def test_is_stratifier(self):
channel = Channel(self.stratified_string)
self.assertTrue(channel.is_stratifier)
channel = Channel(self.undecorated_string)
self.assertFalse(channel.is_stratifier)
channel = Channel(self.arbitrary_decorator_string)
self.assertFalse(channel.is_stratifier)
def test_str_returns_decorated_string(self):
channel = Channel(self.stratified_string)
self.assertEqual(str(channel), self.stratified_string)
channel = Channel(self.undecorated_string)
self.assertEqual(str(channel), self.undecorated_string)
channel = Channel(self.arbitrary_decorator_string)
self.assertEqual(str(channel), self.arbitrary_decorator_string)
def test_decorator_detection(self):
channel = Channel(self.stratified_string)
self.assertEqual(channel.decorator, 's')
channel = Channel(self.undecorated_string)
self.assertEqual(channel.decorator, None)
channel = Channel(self.arbitrary_decorator_string)
self.assertEqual(channel.decorator, 'p')
def test_deconstruct_channel_string_regression(self):
channel_string = 'Asteroid Mining'
parts = Channel.deconstruct_channel_string(channel=channel_string)
self.assertEqual(parts['name'], channel_string)
self.assertEqual(parts['decorator'], None)
channel_string = 's:Asteroid Mining'
parts = Channel.deconstruct_channel_string(channel=channel_string)
self.assertEqual(parts['name'], channel_string.split(':')[1])
self.assertEqual(parts['decorator'], channel_string.split(':')[0])
channel_string = 'd:Asteroid Mining'
parts = Channel.deconstruct_channel_string(channel=channel_string)
self.assertEqual(parts['name'], channel_string.split(':')[1])
self.assertEqual(parts['decorator'], channel_string.split(':')[0])
def test_construct_channel_string_regression(self):
name = 'Asteroid Mining'
channel_string = Channel.construct_channel_string(name=name)
self.assertEqual(channel_string, name)
channel_string = Channel.construct_channel_string(name=name, decorator=None)
self.assertEqual(channel_string, name)
name = 'Carbonate Processing'
decorator = 's'
channel_string = Channel.construct_channel_string(name=name, decorator=decorator)
self.assertEqual(channel_string, '%s:%s' % (decorator, name))
decorator = 'h'
channel_string = Channel.construct_channel_string(name=name, decorator=decorator)
self.assertEqual(channel_string, '%s:%s' % (decorator, name))
def test_df_initialization_with_decorated_stratifiers(self):
cols = list(self.dataframe.columns)
original_cols = deepcopy(cols)
cols[0] = str(
Channel.construct_channel_string(name=cols[0], decorator='s'))
self.dataframe.columns = cols
dfw = DataFrameWrapper(dataframe=self.dataframe)
expected_stratifiers = original_cols[0:1]
self.assertEqual(list(dfw.stratifiers), expected_stratifiers)
def __init__(self, filename=None, dataframe=None, stratifiers=None):
if not ((filename is None) ^ (dataframe is None)):
raise ValueError('filename or dataframe must be provided')
if dataframe is not None:
self._dataframe = dataframe.copy()
else:
_, file_type = os.path.splitext(filename)
file_type = file_type.replace('.', '')
if file_type == self.CSV:
self._dataframe = pd.read_csv(filename)
else:
raise self.UnsupportedFileType(
'Unsupported file type for reading: %s' % file_type)
self._dataframe.reset_index(drop=True, inplace=True)
if not stratifiers:
# determine stratifying channels and remove stratifying decoration from channel names
stratifiers = []
self.data_channels = []
columns = list(self._dataframe.columns)
for i in range(0, len(columns)):
channel = Channel(columns[i])
if channel.is_stratifier:
stratifiers.append(channel.name)
columns[i] = channel.name
else:
self.data_channels.append(channel.name)
self._dataframe.columns = columns
else:
# use provided stratifier list
if not isinstance(stratifiers, list):
stratifiers = [stratifiers]
missing_stratifiers = [
s for s in stratifiers if s not in self._dataframe.columns
]
if len(missing_stratifiers) > 0:
raise self.MissingRequiredData(
'Specified stratifier(s): %s not in dataframe.' %
missing_stratifiers)
self.stratifiers = sorted(set(stratifiers))
self.data_channels = sorted(
set(self._dataframe.columns) - set(self.stratifiers))