def test_table_1(self, mock_fdf):
'''
_make_table():
- "count frequency" is True
- file extension on "pathname" with period
'''
test_wm = WorkflowManager([])
test_wm.settings(None, 'count frequency', True) # just to be 100% clear
mock_fdf.return_value = mock.MagicMock(spec_set=pandas.DataFrame)
test_wm._result = mock_fdf.return_value # to avoid a RuntimeError
test_wm._previous_exp = 'intervals' # to get the proper "exp_name"
top_x = None
threshold = None
exp_name = 'Interval Frequency'
pathname = 'test_path'
test_wm._make_table('CSV', pathname + '.csv', top_x, threshold) # pylint: disable=protected-access
test_wm._make_table('Excel', pathname + '.xlsx', top_x, threshold) # pylint: disable=protected-access
test_wm._make_table('Stata', pathname + '.dta', top_x, threshold) # pylint: disable=protected-access
test_wm._make_table('HTML', pathname + '.html', top_x, threshold) # pylint: disable=protected-access
mock_fdf.return_value.to_csv.assert_called_once_with('test_path.csv')
mock_fdf.return_value.to_stata.assert_called_once_with('test_path.dta')
mock_fdf.return_value.to_excel.assert_called_once_with('test_path.xlsx')
mock_fdf.return_value.to_html.assert_called_once_with('test_path.html')
self.assertSequenceEqual([mock.call(top_x=top_x, threshold=threshold, name=exp_name) for _ in range(4)],
mock_fdf.call_args_list)
def test_table_3(self):
'''
_make_table():
- "count frequency" is False
- file extension not on "pathname"
- there are several IndexedPiece objects
'''
test_wm = WorkflowManager([])
test_wm.settings(None, 'count frequency', False)
test_wm._result = [mock.MagicMock(spec_set=pandas.DataFrame) for _ in range(5)]
test_wm._data = ['boop' for _ in range(len(test_wm._result))]
top_x = None
threshold = None
pathname = 'test_path'
test_wm._make_table('CSV', pathname, top_x, threshold) # pylint: disable=protected-access
test_wm._make_table('Excel', pathname, top_x, threshold) # pylint: disable=protected-access
test_wm._make_table('Stata', pathname, top_x, threshold) # pylint: disable=protected-access
test_wm._make_table('HTML', pathname, top_x, threshold) # pylint: disable=protected-access
for i in range(len(test_wm._result)):
test_wm._result[i].to_csv.assert_called_once_with('{}-{}{}'.format(pathname, i, '.csv'))
test_wm._result[i].to_excel.assert_called_once_with('{}-{}{}'.format(pathname, i, '.xlsx'))
test_wm._result[i].to_stata.assert_called_once_with('{}-{}{}'.format(pathname, i, '.dta'))
test_wm._result[i].to_html.assert_called_once_with('{}-{}{}'.format(pathname, i, '.html'))
def test_lilypond_2(self):
"""one piece with one part; specified pathname; and there's a NaN in the Series!"""
