better arguments handling, some fixes

openlamb.py

@@ -80,6 +80,9 @@ def merge_df(dataframes, sensori):
     df = dataframes[sensori[0]]
     for sensore in sensori[1:]:
         df = pd.merge(df, dataframes[sensore])
+    if len(df) == 0:
+        print('\nERRORE: dati non disponibili per il sensore nel periodo considerato\n')
+        sys.exit(-1)
     return df
 
 
@@ -90,13 +93,13 @@ def get_dataframes(dati_csv, dati, sensori):
         if dati_csv:
             df = process(dati_csv[0], sensore, True)
             for d in dati_csv[1:]:
-                df = pd.concat([df, process(d, sensore, True)], axis=0)
+                df = pd.concat([df, process(d, sensore, True)], axis=0, ignore_index=True)
             df.rename(columns={sensore: sensore + "-csv"}, inplace=True)
             dataframes[sensore + "-csv"] = df
         if dati:
             df = process(dati[0], sensore, False)
             for d in dati[1:]:
-                df = pd.concat([df, process(d, sensore, False)], axis=0)
+                df = pd.concat([df, process(d, sensore, False)], axis=0, ignore_index=True)
             dataframes[sensore] = df
     return dataframes
 
@@ -153,6 +156,7 @@ def check_csv(args, filelist, csv_dict):
         if csv_dict[y][0] not in filelist:
             print("file %s for year %s is not available in folder %s" % (csv_dict[y][0], y, path_to_csv_files))
             download_csv(csv_dict[y][0], csv_dict[y][1], path_to_csv_files)
+        if csv_dict[y][0] not in f:
             f.append(csv_dict[y][0])
     return f
 
@@ -178,6 +182,16 @@ def download_csv(filename, id, path):
         pass
 
 
+def check_year_range(arg):
+    """check if arg is a year or a year range"""
+    if not re.search("\\d{4}-\\d{4}", arg):
+        if not re.search("\\d{4}", arg):
+            print("\nError: syntax for --csv and --dataset parameter: "
+                  "NNNN single year or NNNN-NNNN for years range\n")
+            sys.exit(-1)
+    return True
+
+
 def main():
     parser = argparse.ArgumentParser()
     parser.add_argument("--dataset", nargs='+', required=False,
@@ -193,20 +207,18 @@ def main():
         csv_files = list_of_csv_files(path_to_csv_files)
         dati_csv = []
         if args.csv:
-            if not re.search("\\d{4}-\\d{4}", args.csv[0]):
+            check_year_range(args.csv[0])
-                if not re.search("\\d{4}", args.csv[0]):
-                    print("Error: syntax for --csv parameter:  year for single year or year1-year2 for years range")
-            else:
             dati_csv = check_csv(args.csv[0], csv_files, csv_dict)
-                print("daty csv = %s" % dati_csv)
         dati = []
         if args.dataset:
             if "all" in args.dataset:
                 for k in datasets_ambiente.keys():
                     dati.append(datasets_ambiente[k])
             else:
-                for d in args.dataset:
+                check_year_range(args.dataset[0])
-                    dati.append(datasets_ambiente[d])
+                for d in parse_range(args.dataset[0]):  # args.dataset:
+                    if datasets_ambiente[str(d)] not in dati:
+                        dati.append(datasets_ambiente[str(d)])
         dataframes = get_dataframes(dati_csv, dati, args.sensori)
         datamerged = merge_df(dataframes, list(dataframes.keys()))
         datamerged.to_csv("export.csv")
@@ -217,7 +229,8 @@ def main():
             print('Valore medio per il sensore %s %s: %s' % (sensore, location, datamerged[sensore].mean().round(1)))
         plot_dataframe(datamerged)
     except KeyError:
-        print("\nKeyError: forse hai specificato un dataset che non esiste ?")
+        print("\nKeyError: forse hai specificato un dataset che non esiste ?\n"
+              "i dataset sono disponibili per gli anni %s\n " % list(datasets_ambiente.keys()))
         traceback.print_exc()
     except KeyboardInterrupt:
         print("program terminated by user")