def test_EntityKey_no_name():
type_ = 'population'
measure = 'structure'
key = hdf.EntityKey(f'{type_}.{measure}')
assert key.type == type_
assert key.name == ''
assert key.measure == measure
assert key.group_prefix == '/'
assert key.group_name == type_
assert key.group == f'/{type_}'
assert key.path == f'/{type_}/{measure}'
assert key.with_measure('age_groups') == hdf.EntityKey(
'population.age_groups')
def test_EntityKey_with_name():
type_ = 'cause'
name = 'diarrheal_diseases'
measure = 'incidence'
key = hdf.EntityKey(f'{type_}.{name}.{measure}')
assert key.type == type_
assert key.name == name
assert key.measure == measure
assert key.group_prefix == f'/{type_}'
assert key.group_name == name
assert key.group == f'/{type_}/{name}'
assert key.path == f'/{type_}/{name}/{measure}'
assert key.with_measure('prevalence') == hdf.EntityKey(
f'{type_}.{name}.prevalence')
def test_get_valid_filter_terms_all_valid(hdf_key, hdf_file):
node = hdf_file.get_node(hdf.EntityKey(hdf_key).path)
if not isinstance(node, tables.earray.EArray):
columns = node.table.colnames
valid_filter_terms = _construct_all_valid_filters(columns)
assert set(hdf._get_valid_filter_terms(
valid_filter_terms, columns)) == set(valid_filter_terms)
def test_write_empty_data_frame(hdf_file_path):
key = hdf.EntityKey('cause.test.prevalence')
data = pd.DataFrame(columns=('age', 'year', 'sex', 'draw', 'location',
'value'))
with pytest.raises(ValueError):
hdf._write_pandas_data(hdf_file_path, key, data)
def test_load(hdf_file_path, hdf_key):
key = hdf.EntityKey(hdf_key)
data = hdf.load(hdf_file_path, key, filter_terms=None, column_filters=None)
if 'restrictions' in key or 'versions' in key:
assert isinstance(data, dict)
elif 'metadata' in key:
assert isinstance(data, list)
else:
assert isinstance(data, pd.DataFrame)
def test_touch_existing_file(tmpdir):
path = f'{str(tmpdir)}/test.hdf'
hdf.touch(path)
hdf.write(path, hdf.EntityKey('test.key'), 'data')
assert hdf.get_keys(path) == ['test.key']
# should wipe out and make it again
hdf.touch(path)
assert hdf.get_keys(path) == []
def test_write_empty_data_frame_index(hdf_file_path):
key = hdf.EntityKey('cause.test.prevalence')
data = pd.DataFrame(data={
'age': range(10),
'year': range(10),
'draw': range(10)
})
data = data.set_index(list(data.columns))
hdf._write_pandas_data(hdf_file_path, key, data)
written_data = pd.read_hdf(hdf_file_path, key.path)
written_data = written_data.set_index(
list(written_data)) # write resets index. only calling load undoes it
assert written_data.equals(data)
def test_load_with_valid_filters(hdf_file_path, hdf_key):
key = hdf.EntityKey(hdf_key)
data = hdf.load(hdf_file_path,
key,
filter_terms=["year == 2006"],
column_filters=None)
if 'restrictions' in key or 'versions' in key:
assert isinstance(data, dict)
elif 'metadata' in key:
assert isinstance(data, list)
else:
assert isinstance(data, pd.DataFrame)
if 'year' in data.columns:
assert set(data.year) == {2006}
def test_write_load_empty_data_frame_index(hdf_file_path):
key = hdf.EntityKey('cause.test.prevalence')
data = pd.DataFrame(data={
'age': range(10),
'year': range(10),
'draw': range(10)
})
data = data.set_index(list(data.columns))
hdf._write_pandas_data(hdf_file_path, key, data)
loaded_data = hdf.load(hdf_file_path,
key,
filter_terms=None,
column_filters=None)
assert loaded_data.equals(data)
def test_write_data_frame(hdf_file_path):
key = hdf.EntityKey('cause.test.prevalence')
data = build_table(
[lambda *args, **kwargs: random.choice([0, 1]), "Kenya", 1],
2005,
2010,
columns=('age', 'year', 'sex', 'draw', 'location', 'value'))
non_val_columns = data.columns.difference({'value'})
data = data.set_index(list(non_val_columns))
hdf._write_pandas_data(hdf_file_path, key, data)
written_data = pd.read_hdf(hdf_file_path, key.path)
assert written_data.equals(data)
filter_terms = ['draw == 0']
written_data = pd.read_hdf(hdf_file_path, key.path, where=filter_terms)
assert written_data.equals(data.xs(0, level='draw', drop_level=False))
def mock_key(request):
return hdf.EntityKey(request.param)
def test_EntityKey_init_failure():
bad_keys = ['hello', 'a.b.c.d', '', '.', '.coconut', 'a.', 'a..c']
for k in bad_keys:
with pytest.raises(ValueError):
hdf.EntityKey(k)
def test_get_node_name(hdf_file, hdf_key):
key = hdf.EntityKey(hdf_key)
assert hdf._get_node_name(hdf_file.get_node(key.path)) == key.measure
def test_remove(hdf_file_path, hdf_key):
key = hdf.EntityKey(hdf_key)
hdf.remove(hdf_file_path, key)
with tables.open_file(str(hdf_file_path)) as file:
assert key.path not in file