mock_ips = [MagicMock(spec_set=IndexedPiece)]
mock_ips[0].get_data.return_value = ['ready for LilyPond']
result = [pandas.DataFrame({('analyzer', 'clarinet'): pandas.Series(range(10))})]
result[0][('analyzer', 'clarinet')].iloc[0] = NaN
exp_series = pandas.Series(range(1, 10), index=range(1, 10)) # to test dropna() was run
pathname = 'this_path'
expected = ['this_path.ly']
exp_get_data_calls = [mock.call([lilypond_ind.AnnotationIndexer,
lilypond_exp.AnnotateTheNoteExperimenter,
lilypond_exp.PartNotesExperimenter],
{'part_names': ['analyzer: clarinet'],
'column': 'lilypond.AnnotationIndexer'},
[mock.ANY]),
mock.call([lilypond_exp.LilyPondExperimenter],
{'run_lilypond': True,
'annotation_part': ['ready for LilyPond'],
'output_pathname': expected[0]})]
test_wm = WorkflowManager(mock_ips)
test_wm._result = result
test_wm.settings(None, 'count frequency', False)
test_wm._make_lilypond(pathname)
self.assertEqual(2, mock_ips[0].get_data.call_count)
self.assertSequenceEqual(exp_get_data_calls, mock_ips[0].get_data.call_args_list)
# check that dropna() was run
actual_series = mock_ips[0].get_data.call_args_list[0][0][2][0]
self.assertSequenceEqual(list(exp_series.index), list(actual_series.index))
self.assertSequenceEqual(list(exp_series.values), list(actual_series.values))
def test_lilypond_1b(self):
"""error conditions: if the lengths are different, (but 'count frequency' is okay)"""
test_wm = WorkflowManager(['fake piece'])
test_wm._data = ['fake IndexedPiece']
test_wm._result = ['fake results', 'more fake results', 'so many fake results']
test_wm.settings(None, 'count frequency', False)
with self.assertRaises(RuntimeError) as run_err:
test_wm._make_lilypond(['paths'])
self.assertEqual(WorkflowManager._COUNT_FREQUENCY_MESSAGE, run_err.exception.args[0])
def test_output_4(self):
"""ensure RuntimeError if self._result is None"""
test_wc = WorkflowManager([])
test_wc._result = None # just in case
self.assertRaises(RuntimeError, test_wc.output, 'R histogram')
try:
test_wc.output('R histogram')
except RuntimeError as run_err:
self.assertEqual(WorkflowManager._NO_RESULTS_ERROR, run_err.args[0])
def test_get_dataframe_4(self):
# test with name=auto, top_x=5, threshold=7 (so threshold leaves fewer than 3 results)
test_wc = WorkflowManager([])
test_wc._result = pandas.Series([i for i in xrange(10, 0, -1)])
expected = pandas.DataFrame({'data': pandas.Series([10, 9, 8])})
actual = test_wc._get_dataframe(top_x=5, threshold=7)
self.assertEqual(len(expected.columns), len(actual.columns))
for i in expected.columns:
self.assertSequenceEqual(list(expected.loc[:,i].index), list(actual.loc[:,i].index))
self.assertSequenceEqual(list(expected.loc[:,i].values), list(actual.loc[:,i].values))
def test_lilypond_1a(self):
"""error conditions: if 'count frequency' is True (but the lengths are okay)"""
test_wm = WorkflowManager(['fake piece'])
test_wm._data = ['fake IndexedPiece']
test_wm._result = ['fake results']
# test twice like this to make sure (1) the try/except will definitely catch something, and
# (2) we're not getting hit by another RuntimeError, of which there could be many
with self.assertRaises(RuntimeError) as run_err:
test_wm._make_lilypond(['paths'])
self.assertEqual(WorkflowManager._COUNT_FREQUENCY_MESSAGE, run_err.exception.args[0])
def test_filter_dataframe_3(self):
"""test with top_x=3, threshold=5 (so the top_x still removes after threshold), name=auto"""
test_wc = WorkflowManager([])
test_wc._result = pandas.DataFrame({'data': pandas.Series([i for i in range(10, 0, -1)])})
expected = pandas.DataFrame({'data': pandas.Series([10, 9, 8])})
actual = test_wc._filter_dataframe(top_x=3, threshold=5)
self.assertEqual(len(expected.columns), len(actual.columns))
for i in expected.columns:
self.assertSequenceEqual(list(expected[i].index), list(actual[i].index))
self.assertSequenceEqual(list(expected[i].values), list(actual[i].values))
