def test_dc_replacements(self):
'''make sure we can read in various scenarios in dc_replacements
using a test version of the replacement rules, and ensure any
invalid format of rule is not included.
'''
curr_dc = DataClean(replacements=self.tst_replacements)
print('-------------------------')
for k, v in curr_dc.replacements.items():
print('%s=%s' % (k, v))
print('--------------------------------------------------------')
print('testing for %s, to be replaced with %s' % (k, v))
test_string = 'recording_7.wav %s Wow that is awesome' % k
print('...test string >>%s' % test_string)
#set contents prior to processing,typically users would not do this
#would set contents at instance. doing this just for testing
curr_dc.contents = test_string
curr_dc.process_contents(show_results=False)
print('...processed as>>%s' % curr_dc.contents.decode())
self.assertEqual(
-1,
curr_dc.contents.decode().find(k),
'...seems the original token is still in the string')
self.assertGreater(0,
curr_dc.contents.decode().find(k),
'...seems like the new token was not applied')
def test_dc_process_contents_clean_find_n_replace_symbols(self):
'''ensure that we can replace at the multiple locations of symbols'''
curr_replacements = self.dc.replacements
for k, v in curr_replacements.items():
if k == 'SYMBOL_AS_SPACE':
ignore_sym = ['[', ']', ',']
for each in v:
if each not in ignore_sym:
print(
'--------------------------------------------------------'
)
print(
'testing for symbol "%s", to be removed (i.e. replaced with '
'' % each)
test_string = 'recording_7.wav %s Wow %s%s%s that is awesome %s' % (
each, each, each, each, each)
print('...test string >>%s' % test_string)
tst_dc = DataClean(contents=test_string)
tst_dc.process_contents(show_results=False)
print('...processed as>>%s' % tst_dc.contents.decode())
self.assertEqual(
-1,
tst_dc.contents.decode().find(k),
'...seems the original token is still in the string'
)
self.assertGreater(
0,
tst_dc.contents.decode().find(k),
'...seems like the new token was not applied')
def test_dc_process_search_non_speech_events_multiple_instances_both_over(self):
'''ensure we can locate multiple non speech events, one instance is bad'''
test_limit_1='[%s]'%('a'*21)
test_limit_2='[%s]'%('b'*21)
test_string = 'recording_7.wav %sWow that is %s awesome' % (test_limit_1,test_limit_2)
print('...test string >>%s' % test_string)
tst_dc = DataClean(contents=test_string)
self.assertTrue(tst_dc._search_non_speech_events(),'...expected to be under the limit, confirm')
def test_dc_process_search_non_speech_events_over_limit(self):
'''ensure that OVER the limit will still fail (as True)'''
test_limit='[%s]'%('a'*30)
test_string = 'recording_7.wav %sWow that is awesome' % test_limit
print('...test string >>%s' % test_string)
tst_dc = DataClean(contents=test_string)
self.assertTrue(tst_dc._search_non_speech_events(),'...expected to be over the limit of 20 chars, confirm')
def test_dc_process_search_non_speech_events_at_limit(self):
'''ensure that one the limit will still pass (as False)'''
test_limit='[%s]'%('a'*20)
test_string = 'recording_7.wav %sWow that is awesome' % test_limit
print('...test string >>%s' % test_string)
tst_dc = DataClean(contents=test_string)
self.assertFalse(tst_dc._search_non_speech_events(),'...expected at limit of 20 chars, confirm')
def test_dc_process_search_non_speech_events_nochars(self):
'''make sure we can find the non speech with no chars event and ensure it's under the limit'''
test_limit='[]'
test_string = 'recording_7.wav %sWow that is awesome' % test_limit
print('...test string >>%s' % test_string)
tst_dc = DataClean(contents=test_string)
self.assertFalse(tst_dc._search_non_speech_events(),'...expected to within limit, confirm')
def setUp(self):
