as mock_metadata, mock.patch.object(
AnalysisModel, "_add_data") as mock_data:
model = AnalysisModel()
model.notify(None)
# This line is necessary because the header must be set
# in order to add data to the model.
model.header = ("", )
model.notify(None, metadata=True)
model.notify(None)
mock_header.assert_called_once()
mock_metadata.assert_called_once()
mock_data.assert_called_once()
def test_column(self):
self.model.from_json(self.state_dict)
column_by_id = self.model.column(0)
column_by_label = self.model.column("a")
self.assertListEqual(column_by_id, column_by_label)
self.assertListEqual(column_by_label, [1, 4])
column_by_id = self.model.column(-1)
column_by_label = self.model.column("c")
self.assertListEqual(column_by_id, column_by_label)
self.assertListEqual(column_by_label, [3, 6])
error = ("Column of the AnalysisModel with label 2"
" doesn't exist. The label must be a string or int.")
with self.assertRaisesRegex(ValueError, error):
self.model.column("2")
error = ("Column of the AnalysisModel with label 100"
" doesn't exist. The label must be a string or int.")
with self.assertRaisesRegex(ValueError, error):
self.model.column(100)
def test_is_empty(self):
self.assertTrue(self.model.is_empty)
self.model.from_json(self.state_dict)
self.assertFalse(self.model.is_empty)
def test___getstate__(self):
self.model.notify(self.header)
for entry, meta in zip(self.data, self.metadata):
self.model.notify(meta, metadata=True)
self.model.notify(entry)
state = self.model.__getstate__()
self.assertDictEqual(
{
"header": self.header,
1: {
'data': self.data[0],
'metadata': self.metadata[0]
},
2: {
'data': self.data[1],
'metadata': self.metadata[1]
}
}, state)
with mock.patch.object(AnalysisModel, "__getstate__") as mock_getstate:
AnalysisModel().to_json()
mock_getstate.assert_called_once()
def test_json(self):
error = ("AnalysisModel can't be instantiated from a data dictionary"
" that does not contain a header.")
with LogCapture() as capture:
with self.assertRaisesRegex(KeyError, error):
AnalysisModel().from_json({
"1": {
'data': self.data[0],
'metadata': self.metadata[0]
},
"2": {
'data': self.data[1],
'metadata': self.metadata[1]
}
})
capture.check((
"force_wfmanager.model.analysis_model",
"ERROR",
"AnalysisModel can't be instantiated from a data dictionary "
"that does not contain a header.",
))
with mock.patch.object(AnalysisModel, "clear") as mock_clear:
model = AnalysisModel()
model.from_json(self.state_dict)
mock_clear.assert_called_once()
self.model.from_json(self.state_dict)
self.assertTupleEqual(self.model.header, self.header)
self.assertEqual(2, len(self.model.evaluation_steps))
self.assertEqual(2, len(self.model.step_metadata))
self.assertTupleEqual(self.data[0], self.model.evaluation_steps[0])
self.assertDictEqual(self.metadata[0], self.model.step_metadata[0])
self.assertTupleEqual(self.data[1], self.model.evaluation_steps[1])
self.assertDictEqual(self.metadata[1], self.model.step_metadata[1])
tmp_file = tempfile.NamedTemporaryFile()
filename = tmp_file.name
self.model.dump_json(filename)
with open(filename) as f:
json_data = json.load(f)
self.assertDictEqual(
{
"header": list(self.header),
"1": {
'data': list(self.data[0]),
'metadata': self.metadata[0]
},
"2": {
'data': list(self.data[1]),
'metadata': self.metadata[1]
}
}, json_data)
self.model._export_enabled = False
self.assertFalse(self.model.dump_json(None))
state_dict = {
"header": self.header,
"1": {
'data': self.data[0],
'metadata': self.metadata[0]
},
"3": {
'data': self.data[1],
'metadata': self.metadata[1]
}
}
with LogCapture() as capture:
AnalysisModel().from_json(state_dict)
capture.check((
"force_wfmanager.model.analysis_model",
"WARNING",
"Can't find a row with index 2. This index will "
"be skipped in the AnalysisModel.",
))
with self.subTest("Check deprecated formats"):
# TODO: This test can be removed when issue #414
# is resolved
state_dict = {"header": self.header, "1": list(self.data[0])}
with LogCapture() as capture:
model = AnalysisModel()
model.from_json(state_dict)
capture.check((
"force_wfmanager.model.analysis_model",
"WARNING",
"Project file format is deprecated and will"
" be removed in version 0.7.0",
))
self.assertDictEqual(
{
"header": self.header,
1: {
"data": self.data[0],
"metadata": {}
}
}, model.__getstate__())
def test_write_csv(self):
self.model.from_json(self.state_dict)
tmp_file = tempfile.NamedTemporaryFile()
filename = tmp_file.name
self.model.dump_csv(filename)
with open(filename) as f:
csv_data = f.read()
self.assertEqual("a,b,c\n1,2,3\n4,5,6\n", csv_data)
self.model._export_enabled = False
self.assertFalse(self.model.dump_csv(None))
def test_write(self):
with mock.patch.object(AnalysisModel, "dump_csv") as mock_csv:
AnalysisModel().write("filename.csv")
mock_csv.assert_called_with("filename.csv", mode="w")
with mock.patch.object(AnalysisModel, "dump_json") as mock_json:
AnalysisModel().write("filename.json")
mock_json.assert_called_with("filename.json", mode="w")
error = "AnalysisModel can only write to .json or .csv formats."
with self.assertRaisesRegex(IOError, error):
AnalysisModel().write("filename.format")
def test_clear(self):
self.model.notify(self.header)
for entry, meta in zip(self.data, self.metadata):
self.model.notify(meta, metadata=True)
self.model.notify(entry)
self.model.clear_steps()
self.assertFalse(self.model.export_enabled)
self.assertEqual([], self.model.evaluation_steps)
self.assertEqual([], self.model.step_metadata)
self.assertTupleEqual(self.model.header, self.header)
self.model.notify(self.header)
for entry in self.data:
self.model.notify(entry)
self.model.clear()
self.assertFalse(self.model.export_enabled)
self.assertEqual([], self.model.evaluation_steps)
self.assertEqual([], self.model.step_metadata)
self.assertTupleEqual(self.model.header, ())
def test_selected_step_indices(self):
self.assertIsNone(self.model.selected_step_indices)
self.model.notify(self.header)
for entry in self.data:
self.model.notify(entry)
self.assertIsNone(self.model.selected_step_indices)
self.model.selected_step_indices = [1]
self.assertListEqual(self.model.selected_step_indices, [1])
self.model.selected_step_indices = [1, 2]
self.assertListEqual(self.model.selected_step_indices, [1, 2])
error = ("Invalid value for selection index 3. "
"It must be a positive Int less or equal to 1")
with self.assertRaisesRegex(ValueError, error):
self.model.selected_step_indices = [1, 3]
error = ("Invalid value for selection index -1. "
"It must be a positive Int less or equal to 1")
with self.assertRaisesRegex(ValueError, error):
self.model.selected_step_indices = [-1]