def test_get_dataframe_2(self):
# test with name='asdf', top_x=3, threshold=auto
test_wc = WorkflowManager([])
test_wc._result = pandas.Series([i for i in xrange(10, 0, -1)])
expected = pandas.DataFrame({'asdf': pandas.Series([10, 9, 8])})
actual = test_wc._get_dataframe('asdf', 3)
self.assertEqual(len(expected.columns), len(actual.columns))
for i in expected.columns:
self.assertSequenceEqual(list(expected.loc[:,i].index), list(actual.loc[:,i].index))
self.assertSequenceEqual(list(expected.loc[:,i].values), list(actual.loc[:,i].values))
def test_output_3(self):
"""ensure RuntimeError if there's an invalid instruction"""
test_wc = WorkflowManager([])
test_wc._result = [5] # make sure that's not what causes it
bad_instruction = 'eat dirt'
self.assertRaises(RuntimeError, test_wc.output, bad_instruction)
try:
test_wc.output(bad_instruction)
except RuntimeError as run_err:
self.assertEqual(WorkflowManager._UNRECOGNIZED_INSTRUCTION.format(bad_instruction),
run_err.args[0])
def test_lilypond_1b(self):
# error conditions: if the lengths are different, (but 'count frequency' is okay)
test_wm = WorkflowManager(['fake piece'])
test_wm._data = ['fake IndexedPiece']
test_wm._result = ['fake results', 'more fake results', 'so many fake results']
test_wm.settings(None, 'count frequency', False)
self.assertRaises(RuntimeError, test_wm._make_lilypond, ['paths'])
try:
test_wm._make_lilypond(['paths'])
except RuntimeError as the_err:
self.assertEqual(WorkflowManager._COUNT_FREQUENCY_MESSAGE, the_err.message)
def test_filter_dataframe_5(self):
"""test with top_x=3, threshold=auto, name='asdf'; many input columns"""
test_wc = WorkflowManager([])
test_wc._result = pandas.DataFrame({('1', 'b'): pandas.Series([i for i in range(10, 0, -1)]),
('1', 'z'): pandas.Series([i for i in range(10, 20)]),
('2', 'e'): pandas.Series([i for i in range(40, 900)])})
expected = pandas.DataFrame({'asdf': pandas.Series([10, 9, 8])})
actual = test_wc._filter_dataframe(top_x=3, name='asdf')
self.assertEqual(len(expected.columns), len(actual.columns))
for i in expected.columns:
self.assertSequenceEqual(list(expected[i].index), list(actual[i].index))
self.assertSequenceEqual(list(expected[i].values), list(actual[i].values))
def test_export_3(self, mock_gdf):
# --> the method works as expected for CSV, Excel, and Stata when _result is a DataFrame
test_wm = WorkflowManager([])
test_wm._result = mock.MagicMock(spec=pandas.DataFrame)
test_wm.export(u'CSV', u'test_path')
test_wm.export(u'Excel', u'test_path')
test_wm.export(u'Stata', u'test_path')
test_wm.export(u'HTML', u'test_path')
test_wm._result.to_csv.assert_called_once_with(u'test_path.csv')
test_wm._result.to_stata.assert_called_once_with(u'test_path.dta')
test_wm._result.to_excel.assert_called_once_with(u'test_path.xlsx')
test_wm._result.to_html.assert_called_once_with(u'test_path.html')
self.assertEqual(0, mock_gdf.call_count)
def test_export_4(self, mock_gdf):
# --> test_export_3() with a valid extension already on
test_wm = WorkflowManager([])
test_wm._result = mock.MagicMock(spec=pandas.DataFrame)
test_wm.export(u'CSV', u'test_path.csv')
test_wm.export(u'Excel', u'test_path.xlsx')
test_wm.export(u'Stata', u'test_path.dta')
test_wm.export(u'HTML', u'test_path.html')
test_wm._result.to_csv.assert_called_once_with(u'test_path.csv')
test_wm._result.to_stata.assert_called_once_with(u'test_path.dta')
test_wm._result.to_excel.assert_called_once_with(u'test_path.xlsx')
test_wm._result.to_html.assert_called_once_with(u'test_path.html')
self.assertEqual(0, mock_gdf.call_count)
def test_histogram_1(self, mock_call, mock_gdf, mock_join):
# with specified pathname; last experiment was intervals with 20 pieces; self._result is DF
test_wc = WorkflowManager([])
test_wc._previous_exp = u'intervals'
test_wc._data = [1 for _ in xrange(20)]
test_wc._result = MagicMock(spec=pandas.DataFrame)
path = u'pathname!'