self.filename = os.path.join('datasources','metadata_for_audio.txt')
self.project_path = os.path.dirname(os.path.abspath(__file__))
self.full_filename = os.path.join(self.project_path, self.filename)
self.replacements_txt = os.path.join('datasources', 'data_clean_replacements_test.txt')
self.tst_replacements = os.path.join(self.project_path, self.replacements_txt)
self.dc = DataClean()
self.s = swat.CAS()
def test_dc_datasource_data_to_castable(self):
'''process everything and build a castable from results'''
curr_dc = DataClean(conn=self.conn,
contents_as_path=self.full_filename)
response = curr_dc.process_contents(show_results=False)
curr_dc.create_castable(response['results'],
'cool_cas',
replace=True,
promote=True)
def test_dc_process_consecspaces_multiple_locations(self):
'''ensure we can replace consecutive spaces for processing 2 locations, one at end'''
test_limit_1='%s'%(' '*1)
test_limit_2='%s'%(' '*2)
test_string = 'recording_7.wav %sWow that is %s awesome%s' % (test_limit_1,test_limit_2,test_limit_2)
print('...test string >>%s' % test_string)
tst_dc = DataClean(contents=test_string)
status,updated_string = tst_dc._search_replace()
print(updated_string)
self.assertLess(len(updated_string), len(test_string), '...expected string to less due to replacements')
self.assertTrue(status,'...expected to be under the limit, confirm')
def test_dc_process_consecspaces_beginning_one_single_and_double_single(self):
'''ensure we can replace consecutive spaces for processing, second one should be replaced'''
test_limit_1='>>%s<<'%(' '*1)
test_limit_2='>>%s<<'%(' '*2)
test_string = 'recording_7.wav %sWow that is %s awesome' % (test_limit_1,test_limit_2)
print('...test string >>%s' % test_string)
tst_dc = DataClean(contents=test_string)
status,updated_string = tst_dc._search_replace()
print(updated_string)
self.assertLess(len(updated_string), len(test_string), '...expected string to less due to replacements')
self.assertTrue(status,'...expected to be under the limit, confirm')
def test_dc_process_consecspaces_beginning_both_single(self):
'''ensure we can replace consecutive spaces for processing, none should return false'''
test_limit_1='>>%s<<'%(' '*1)
test_limit_2='>>%s<<'%(' '*1)
test_string = 'recording_7.wav %sWow that is %s awesome' % (test_limit_1,test_limit_2)
print('...test string >>%s' % test_string)
tst_dc = DataClean(contents=test_string)
status,updated_string = tst_dc._search_replace()
print(updated_string)
self.assertEqual(len(updated_string), len(test_string), '...expected string to equal since no replacement')
self.assertFalse(status,'...nothing should have been replaced based on default regex')
def test_dc_process_consecspaces_beginning_both_multiple(self):
'''ensure we can replace consecutive spaces for processing, both need to be replaced'''
test_limit_1='>>%s<<'%(' '*30)
test_limit_2='>>%s<<'%(' '*25)
test_string = 'recording_7.wav %sWow that is %s awesome' % (test_limit_1,test_limit_2)
tst_dc = DataClean(contents=test_string)
status,updated_string = tst_dc._search_replace()
print('...replace status: %s' % status)
print('...test string >>:%s' % test_string)
print('...updated string >>:%s' % updated_string)
self.assertLess(len(updated_string), len(test_string), '...expected string to less due to replacements')
self.assertTrue(status,'...expected to be under the limit, confirm')
def test_dc_process_contents_clean_find_n_replace_end_of_string(self):
'''ensure that we can replace at the end of the string'''
curr_replacements = self.dc.replacements
for k,v in curr_replacements.items():
if k != 'SYMBOL_AS_SPACE':
print('--------------------------------------------------------')
print('testing for %s, to be replaced with %s' % (k,v))
