def test_attributes_flat_dict(self):
from collections.abc import MutableMapping
# code to convert ini_dict to flattened dictionary
# default seperater '_'
def convert_flatten(d, parent_key='', sep='_'):
items = []
for k, v in d.items():
new_key = parent_key + sep + k if parent_key else k
if isinstance(v, MutableMapping):
items.extend(convert_flatten(v, new_key, sep=sep).items())
else:
items.append((new_key, v))
return dict(items)
shape = (10, 10, 15)
data = np.random.normal(size=shape)
h5_file = h5py.File('test2.h5', 'w')
h5_group = h5_file.create_group('MyGroup')
data_set = sidpy.Dataset.from_array(data[:, :, :], title='Image')
attributes = {'Key1': 10, 'Key2': np.linspace(0, 1, 10), 'NestedDict': {'KeyLevel2_0': 10,
'KeyLevel2_1': np.linspace(0, 1, 10)}}
flattened_attributes = convert_flatten(attributes)
# pass data with flattened dictionary, make sure it doesn't complain
hdf_io.write_results(h5_group, dataset=data_set, attributes=flattened_attributes, process_name='TestProcess')
h5_file.close()
remove('test2.h5')
def test_no_sidpy_dataset_provided(self):
h5_file = h5py.File('test2.h5', 'w')
h5_group = h5_file.create_group('MyGroup')
# going to pass a numpy array instead of a sidpy dataset
with self.assertRaises(ValueError):
hdf_io.write_results(h5_group, dataset=None, attributes=None, process_name='TestProcess')
h5_file.close()
remove('test2.h5')
def test_not_h5py_group_obj(self):
#Set h5_group to be a list instead of an actual hdf5 group object
h5_group = [10,115]
shape = (5, 15, 16)
data = np.random.randn(shape[0], shape[1], shape[2])
data_set = sidpy.Dataset.from_array(data[:, :, :], title='Image')
with self.assertRaises(TypeError):
hdf_io.write_results(h5_group, dataset=data_set, attributes=None, process_name='TestProcess')
def test_process_name_no_name_clashes(self):
shape = (10, 10, 15)
data = np.random.normal(size=shape)
h5_file = h5py.File('test2.h5', 'w')
h5_group = h5_file.create_group('MyGroup')
data_set = sidpy.Dataset.from_array(data[:, :, :], title='Image')
# pass data with process_name being something other than a string
hdf_io.write_results(h5_group, dataset=data_set, attributes=None, process_name='This should work')
h5_file.close()
remove('test2.h5')
def test_not_a_sidpy_Dataset(self):
shape = (10, 10, 15)
data = np.random.normal(size=shape)
h5_file = h5py.File('test2.h5', 'w')
h5_group = h5_file.create_group('MyGroup')
#should fail if standard numpy array is passed
with self.assertRaises(ValueError):
hdf_io.write_results(h5_group, dataset=data, attributes=None, process_name='TestProcess')
h5_file.close()
remove('test2.h5')
def test_simple(self):
h5_f = h5py.File('test_write_results.h5', 'w')
h5_group = h5_f.create_group('MyGroup')
shape = (5, 15, 16)
data = np.random.randn(shape[0], shape[1], shape[2])
data_set = sidpy.Dataset.from_array(data[:, :, :], title='Image')
hdf_io.write_results(h5_group, dataset=data_set, attributes=None, process_name='TestProcess')
# TODO: Add some more assertions
h5_f.close()
remove('test_write_results.h5')
def test_no_attributes_provided(self):
shape = (10, 10, 15)
data = np.random.normal(size=shape)
h5_file = h5py.File('test2.h5', 'w')
h5_group = h5_file.create_group('MyGroup')
data_set = sidpy.Dataset.from_array(data[:, :, :], title='Image')
# pass data without attributes
hdf_io.write_results(h5_group, dataset=data_set, attributes=None, process_name='TestProcess')
h5_file.close()
remove('test2.h5')
def test_group_already_contains_objects_name_clashes(self):
# Set h5_group to be a list instead of an actual hdf5 group object
h5_file = h5py.File('test2.h5', 'w')
h5_group = h5_file.create_group('MyGroup')
shape = (5, 15, 16)
data = np.random.randn(shape[0], shape[1], shape[2])
data_set1 = sidpy.Dataset.from_array(data[:, :, :], title='Image')
data_set2 = sidpy.Dataset.from_array(data[:, :, :], title='Image')
hdf_io.write_results(h5_group, dataset=data_set1, attributes=None, process_name='TestProcess')
# If we try to write data_set2 which has the same name, it probably should fail??
# with self.assertRaises(ValueError):
hdf_io.write_results(h5_group, dataset=data_set2, attributes=None, process_name='TestProcess')
def test_attributes_not_dict(self):
from collections import namedtuple
shape = (10, 10, 15)
data = np.random.normal(size=shape)
h5_file = h5py.File('test2.h5', 'w')
h5_group = h5_file.create_group('MyGroup')
data_set = sidpy.Dataset.from_array(data[:, :, :], title='Image')
attributes = namedtuple('Dimensions', ['x', 'y'])
# pass data with attributes being something other than dictionary
with self.assertRaises(TypeError):
hdf_io.write_results(h5_group, dataset=data_set, attributes=attributes, process_name='TestProcess')
h5_file.close()
remove('test2.h5')
def test_multiple_sidpy_datasets_as_results(self):
shape = (10, 10, 15)
data = np.random.normal(size=shape)
data2 = np.random.normal(size=shape)
h5_file = h5py.File('test2.h5', 'w')
h5_group = h5_file.create_group('MyGroup')
data_set = sidpy.Dataset.from_array(data[:, :, :], title='Image')
data_set2 = sidpy.Dataset.from_array(data2[:, :, :], title='Image2')
results = [data_set, data_set2]
hdf_io.write_results(h5_group, dataset=results, attributes=None, process_name='This should work')
h5_file.close()
remove('test2.h5')
def test_attributes_nested_dict(self):
shape = (10, 10, 15)
data = np.random.normal(size=shape)
h5_file = h5py.File('test2.h5', 'w')
h5_group = h5_file.create_group('MyGroup')
data_set = sidpy.Dataset.from_array(data[:, :, :], title='Image')
attributes = {'Key1':10, 'Key2': np.linspace(0,1,10), 'NestedDict': {'KeyLevel2_0':10,
'KeyLevel2_1': np.linspace(0,1,10)}}
# pass data with nested dictionary, make sure it doesn't complain
hdf_io.write_results(h5_group, dataset=data_set, attributes=attributes, process_name='TestProcess')
h5_file.close()
remove('test2.h5')