actual = test_wc._make_histogram(path)
self.assertEqual(0, mock_gdf.call_count)
expected_args = [u'Rscript', u'--vanilla', mock_join.return_value,
path + u'.dta', path + u'.png', u'int', u'20']
mock_call.assert_called_once_with(expected_args)
self.assertEqual(path + u'.png', actual)
self.assertEqual(1, mock_join.call_count)
def test_histogram_3(self, mock_call, mock_gdf, mock_join):
# test_ouput_6, plus top_x and threshold
test_wc = WorkflowManager([])
test_wc._previous_exp = u'n-grams'
test_wc._data = [1]
test_wc._shared_settings[u'n'] = 14
test_wc._result = MagicMock(spec=pandas.Series)
path = u'test_output/output_result'
actual = test_wc._make_histogram(top_x=420, threshold=1987)
mock_gdf.assert_called_once_with(u'freq', 420, 1987)
expected_args = [u'Rscript', u'--vanilla', mock_join.return_value,
path + u'.dta', path + u'.png', u'14', u'1']
mock_call.assert_called_once_with(expected_args)
self.assertEqual(path + u'.png', actual)
self.assertEqual(1, mock_join.call_count)
def test_histogram_2(self, mock_call, mock_gdf, mock_join):
# with unspecified pathname; last experiment was 14-grams with 1 piece; self._result is S
test_wc = WorkflowManager([])
test_wc._previous_exp = u'n-grams'
test_wc._data = [1]
test_wc._shared_settings[u'n'] = 14
test_wc._result = MagicMock(spec=pandas.Series)
path = u'test_output/output_result'
actual = test_wc._make_histogram()
mock_gdf.assert_called_once_with(u'freq', None, None)
expected_args = [u'Rscript', u'--vanilla', mock_join.return_value,
path + u'.dta', path + u'.png', u'14', u'1']
mock_call.assert_called_once_with(expected_args)
self.assertEqual(path + u'.png', actual)
self.assertEqual(1, mock_join.call_count)
def test_output_5(self, mock_table):
"""ensure output() calls export() as required"""
# 1: prepare
export_path = 'the_path'
mock_table.return_value = export_path
test_wc = WorkflowManager([])
test_wc._previous_exp = 'intervals'
test_wc._data = [1 for _ in range(20)]
test_wc._result = MagicMock(spec=pandas.DataFrame)
path = 'pathname!'
expected_args = ['Excel', path, None, None]
# 2: run
actual = test_wc.output('Excel', path)
# 3: check
self.assertEqual(export_path, actual)
mock_table.assert_called_once_with(*expected_args)
def test_output_2(self, mock_lily):
# ensure output() calls _make_lilypond() as required
# 1: prepare
lily_path = u'the_path'
mock_lily.return_value = lily_path
test_wc = WorkflowManager([])
test_wc._previous_exp = u'intervals'
test_wc._data = [1 for _ in xrange(20)]
test_wc._result = MagicMock(spec=pandas.DataFrame)
path = u'pathname!'
expected_args = [path]
# 2: run
actual = test_wc.output('LilyPond', path)
# 3: check
self.assertEqual(lily_path, actual)
mock_lily.assert_called_once_with(*expected_args)
def test_output_1b(self, mock_histo):
# ensure output() calls _make_histogram() as required (with 'R histogram' instruction)
# 1: prepare
histo_path = u'the_path.svg'
mock_histo.return_value = histo_path
test_wc = WorkflowManager([])
test_wc._previous_exp = u'intervals'
test_wc._data = [1 for _ in xrange(20)]
test_wc._result = MagicMock(spec=pandas.DataFrame)
path = u'pathname!'