test_string = 'recording_7.wav Wow that is awesome %s' % k
print('...test string >>%s' % test_string)
tst_dc = DataClean(contents=test_string)
tst_dc.process_contents(show_results=False)
print('...processed as>>%s' % tst_dc.contents.decode())
self.assertEqual(-1,tst_dc.contents.decode().find(k),'...seems the original token is still in the string')
self.assertGreater(0, tst_dc.contents.decode().find(k),'...seems like the new token was not applied')
def test_dc_datasource_data(self):
'''just process the test example and ensure everything works'''
print ('---------------------------------------')
print ('Metadata Prior to Cleaning')
print ('---------------------------------------')
# curr_dc=DataClean()
# pre_data = curr_dc.readin_file_as_bytearray(self.full_filename)
# for each in pre_data.decode():
# print (each)
curr_dc=DataClean(contents_as_path=self.full_filename).process_contents(show_results=True)
class TestDataCleaning(tm.TestCase):
def setUp(self):
self.filename = os.path.join('datasources','metadata_for_audio.txt')
self.project_path = os.path.dirname(os.path.abspath(__file__))
self.full_filename = os.path.join(self.project_path, self.filename)
self.replacements_txt = os.path.join('datasources', 'data_clean_replacements_test.txt')
self.tst_replacements = os.path.join(self.project_path, self.replacements_txt)
self.dc = DataClean()
self.s = swat.CAS()
def tearDown(self):
# tear down tests
try:
self.s.terminate()
except swat.SWATError:
pass
del self.s
swat.reset_option()
def test_dc_valid_file(self):
'''read in a valid text file from datasource'''
self.dc.readin_file_as_bytearray(self.full_filename)
def test_dc_replacements(self):
'''make sure we can read in various scenarios in dc_replacements
using a test version of the replacement rules, and ensure any
invalid format of rule is not included.
'''
curr_dc=DataClean(replacements=self.tst_replacements)
print ('-------------------------')
for k,v in curr_dc.replacements.items():
print('%s=%s' % (k, v))
print('--------------------------------------------------------')
print('testing for %s, to be replaced with %s' % (k,v))
test_string = 'recording_7.wav %s Wow that is awesome' % k
print('...test string >>%s' % test_string)
#set contents prior to processing,typically users would not do this
#would set contents at instance. doing this just for testing
curr_dc.contents=test_string
curr_dc.process_contents(show_results=False)
print('...processed as>>%s' % curr_dc.contents.decode())
self.assertEqual(-1,curr_dc.contents.decode().find(k),'...seems the original token is still in the string')
self.assertGreater(0, curr_dc.contents.decode().find(k),'...seems like the new token was not applied')
def test_dc_load_replacemetns(self):
self.dc.load_dc_replacements()
def test_dc_process_contents_clean_find_n_replace_start_of_string(self):
'''ensure that we can replace at the start of the string'''
curr_replacements = self.dc.replacements
for k,v in curr_replacements.items():
if k != 'SYMBOL_AS_SPACE':
print('--------------------------------------------------------')
print('testing for %s, to be replaced with %s' % (k,v))
test_string = 'recording_7.wav %s Wow that is awesome' % k
print('...test string >>%s' % test_string)
tst_dc = DataClean(contents=test_string)
tst_dc.process_contents(show_results=False)
print('...processed as>>%s' % tst_dc.contents.decode())
self.assertEqual(-1,tst_dc.contents.decode().find(k),'...seems the original token is still in the string')
self.assertGreater(0, tst_dc.contents.decode().find(k),'...seems like the new token was not applied')
def test_dc_process_contents_clean_find_n_replace_end_of_string(self):
'''ensure that we can replace at the end of the string'''
curr_replacements = self.dc.replacements
for k,v in curr_replacements.items():
