def test_compare_data_frames(self):
"""Test a function to compare data frames."""
df1 = load_expected_case(1)[0]
df2 = load_expected_case(1)[0]
df3 = load_expected_case(2)[0]
self.assertTrue(compare_data_frames(df1, df2))
with self.assertRaises(AssertionError):
compare_data_frames(df1, df3)
def run_case_without_some_parameters(self,
case_num,
specific_params=None,
special_case=None):
"""Run a test case deleting some parameters.
Args:
case_num: The test case number to run.
"""
test_wb = get_orig_cases_path(case_num)
params = load_parameters_case(case_num)
exp_dfs = load_expected_case(case_num, special_case)
params.remove_non_critical()
if specific_params:
for specific_param, value in specific_params.items():
params[specific_param] = value
# change safe_mode to True, for complete test in safe_mode (very slow)
safe_mode = False
series = XlSeries(test_wb)
test_dfs = series.get_data_frames(params, safe_mode=safe_mode)
# get them always into a list
if type(test_dfs) != list:
test_dfs = [test_dfs]
if type(exp_dfs) != list:
exp_dfs = [exp_dfs]
for test_df, exp_df in zip(test_dfs, exp_dfs):
self.assertTrue(compare_data_frames(test_df, exp_df))
def run_case_with_parameters(self,
case_num,
specific_params=None,
special_case=None):
"""Run a test case with parameters using ParameterDiscovery strategy.
Args:
case_num (int): The test case number to run.
"""
test_wb = load_original_case(case_num)
params = load_parameters_case(case_num)
params["data_ends"] = None
exp_dfs = load_expected_case(case_num, special_case)
if specific_params:
for specific_param, value in specific_params.items():
params[specific_param] = value
# get dfs from the strategy
series = XlSeries(test_wb)
test_dfs = series.get_data_frames(params)
if type(test_dfs) != list:
test_dfs = [test_dfs]
if type(exp_dfs) != list:
exp_dfs = [exp_dfs]
for test_df, exp_df in zip(test_dfs, exp_dfs):
print(test_df.columns, exp_df.columns)
self.assertTrue(compare_data_frames(test_df, exp_df))
def _get_data_frames(cls, ws, params, safe_mode):
"""Extract time data series and return them as data frames."""
# FIRST: discover missing parameters generating attempts
attempts = cls._discover_parameters(ws, params)
# there is only one attempt, probably the user passed all the params
if len(attempts) == 1:
params = attempts[0]
# SECOND: clean the data
cls._clean_data(ws, params)
# THIRD: get the data from a cleaned worksheet
dfs = cls._get_data(ws, params)
return (dfs, params)
# there is multiple combinations of parameters to try
else:
results = []
for params_attempt in attempts:
ws_temp = make_ws_copy(ws)
try:
# SECOND: clean the data
cls._clean_data(ws_temp, params_attempt)
# THIRD: get the data from a cleaned worksheet
dfs = cls._get_data(ws_temp, params_attempt)
# don't return a list with only one element
if type(dfs) == list and len(dfs) == 1:
dfs = dfs[0]
if (type(params_attempt) == list
and len(params_attempt) == 1):
params_attempt = params_attempt[0]
results.append((dfs, params_attempt))
# stops with the first successful result
if not safe_mode:
break
except:
continue
# remove duplicates
unique_results = []
for res in results:
repeated = False # first result will not be repeated!
for unique_res in unique_results:
repeated = True
for df_a, df_b in zip(res[0], unique_res[0]):
try:
compare_data_frames(df_a, df_b)
except AssertionError:
repeated = False
if repeated:
break
# if True or not repeated:
if not repeated:
unique_results.append(res)
# return results
if len(unique_results) == 0:
raise Exception("""
File couldn't be parsed with provided parameters:
{}
Last attempt was:
{}
""".format(repr(params), repr(params_attempt)))
elif len(unique_results) == 1:
return unique_results[0]
else:
print("There is more than one result with given parameters.")
dfs = [res[0] for res in unique_results]
params = [res[1] for res in unique_results]
return (dfs, params_attempt)