step2 abgeschlossen

src/step2.py

@@ -2,28 +2,47 @@ import pandas as pd
 from Levenshtein import distance
 
 
-def print_status(new, sys, bool_class):
+def compare_data(new, sys, bool_class):
     print(f"\nEinträge in Import Liste: {len(new)}")
     print(f"Einträge in System Liste: {len(sys)}")
 
     if bool_class:
-        merged_df = pd.merge(new, sys, on=['name', 'vorname'], how='outer', indicator=True)
+        if 'index' in new.columns:
+            new = new.drop('index', axis=1)
+        sys = sys.drop(columns=['klasse'])
+        merged_df = pd.merge(new[['name', 'vorname', 'klasse']], sys, on=['name', 'vorname'], how='outer', indicator=True)
         matches = pd.merge(new, sys, on=['name', 'vorname'])
-        matches = matches[['name', 'vorname', 'klasse']]
+
     else:
-        merged_df = pd.merge(new, sys, on=['name', 'vorname'], how='outer', indicator=True)
+        merged_df = pd.merge(new[['name', 'vorname']], sys, on=['name', 'vorname'], how='outer', indicator=True)
         matches = pd.merge(new, sys, on=['name', 'vorname'])
-        matches = matches[['name', 'vorname']]
+        # matches = matches[['name', 'vorname', 'schuelerid']]
 
     # Subsets für Zeilen erstellen, die nur in einem der DataFrames vorhanden sind
     only_new = merged_df[merged_df['_merge'] == 'left_only'].drop(columns=['_merge'])
     only_sys = merged_df[merged_df['_merge'] == 'right_only'].drop(columns=['_merge'])
 
-    print("\nAnzahl Übereinstimmungen:", len(matches))
-    print("Anzahl neuer Nutzer:", len(only_new))
-    print("Anzahl veralteter Nutzer:", len(only_sys))
-    print(matches)
+    print_status(matches, only_new, only_sys)
+    search_typos(only_new[['name', 'vorname']], only_sys[['name', 'vorname']])
+
+    return matches, only_new
 
 
 def search_typos(new, sys):
-    matches = pd.merge(new, sys, on=['name', 'vorname'])
+    typos = []
+    col1 = 'vorname'
+    col2 = 'name'
+    for idx1, row1 in new.iterrows():
+        for idx2, row2 in sys.iterrows():
+            if distance(row1[col1], row2[col1]) <= 2 and distance(row1[col2], row2[col2]) <= 2:
+                typos.append((row1[col1], row1[col2], row2[col1], row2[col2]))
+    if len(typos) > 0:
+        print('Mögliche Tippfehler:', len(typos), '\n', typos)
+    else:
+        print('Mögliche Tippfehler: keine Fehler gefunden!')
+
+
+def print_status(matches, new, old):
+    print("\nAnzahl Übereinstimmungen:", len(matches))
+    print("Anzahl neuer Nutzer:", len(new))
+    print("Anzahl veralteter Nutzer:", len(old))