if k != 'SYMBOL_AS_SPACE':
print('--------------------------------------------------------')
print('testing for %s, to be replaced with %s' % (k,v))
test_string = 'recording_7.wav Wow that is awesome %s' % k
print('...test string >>%s' % test_string)
tst_dc = DataClean(contents=test_string)
tst_dc.process_contents(show_results=False)
print('...processed as>>%s' % tst_dc.contents.decode())
self.assertEqual(-1,tst_dc.contents.decode().find(k),'...seems the original token is still in the string')
self.assertGreater(0, tst_dc.contents.decode().find(k),'...seems like the new token was not applied')
def test_dc_process_contents_clean_find_n_replace_middle_of_string(self):
'''ensure that we can replace at the middle of the string'''
curr_replacements = self.dc.replacements
for k,v in curr_replacements.items():
if k != 'SYMBOL_AS_SPACE':
print('--------------------------------------------------------')
print('testing for %s, to be replaced with %s' % (k,v))
test_string = 'recording_7.wav Wow that is %s awesome' % k
print('...test string >>%s' % test_string)
tst_dc = DataClean(contents=test_string)
tst_dc.process_contents(show_results=False)
print('...processed as>>%s' % tst_dc.contents.decode())
self.assertEqual(-1,tst_dc.contents.decode().find(k),'...seems the original token is still in the string')
self.assertGreater(0, tst_dc.contents.decode().find(k),'...seems like the new token was not applied')
def test_dc_process_contents_clean_find_n_replace_multiple_instances_in_string(self):
'''ensure that we can replace at the multiple locations within the string'''
curr_replacements = self.dc.replacements
for k,v in curr_replacements.items():
if k != 'SYMBOL_AS_SPACE':
print('--------------------------------------------------------')
print('testing for %s, to be replaced with %s' % (k,v))
test_string = 'recording_7.wav %s Wow %s that is %s awesome %s' % (k,k,k,k)
print('...test string >>%s' % test_string)
tst_dc = DataClean(contents=test_string)
tst_dc.process_contents(show_results=False)
print('...processed as>>%s' % tst_dc.contents.decode())
self.assertEqual(-1,tst_dc.contents.decode().find(k),'...seems the original token is still in the string')
self.assertGreater(0, tst_dc.contents.decode().find(k),'...seems like the new token was not applied')
def test_dc_process_contents_clean_find_n_replace_symbols(self):
'''ensure that we can replace at the multiple locations of symbols'''
curr_replacements = self.dc.replacements
for k,v in curr_replacements.items():
if k == 'SYMBOL_AS_SPACE':
ignore_sym=['[',']',',']
for each in v:
if each not in ignore_sym:
print('--------------------------------------------------------')
print('testing for symbol "%s", to be removed (i.e. replaced with ''' % each)
test_string = 'recording_7.wav %s Wow %s%s%s that is awesome %s' % (each,each,each,each,each)
print('...test string >>%s' % test_string)
tst_dc = DataClean(contents=test_string)
tst_dc.process_contents(show_results=False)
print('...processed as>>%s' % tst_dc.contents.decode())
self.assertEqual(-1,tst_dc.contents.decode().find(k),'...seems the original token is still in the string')
self.assertGreater(0, tst_dc.contents.decode().find(k),'...seems like the new token was not applied')
def test_dc_process_search_non_speech_events_singlechar(self):
'''make sure we can find the non speech event and ensure it's under the limit'''
test_limit='[%s]'%('a'*1)
test_string = 'recording_7.wav %sWow that is awesome' % test_limit
print('...test string >>%s' % test_string)
tst_dc = DataClean(contents=test_string)
self.assertFalse(tst_dc._search_non_speech_events(),'...expected to within limit, confirm')
def test_dc_process_search_non_speech_events_nochars(self):