top_x = 20
threshold = 10
expected_args = [path, top_x, threshold]
# 2: run
actual = test_wc.output('R histogram', path, top_x, threshold)
# 3: check
self.assertEqual(histo_path, actual)
mock_histo.assert_called_once_with(*expected_args)
def test_lilypond_4(self, mock_metadata, test_ip):
# make sure it works correctly with three pieces that have three parts
# ("voice combinations" is "all")
# 1: prepare
input_path = u'carpathia'
get_data_ret = lambda *x: ['** ' + str(x[1]) if len(x) == 2 else '** ' + str(x[2][0][-3:])]
num_parts = 3 # how many parts per piece? -- NB: diffferent from first test
piece_list = ['test_piece_1.mei', 'test_piece_2.mei', 'test_piece_3.mei']
test_wm = WorkflowManager(piece_list)
for i in xrange(len(piece_list)):
test_wm._data[i] = mock.MagicMock(spec_set=IndexedPiece)
test_wm._data[i].get_data.side_effect = get_data_ret
mock_metadata.return_value = ['part %i' % x for x in xrange(num_parts)]
# the results will be like this: [['fake result 0-0', 'fake result 0-1'],
# ['fake result 1-0', 'fake result 1-1']]
exp_results = [['fake result ' + str(i) + '-' + str(j) for j in xrange(num_parts)] \
for i in xrange(len(piece_list))]
test_wm._result = exp_results
test_wm.settings(None, 'count frequency', False)
test_wm.settings(0, 'voice combinations', 'all')
test_wm.settings(1, 'voice combinations', 'all')
test_wm.settings(2, 'voice combinations', 'all')
#exp_part_labels = [[[0, 2], [1, 2]] for _ in xrange(len(piece_list))]
# 2: run
test_wm._make_lilypond(input_path)
# 3: check
self.assertEqual(len(piece_list), test_ip.call_count) # even though we don't use them
lily_ind_list = [lilypond.AnnotationIndexer,
lilypond.AnnotateTheNoteIndexer,
lilypond.PartNotesIndexer]
for i, piece in enumerate(test_wm._data):
self.assertEqual(num_parts + 1, piece.get_data.call_count)
for j in xrange(num_parts):
piece.get_data.assert_any_call(lily_ind_list,
None, # {'part_names': exp_part_labels[j]},
[exp_results[i][j]])
# NB: the output_pathname is different from the previous two tests
sett_dict = {u'run_lilypond': True,
u'output_pathname': input_path + '-' + str(i) + '.ly',
u'annotation_part': [get_data_ret(0, 0, [z])[0] for z in exp_results[i]]}
piece.get_data.assert_any_call([lilypond.LilyPondIndexer], sett_dict)
def test_histogram_4(self, mock_call, mock_gdf, mock_join):
# test_output_4() but the subprocess thing fails
def raiser(*args):
raise CalledProcessError(u'Bach!', 42, u'CPE')
mock_call.side_effect = raiser
test_wc = WorkflowManager([])
test_wc._previous_exp = u'intervals'
test_wc._data = [1 for _ in xrange(20)]
test_wc._result = MagicMock(spec=pandas.DataFrame)
path = u'pathname!'
expected_msg = u'Error during call to R: CPE (return code: Bach!)'
actual = None
try:
test_wc._make_histogram(path)
except RuntimeError as run_e:
actual = run_e
self.assertEqual(expected_msg, actual.message)
self.assertEqual(0, mock_gdf.call_count)
expected_args = [u'Rscript', u'--vanilla', mock_join.return_value,
path + u'.dta', path + u'.png', u'int', u'20']
mock_call.assert_called_once_with(expected_args)
self.assertEqual(1, mock_join.call_count)
def test_lilypond_2(self, test_ip):
# make sure it works correctly with one piece that has one part
# ("voice combinations" with literal_eval())
# 1: prepare
input_path = u'carpathia'
get_data_ret = lambda *x: ['** ' + str(x[1]) if len(x) == 2 else '** ' + str(x[2][0][-3:])]
num_parts = 1 # how many parts per piece?
