Refactored the 'check_data' function in 'input_checks.py'. It now returns a data frame.

toarstats/input_checks.py

 """Functions to check the given user inputs.
 
-This module contains the following function:
+This module contains the following functions:
 check_sampling - check the given sampling argument
+check_statistics - check the given statistics argument
+check_data - check the given data argument
 """
 
 import numpy as np
@@ -56,46 +58,46 @@ def check_statistics(statistics_in):
     return statistics_out
 
 
-def check_data(input_argument, alt_input_idx, alt_input_vals):
+def check_data(data_in, datetimes_in, values_in):
     """Check the given data argument.
 
-    :param input_argument: the given argument
-    :param alt_input_idx: the alternative input's index
-    :param alt_input_vals: the alternative input's values
+    :param data_in: the given data argument
+    :param datetimes_in: the given datetimes argument
+    :param values_in: the given values argument
 
-    :return: The processed arguments or None if there was a problem with
-             the input
+    :return: The processed data argument or None if there was a problem
+             while processing the given data argument
     """
-    if ((input_argument is not None and not (alt_input_idx is None
-                                             and alt_input_vals is None))
-            or (input_argument is None and (alt_input_idx is None
-                                            or alt_input_vals is None))):
-        return None, None
-    if isinstance(input_argument, (pd.DataFrame, pd.Series)):
-        index = input_argument.index
-        if isinstance(input_argument, pd.Series):
-            param_vals = input_argument.to_numpy()
-        elif "value" in input_argument.columns.to_numpy():
-            param_vals = input_argument["value"].to_numpy()
-        elif "values" in input_argument.columns.to_numpy():
-            param_vals = input_argument["values"].to_numpy()
-        elif input_argument.shape[1] > 0:
-            param_vals = input_argument.iloc[:, 0].to_numpy()
+    if (data_in is not None
+            and not (datetimes_in is None and values_in is None)):
+        return None
+    if data_in is None and (datetimes_in is None or values_in is None):
+        return None
+    if isinstance(data_in, (pd.DataFrame, pd.Series)):
+        index_out = data_in.index
+        if isinstance(data_in, pd.Series):
+            values_out = data_in.to_numpy()
+        elif "value" in data_in.columns.to_numpy():
+            values_out = data_in["value"].to_numpy()
+        elif "values" in data_in.columns.to_numpy():
+            values_out = data_in["values"].to_numpy()
+        elif data_in.shape[1] > 0:
+            values_out = data_in.iloc[:, 0].to_numpy()
         else:
-            return None, None
+            values_out = np.array([])
     else:
-        index = alt_input_idx
-        param_vals = alt_input_vals
-    if not isinstance(index, pd.DatetimeIndex):
-        return None, None
-    if not isinstance(param_vals, np.ndarray) or param_vals.ndim != 1:
-        return None, None
-    if (index.dropna().empty or all(np.isnan(param_vals))
-            or index.size != param_vals.size):
-        return None, None
-    if not (np.issubdtype(param_vals.dtype, np.floating)
-            or np.issubdtype(param_vals.dtype, np.integer)):
-        return None, None
-    if index.tz:
-        index = index.tz_localize(None)
-    return index, param_vals
+        index_out = datetimes_in
+        values_out = values_in
+    if not isinstance(index_out, pd.DatetimeIndex):
+        return None
+    if not isinstance(values_out, np.ndarray) or values_out.ndim != 1:
+        return None
+    if not (np.issubdtype(values_out.dtype, np.floating)
+            or np.issubdtype(values_out.dtype, np.integer)):
+        return None
+    if (index_out.dropna().empty or all(np.isnan(values_out))
+            or index_out.size != values_out.size):
+        return None
+    if index_out.tz:
+        index_out = index_out.tz_localize(None)
+    return pd.DataFrame({"values": values_out}, index=index_out)

toarstats/interface.py

@@ -92,4 +92,4 @@ def calculate_statistics(sampling, statistics, data=None, metadata=None,
     """
     sampling_method = check_sampling(sampling)
     stats_list = check_statistics(statistics)
-    datetime_index, parameter_values = check_data(data, datetimes, values)
+    data_frame = check_data(data, datetimes, values)