'''make sure we can find the non speech with no chars event and ensure it's under the limit'''
test_limit='[]'
test_string = 'recording_7.wav %sWow that is awesome' % test_limit
print('...test string >>%s' % test_string)
tst_dc = DataClean(contents=test_string)
self.assertFalse(tst_dc._search_non_speech_events(),'...expected to within limit, confirm')
def test_dc_process_search_non_speech_events_one_less_limit(self):
'''ensure that one less the limit will still pass (as False)'''
test_limit='[%s]'%('a'*19)
test_string = 'recording_7.wav %sWow that is awesome' % test_limit
print('...test string >>%s' % test_string)
tst_dc = DataClean(contents=test_string)
self.assertFalse(tst_dc._search_non_speech_events(),'...expected to within limit, confirm')
def test_dc_process_search_non_speech_events_at_limit(self):
'''ensure that one the limit will still pass (as False)'''
test_limit='[%s]'%('a'*20)
test_string = 'recording_7.wav %sWow that is awesome' % test_limit
print('...test string >>%s' % test_string)
tst_dc = DataClean(contents=test_string)
self.assertFalse(tst_dc._search_non_speech_events(),'...expected at limit of 20 chars, confirm')
def test_dc_process_search_non_speech_events_over_limit(self):
'''ensure that OVER the limit will still fail (as True)'''
test_limit='[%s]'%('a'*30)
test_string = 'recording_7.wav %sWow that is awesome' % test_limit
print('...test string >>%s' % test_string)
tst_dc = DataClean(contents=test_string)
self.assertTrue(tst_dc._search_non_speech_events(),'...expected to be over the limit of 20 chars, confirm')
def test_dc_process_search_non_speech_events_multiple_instances_both_okay(self):
'''ensure we can locate multiple non speech events'''
test_limit_1='[%s]'%('a'*10)
test_limit_2='[%s]'%('b'*15)
test_string = 'recording_7.wav %sWow that is %s awesome' % (test_limit_1,test_limit_2)
print('...test string >>%s' % test_string)
tst_dc = DataClean(contents=test_string)
self.assertFalse(tst_dc._search_non_speech_events(),'...expected to be under the limit, confirm')
def test_dc_process_search_non_speech_events_multiple_instances_2nd_okay(self):
'''ensure we can locate multiple non speech events, one instance is bad'''
test_limit_1='[%s]'%('a'*30)
test_limit_2='[%s]'%('b'*15)
test_string = 'recording_7.wav %sWow that is %s awesome' % (test_limit_1,test_limit_2)
print('...test string >>%s' % test_string)
tst_dc = DataClean(contents=test_string)
self.assertTrue(tst_dc._search_non_speech_events(),'...expected to be under the limit, confirm')
def test_dc_process_search_non_speech_events_multiple_instances_1st_okay(self):
'''ensure we can locate multiple non speech events, one instance is bad'''
test_limit_1='[%s]'%('a'*10)
test_limit_2='[%s]'%('b'*40)
test_string = 'recording_7.wav %sWow that is %s awesome' % (test_limit_1,test_limit_2)
print('...test string >>%s' % test_string)
tst_dc = DataClean(contents=test_string)
self.assertTrue(tst_dc._search_non_speech_events(),'...expected to be under the limit, confirm')
def test_dc_process_search_non_speech_events_multiple_instances_both_over(self):
'''ensure we can locate multiple non speech events, one instance is bad'''
test_limit_1='[%s]'%('a'*21)
test_limit_2='[%s]'%('b'*21)
test_string = 'recording_7.wav %sWow that is %s awesome' % (test_limit_1,test_limit_2)
print('...test string >>%s' % test_string)
tst_dc = DataClean(contents=test_string)
self.assertTrue(tst_dc._search_non_speech_events(),'...expected to be under the limit, confirm')
def test_dc_process_consecspaces_beginning_one_double_and_one_single(self):
'''ensure we can replace consecutive spaces for processing, 1 needs it 1 doesn't'''
test_limit_1='>>%s<<'%(' '*2)
test_limit_2='>>%s<<'%(' '*1)
test_string = 'recording_7.wav %sWow that is %s awesome' % (test_limit_1,test_limit_2)
