Python: add + processing for columns in dataplot

Author: rmcdermo

Utilities/Python/fdsplotlib.py

@@ -67,8 +67,16 @@ def dataplot(config_filename,**kwargs):
         plot_range_in = kwargs.get('plot_range')
         plot_range = range(plot_range_in[0]-2,plot_range_in[-1]-1)
 
+    # Rebuild the default NA strings *excluding* 'N/A'
+    default_na = {
+        '', '#N/A', '#N/A N/A', '#NA', '-1.#IND', '-1.#QNAN',
+        '1.#IND', '1.#QNAN', '<NA>', 'NA', 'NULL', 'NaN',
+        'n/a', 'nan', 'null'
+    }
+    safe_na_values = default_na  # 'N/A' intentionally not included
+
     # read the config file
-    df = pd.read_csv(configdir+config_filename, sep=',', engine='python', comment='#', quotechar='"')
+    df = pd.read_csv(configdir+config_filename, sep=',', engine='python', comment='#', quotechar='"', na_values=safe_na_values, keep_default_na=False)
     C = df.where(pd.notnull(df), None)
 
     Plot_Filename_Last = None
@@ -103,12 +111,17 @@ def dataplot(config_filename,**kwargs):
             # read data from exp file
             # set header to the row where column names are stored (Python is 0 based)
             E = pd.read_csv(expdir+pp.d1_Filename, header=int(pp.d1_Col_Name_Row-1), sep=',', engine='python', comment='#', quotechar='"')
+            E.columns = E.columns.str.strip()  # <-- Strip whitespace from headers
 
             x = E[pp.d1_Ind_Col_Name].values[:].astype(float)
             # y = E[pp.d1_Dep_Col_Name].values[:].astype(float)
             col_names = [c.strip() for c in pp.d1_Dep_Col_Name.split('|')]
-            # print(col_names)
-            y = E[col_names].values.astype(float)
+            # y = E[col_names].values.astype(float)
+            y = np.column_stack([
+                E[cols].astype(float).sum(axis=1) if '+' in name else E[[name]].astype(float).values.ravel()
+                for name in col_names
+                for cols in [name.split('+')]
+            ])
 
             styles = [c.strip() for c in pp.d1_Style.split('|')]
             key_labels = [c.strip() for c in pp.d1_Key.split('|')]
@@ -147,6 +160,7 @@ def dataplot(config_filename,**kwargs):
                 # read data from exp file
                 # set header to the row where column names are stored (Python is 0 based)
                 E = pd.read_csv(expdir+pp.d1_Filename, header=int(pp.d1_Col_Name_Row-1), sep=',', engine='python', comment='#', quotechar='"')
+                E.columns = E.columns.str.strip()  # <-- Strip whitespace from headers
                 x = E[pp.d1_Ind_Col_Name].values[:].astype(float)
                 y = E[pp.d1_Dep_Col_Name].values[:].astype(float)
 
@@ -164,10 +178,16 @@ def dataplot(config_filename,**kwargs):
 
         # get the model results
         M = pd.read_csv(cmpdir+pp.d2_Filename, header=int(pp.d2_Col_Name_Row-1), sep=',', engine='python', comment='#', quotechar='"')
+        M.columns = M.columns.str.strip()  # <-- Strip whitespace from headers
         x = M[pp.d2_Ind_Col_Name].values[:].astype(float)
         # y = M[pp.d2_Dep_Col_Name].values[:].astype(float)
         col_names = [c.strip() for c in pp.d2_Dep_Col_Name.split('|')]
-        y = M[col_names].values.astype(float)
+        # y = M[col_names].values.astype(float)
+        y = np.column_stack([
+            M[cols].astype(float).sum(axis=1) if '+' in name else M[[name]].astype(float).values.ravel()
+            for name in col_names
+            for cols in [name.split('+')]
+        ])
 
         version_string = revision
         if (pp.VerStr_Filename):