def test_add_column_from_element_to_elements():
net = nw.create_cigre_network_mv()
pp.create_measurement(net, "i", "trafo", 5, 3, 0, side="hv")
pp.create_measurement(net, "i", "line", 5, 3, 0, side="to")
pp.create_measurement(net, "p", "bus", 5, 3, 2)
assert net.measurement.name.isnull().all()
assert ~net.switch.name.isnull().all()
orig_switch_names = copy.deepcopy(net.switch.name.values)
expected_measurement_names = np.array(
[net.trafo.name.loc[0], net.line.name.loc[0], net.bus.name.loc[2]])
expected_switch_names = np.append(
net.line.name.loc[net.switch.element.loc[net.switch.et == "l"]].values,
net.trafo.name.loc[net.switch.element.loc[net.switch.et ==
"t"]].values)
pp.add_column_from_element_to_elements(net, "name", False)
assert all(
pp.compare_arrays(net.measurement.name.values,
expected_measurement_names))
assert all(pp.compare_arrays(net.switch.name.values, orig_switch_names))
del net.measurement["name"]
pp.add_column_from_element_to_elements(net, "name", True)
assert all(
pp.compare_arrays(net.measurement.name.values,
expected_measurement_names))
assert all(pp.compare_arrays(net.switch.name.values,
expected_switch_names))
def check_subnet_correctness(net, elements, branch_bus):
for elm in elements:
if "bus" in net[elm].columns:
assert all(pp.compare_arrays(net[elm]["subnet"].values,
np.array(["subnet_%i" % bus for bus in net[elm].bus])))
elif branch_bus[0] in net[elm].columns:
assert all(pp.compare_arrays(net[elm]["subnet"].values, np.array([
"subnet_%i" % bus for bus in net[elm][branch_bus[0]]])))
elif branch_bus[1] in net[elm].columns:
assert all(pp.compare_arrays(net[elm]["subnet"].values, np.array([
"subnet_%i" % bus for bus in net[elm][branch_bus[1]]])))
def test_column_indices():
df = pd.DataFrame([[0, 3, 2, 4, 5, 6]],
columns=["b", "q", "a", "g", "f", "c"])
query_cols = ["b", "g", "f", "c", "q", "a", "q", "a", "f"]
col_idx = sb.column_indices(df, query_cols)
assert all(
pp.compare_arrays(col_idx, np.array([0, 3, 4, 5, 1, 2, 1, 2, 4])))
def _get_unique_duplicated_dict(df, subset=None):
""" Returns a dict which keys are the indices of unique row of the dataframe 'df'. The values
of the dict are the indices which are duplicated to each key index. """
subset = subset or df.columns
dupl = df.index[df.duplicated(subset=subset)]
uniq = df.index[~df.duplicated(subset=subset)]
uniq_dupl_dict = {}
# nan_str only needed since compare_arrays() using old numpy versions connected to python 3.4
# don't detect reliably nans as equal
nan_str = "nan"
while nan_str in df.values:
nan_str += "n"
for uni in uniq:
do_dupl_fit = compare_arrays(
np.repeat(df.loc[uni,
subset].fillna(nan_str).values.reshape(1, -1),
len(dupl),
axis=0),
df.loc[dupl, subset].fillna(nan_str).values).all(axis=1)
uniq_dupl_dict[uni] = list(dupl[do_dupl_fit])
return uniq_dupl_dict