print('...test string >>%s' % test_string)
tst_dc = DataClean(contents=test_string)
status,updated_string = tst_dc._search_replace()
print(updated_string)
self.assertLess(len(updated_string), len(test_string), '...expected string to less due to replacements')
self.assertTrue(status,'...expected to be under the limit, confirm')
def test_dc_process_consecspaces_beginning_both_multiple(self):
'''ensure we can replace consecutive spaces for processing, both need to be replaced'''
test_limit_1='>>%s<<'%(' '*30)
test_limit_2='>>%s<<'%(' '*25)
test_string = 'recording_7.wav %sWow that is %s awesome' % (test_limit_1,test_limit_2)
tst_dc = DataClean(contents=test_string)
status,updated_string = tst_dc._search_replace()
print('...replace status: %s' % status)
print('...test string >>:%s' % test_string)
print('...updated string >>:%s' % updated_string)
self.assertLess(len(updated_string), len(test_string), '...expected string to less due to replacements')
self.assertTrue(status,'...expected to be under the limit, confirm')
def test_dc_process_consecspaces_beginning_both_single(self):
'''ensure we can replace consecutive spaces for processing, none should return false'''
test_limit_1='>>%s<<'%(' '*1)
test_limit_2='>>%s<<'%(' '*1)
test_string = 'recording_7.wav %sWow that is %s awesome' % (test_limit_1,test_limit_2)
print('...test string >>%s' % test_string)
tst_dc = DataClean(contents=test_string)
status,updated_string = tst_dc._search_replace()
print(updated_string)
self.assertEqual(len(updated_string), len(test_string), '...expected string to equal since no replacement')
self.assertFalse(status,'...nothing should have been replaced based on default regex')
def test_dc_process_consecspaces_beginning_one_single_and_double_single(self):
'''ensure we can replace consecutive spaces for processing, second one should be replaced'''
test_limit_1='>>%s<<'%(' '*1)
test_limit_2='>>%s<<'%(' '*2)
test_string = 'recording_7.wav %sWow that is %s awesome' % (test_limit_1,test_limit_2)
print('...test string >>%s' % test_string)
tst_dc = DataClean(contents=test_string)
status,updated_string = tst_dc._search_replace()
print(updated_string)
self.assertLess(len(updated_string), len(test_string), '...expected string to less due to replacements')
self.assertTrue(status,'...expected to be under the limit, confirm')
def test_dc_process_consecspaces_multiple_locations(self):
'''ensure we can replace consecutive spaces for processing 2 locations, one at end'''
test_limit_1='%s'%(' '*1)
test_limit_2='%s'%(' '*2)
test_string = 'recording_7.wav %sWow that is %s awesome%s' % (test_limit_1,test_limit_2,test_limit_2)
print('...test string >>%s' % test_string)
tst_dc = DataClean(contents=test_string)
status,updated_string = tst_dc._search_replace()
print(updated_string)
self.assertLess(len(updated_string), len(test_string), '...expected string to less due to replacements')
self.assertTrue(status,'...expected to be under the limit, confirm')
def test_dc_datasource_data(self):
'''just process the test example and ensure everything works'''
print ('---------------------------------------')
print ('Metadata Prior to Cleaning')
print ('---------------------------------------')
# curr_dc=DataClean()
# pre_data = curr_dc.readin_file_as_bytearray(self.full_filename)
# for each in pre_data.decode():
# print (each)
curr_dc=DataClean(contents_as_path=self.full_filename).process_contents(show_results=True)
def test_dc_datasource_data_to_castable(self):
'''process everything and build a castable from results'''
curr_dc=DataClean(conn=self.s,contents_as_path=self.full_filename)
response = curr_dc.process_contents(show_results=False)
curr_dc.create_castable(response['results'],'cool_cas',replace=True,promote=True)
