Update app.py

app.py

@@ -122,7 +122,6 @@ def extract_numeric_values(pdf_file, schedule1_file=None, client_name="Unknown C
 
 
 def save_to_csv_flat(all_extracted_values, schedule1_values, client_name="Unknown Client", csv_path=None):
-
     # Define the directory path explicitly
     if csv_path is None:
         csv_path = "./Clients_Output_Data_Form_1040.csv"
@@ -147,24 +146,16 @@ def save_to_csv_flat(all_extracted_values, schedule1_values, client_name="Unknow
         else (h1.strip() + h2.strip()) for h1, h2 in zip(header_level_1, header_level_2)
     ]
 
-    # If file doesn't exist, create new DataFrame and write headers
-    # if os.path.exists(csv_path):
-    #     df = pd.read_csv(csv_path)
-    # else:
-    #     df = pd.DataFrame(columns=flat_columns)
-
+    # Read existing CSV or create new DataFrame
     if os.path.exists(csv_path):
-        print(f"Reading existing CSV file: {csv_path}")
         df = pd.read_csv(csv_path)
-        print(f"CSV columns: {df.columns.tolist()}")
-        print(f"CSV rows before append: {len(df)}")
     else:
-        print("CSV does not exist. Creating new DataFrame.")
         df = pd.DataFrame(columns=flat_columns)
 
-    # Create new row with None
+    # Create new row
     new_row = pd.Series([None] * len(flat_columns), index=flat_columns)
     new_row.iloc[0] = client_name
+    
     # Map Page 1-2 values
     line_mapping = {
         "Taxable Wages - Line 1": 0,
@@ -193,12 +184,35 @@ def save_to_csv_flat(all_extracted_values, schedule1_values, client_name="Unknow
         new_row["Rent/ Royalty (Schedule E) - Schedule 1, Line 5"] = schedule1_values[1] if schedule1_values[1] != '' else '0'
         new_row["Other Income - Schedule 1, Line 8"] = schedule1_values[2] if schedule1_values[2] != '' else '0'
 
-    # Append and save
-    if not ((df == new_row.iloc[0]).all(axis=1)).any():   
-        df = pd.concat([df, pd.DataFrame([new_row])], ignore_index=True)
-    print(f"CSV rows after append: {len(df)}")
-    df.to_csv(csv_path, index=False)
-    print(f" Data saved to CSV: {csv_path}")
+    # Check for duplicates - improved method
+    is_duplicate = False
+    
+    # Convert new_row to DataFrame for comparison
+    new_row_df = pd.DataFrame([new_row])
+    
+    # Compare all columns except possibly the client name
+    comparison_cols = [col for col in flat_columns if col != "Client Name"]
+    
+    if not df.empty:
+        # Check if any existing row matches the new data
+        for _, existing_row in df.iterrows():
+            match = True
+            for col in comparison_cols:
+                if str(existing_row[col]) != str(new_row[col]):
+                    match = False
+                    break
+            if match:
+                is_duplicate = True
+                print("Duplicate found - not adding new row")
+                break
+
+    # Append if not duplicate
+    if not is_duplicate:
+        df = pd.concat([df, new_row_df], ignore_index=True)
+        df.to_csv(csv_path, index=False)
+        print(f"New data saved to CSV: {csv_path}")
+    else:
+        print("Duplicate data detected - no changes made to CSV")
 
 # Gradio UI
 iface = gr.Interface(