piece_list = ['test_piece.mei']
test_wm = WorkflowManager(piece_list)
for i in xrange(len(piece_list)):
test_wm._data[i] = mock.MagicMock(spec_set=IndexedPiece)
test_wm._data[i].get_data.side_effect = get_data_ret
# the results will be like this: [['fake result 0-0', 'fake result 0-1'],
# ['fake result 1-0', 'fake result 1-1']]
exp_results = [['fake result ' + str(i) + '-' + str(j) for j in xrange(num_parts)] \
for i in xrange(len(piece_list))]
test_wm._result = exp_results
test_wm.settings(None, 'count frequency', False)
test_wm.settings(0, 'voice combinations', '[[0]]')
#exp_part_labels = [{'part_names': [[0]]}]
# 2: run
test_wm._make_lilypond(input_path)
# 3: check
self.assertEqual(len(piece_list), test_ip.call_count) # even though we don't use them
lily_ind_list = [lilypond.AnnotationIndexer,
lilypond.AnnotateTheNoteIndexer,
lilypond.PartNotesIndexer]
for i, piece in enumerate(test_wm._data):
self.assertEqual(num_parts + 1, piece.get_data.call_count)
for j in xrange(num_parts):
piece.get_data.assert_any_call(lily_ind_list,
None, # {'part_names': exp_part_labels[i]},
[exp_results[i][j]])
sett_dict = {u'run_lilypond': True,
u'output_pathname': input_path + '.ly',
u'annotation_part': [get_data_ret(0, 0, [z])[0] for z in exp_results[i]]}
piece.get_data.assert_any_call([lilypond.LilyPondIndexer], sett_dict)
def test_export_5(self, mock_gdf):
# --> test_export_3() with a Series that requires calling _get_dataframe()
test_wm = WorkflowManager([])
test_wm._result = mock.MagicMock(spec=pandas.Series)
# CSV
mock_gdf.return_value = MagicMock(spec=pandas.DataFrame)
test_wm.export(u'CSV', u'test_path')
mock_gdf.assert_called_once_with(u'data', None, None)
mock_gdf.return_value.to_csv.assert_called_once_with(u'test_path.csv')
mock_gdf.reset_mock()
# Excel
test_wm.export(u'Excel', u'test_path', 5)
mock_gdf.assert_called_once_with(u'data', 5, None)
mock_gdf.return_value.to_excel.assert_called_once_with(u'test_path.xlsx')
mock_gdf.reset_mock()
# Stata
test_wm.export(u'Stata', u'test_path', 5, 10)
mock_gdf.assert_called_once_with(u'data', 5, 10)
mock_gdf.return_value.to_stata.assert_called_once_with(u'test_path.dta')
mock_gdf.reset_mock()
# HTML
test_wm.export(u'HTML', u'test_path', threshold=10)
mock_gdf.assert_called_once_with(u'data', None, 10)
mock_gdf.return_value.to_html.assert_called_once_with(u'test_path.html')
def test_output_3(self):
test_wc = WorkflowManager([])
test_wc._result = [5] # make sure that's not what causes it
self.assertRaises(RuntimeError, test_wc.output, u'LJKDSFLAESFLKJ')
def test_lilypond_3(self):
"""two pieces with two parts; unspecified pathname"""
mock_ips = [MagicMock(spec_set=IndexedPiece), MagicMock(spec_set=IndexedPiece)]
mock_ips[0].get_data.return_value = ['0 ready for LilyPond']
mock_ips[1].get_data.return_value = ['1 ready for LilyPond']
result = [pandas.DataFrame({('analyzer', 'clarinet'): pandas.Series(range(10)),
('analyzer', 'tuba'): pandas.Series(range(10))}),
pandas.DataFrame({('analyzer', 'flute'): pandas.Series(range(10)),
('analyzer', 'horn'): pandas.Series(range(10))})]
expected = ['test_output/output_result-0.ly', 'test_output/output_result-1.ly']
exp_get_data_calls_0 = [mock.call([lilypond_ind.AnnotationIndexer,
lilypond_exp.AnnotateTheNoteExperimenter,
lilypond_exp.PartNotesExperimenter],
{'part_names': ['analyzer: clarinet'],
'column': 'lilypond.AnnotationIndexer'},
[mock.ANY]),
mock.call([lilypond_ind.AnnotationIndexer,
lilypond_exp.AnnotateTheNoteExperimenter,
lilypond_exp.PartNotesExperimenter],
{'part_names': ['analyzer: tuba'],
'column': 'lilypond.AnnotationIndexer'},
[mock.ANY]),
mock.call([lilypond_exp.LilyPondExperimenter],
