def test_hdf5_array_repository_remove_records(array_repository_w_saved_data):
_, repo, _, ids = array_repository_w_saved_data
prev_indices = {}
for index in repo.indices:
prev_indices[index] = repo.get_index(index)
removed_ids = ranges.Key([ids[0]])
repo.remove_records(removed_ids)
removed_locations = repo.find(removed_ids)
locations = repo.find(ids)
found_ids = []
for path, indices in locations.items():
with hdf5.HDF5File() as file:
dataset = file[path]
for _id, idx in dataset.attrs.items():
if idx in indices:
found_ids.append(_id)
assert repo.get_num_records() == len(ids) - 1
assert len(removed_locations) == 0
assert len(found_ids) == len(ids) - 1
assert sorted(found_ids) != ids
for index in repo.indices:
assert prev_indices[index] != repo.get_index(index)
def test_hdf5_file_context_manager():
with hdf5.HDF5File() as file:
assert os.path.exists(config.HDF5_FILEPATH)
assert file
assert len(list(file.keys())) == 0
assert not file
def test_hdf5_array_repository_get_datasets(array_repository_w_saved_data):
_, repo, datasets, ids = array_repository_w_saved_data
with hdf5.HDF5File() as file:
found_datasets = recursive_search(file[repo.GROUP], repo.GROUP)
remove_meta_data_datasets(repo, found_datasets)
assert repo.get_datasets() == found_datasets
def test_hdf5_file_create_group():
group_name = 'test'
with hdf5.HDF5File() as file:
group1 = file.create_group(group_name)
group2 = file.create_group(group_name)
assert isinstance(group1, h5py.Group)
assert isinstance(group2, h5py.Group)
assert group1.name == group2.name
def test_hdf5_file_regular_open():
file = hdf5.HDF5File()
assert os.path.exists(config.HDF5_FILEPATH)
assert file
assert len(list(file.keys())) == 0
file.close()
assert not file
def test_hd5f_array_repository_save(array_repository_w_saved_data):
arrays, repo, datasets, _ = array_repository_w_saved_data
num_records = len(arrays.reg_combos) + len(arrays.cap_combos)
# verify datasets have been made properly
with hdf5.HDF5File() as file:
found_datasets = recursive_search(file[repo.GROUP], repo.GROUP)
assert found_datasets == datasets
for dataset in datasets:
if repo.INDEX_DATASET not in dataset:
num_records -= len(file[repo.GROUP][dataset])
assert num_records == 0
def test_hdf5_array_repository_find(array_repository_w_saved_data):
_, repo, _, ids = array_repository_w_saved_data
locations = repo.find(ids)
found_ids = []
for path, indices in locations.items():
with hdf5.HDF5File() as file:
dataset = file[path]
for _id, idx in dataset.attrs.items():
if idx in indices:
found_ids.append(_id)
assert sorted(found_ids) == ids
def test_hdf5_array_repository_remove_dataset(array_repository_w_saved_data):
_, repo, datasets, ids = array_repository_w_saved_data
prev_indices = {}
for index in repo.indices:
prev_indices[index] = repo.get_index(index)
repo.remove_dataset(datasets[0])
with hdf5.HDF5File() as file:
group = file[repo.GROUP]
found_datasets = sorted(list(group.keys()))
assert datasets != found_datasets
for index in repo.indices:
assert prev_indices[index] != repo.get_index(index)
def test_hdf5_obj_repository_save(obj_repository_w_saved_data):
dicts, repo, datasets, _ = obj_repository_w_saved_data
# verify datasets have been made properly
with hdf5.HDF5File() as file:
found_datasets = sorted(recursive_search(file[repo.GROUP], repo.GROUP))
assert found_datasets == datasets
assert len(file[repo.GROUP][datasets[0]]) == len(dicts)
# verify indices have been made properly
for index in repo.indices:
stored_index = repo.get_index(index)
for doc in dicts:
attr_val = doc[index]
assert attr_val in stored_index
stored_index.pop(attr_val)