{'run_lilypond': True,
'annotation_part': ['0 ready for LilyPond',
'0 ready for LilyPond'],
'output_pathname': expected[0]})]
exp_get_data_calls_1 = [mock.call([lilypond_ind.AnnotationIndexer,
lilypond_exp.AnnotateTheNoteExperimenter,
lilypond_exp.PartNotesExperimenter],
{'part_names': ['analyzer: flute'],
'column': 'lilypond.AnnotationIndexer'},
[mock.ANY]),
mock.call([lilypond_ind.AnnotationIndexer,
lilypond_exp.AnnotateTheNoteExperimenter,
lilypond_exp.PartNotesExperimenter],
{'part_names': ['analyzer: horn'],
'column': 'lilypond.AnnotationIndexer'},
[mock.ANY]),
mock.call([lilypond_exp.LilyPondExperimenter],
{'run_lilypond': True,
'annotation_part': ['1 ready for LilyPond',
'1 ready for LilyPond'],
'output_pathname': expected[1]})]
test_wm = WorkflowManager(mock_ips)
test_wm._result = result
test_wm.settings(None, 'count frequency', False)
test_wm._make_lilypond()
self.assertEqual(3, mock_ips[0].get_data.call_count)
self.assertSequenceEqual(exp_get_data_calls_0, mock_ips[0].get_data.call_args_list)
self.assertEqual(3, mock_ips[1].get_data.call_count)
self.assertSequenceEqual(exp_get_data_calls_1, mock_ips[1].get_data.call_args_list)
# check the Series are the ones we expected---this is pretty weird and I'm sorry
self.assertEqual(result[0][('analyzer', 'clarinet')].name,
mock_ips[0].get_data.call_args_list[0][0][2][0].name)
self.assertEqual(result[0][('analyzer', 'tuba')].name,
mock_ips[0].get_data.call_args_list[1][0][2][0].name)
self.assertEqual(result[1][('analyzer', 'flute')].name,
mock_ips[1].get_data.call_args_list[0][0][2][0].name)
self.assertEqual(result[1][('analyzer', 'horn')].name,
mock_ips[1].get_data.call_args_list[1][0][2][0].name)
def test_export_1(self):
# --> raise RuntimeError with unrecognized output format
test_wm = WorkflowManager([])
test_wm._result = pandas.Series(xrange(100))
self.assertRaises(RuntimeError, test_wm.export, u'PowerPoint')
# aggregate results and count frequency, then output that
# NOTE: for this to work, I'll have to prepare ColumnAggregator and FrequencyExperimenter for vis-framework-2.0.0
#print('\nCalculating and outputting aggregated frequencies\n')
#freq_results = the_piece.get_data([aggregator.ColumnAggregator, frequency.FrequencyExperimenter],
#None,
#new_df['dissonance.SuspensionIndexer'])
#freq_results.sort(ascending=False)
#freq_results.to_excel('test_output/aggregated_results.xlsx')
## LilyPond Output! ##
# break a WorkflowManager so we can get annotated score output
print(u'\n\nPreparing and outputting the score, running LilyPond, etc.\n')
# 1.) collect indices for this part combo
#part_diss_orig = dissonances[u'0,1']
#beats_zero_orig = beat_strengths[0]
#beats_one_orig = beat_strengths[1]
# 2.) filter out where there isn't a dissonance
#susp_index = new_df['dissonance.SuspensionIndexer']['0,1']
#part_diss = part_diss_orig[~part_diss_orig.isin([None])]
#beats_zero = beats_zero_orig[~part_diss_orig.isin([None])]
#beats_one = beats_one_orig[~part_diss_orig.isin([None])]
# 3.) mangle the WorkflowManager
workm = WorkflowManager([piece_path])
workm.settings(None, 'voice combinations', '[[0, 1]]')
workm.settings(None, 'count frequency', False)
#workm._result = [new_df['dissonance.SuspensionIndexer']]
#workm._result = [new_df['dissonance.NeighbourNoteIndexer']]
#workm._result = [new_df['dissonance.PassingNoteIndexer']]
workm._result = [combined_df]
workm.output('LilyPond', 'test_output/combined_dissonances')
def test_output_4(self):
# ensure RuntimeError if self._result is None
test_wc = WorkflowManager([])
test_wc._result = None # just in case
self.assertRaises(RuntimeError, test_wc.output, u'R histogram')
