Update app.py

app.py

@@ -20,108 +20,64 @@ def adjust_excel_formatting(file_path):
 
 def process_file_a_to_b(input_file):
     try:
-        # Read the Excel file, skipping the first row (OVERNIGHT/MORNING header)
         input_df = pd.read_excel(input_file.name, header=1)
-        
-        # Get the date columns (all columns except the first one which contains model names)
         date_columns = input_df.columns[1:].tolist()
-        
-        # Melt the dataframe to long format
         df_long = input_df.melt(
-            id_vars=[input_df.columns[0]],  # First column (Model names)
+            id_vars=[input_df.columns[0]],
             var_name='DATE',
             value_name='CHATTER'
         )
-        
-        # Clean up the data
-        df_long = df_long[df_long['CHATTER'].notna()]  # Remove empty cells
-        df_long = df_long[df_long['CHATTER'] != '']    # Remove empty strings
-        df_long = df_long[df_long['CHATTER'] != 'OFF'] # Remove 'OFF' entries
-        
-        # Group by chatter and date, collect all models
+        df_long = df_long[df_long['CHATTER'].notna()]
+        df_long = df_long[df_long['CHATTER'] != '']
+        df_long = df_long[df_long['CHATTER'] != 'OFF']
         grouped = df_long.groupby(['CHATTER', 'DATE'])[input_df.columns[0]].apply(
             lambda x: ', '.join(sorted(x))
         ).reset_index()
-        
-        # Pivot to get chatters as rows and dates as columns
         pivoted = grouped.pivot(
             index='CHATTER',
             columns='DATE',
             values=input_df.columns[0]
         )
-        
-        # Reorder columns to match original date order
         pivoted = pivoted[date_columns]
-        
-        # Define the expected order of chatters
-        expected_chatters = [
-            'VELJKO2', 'VELJKO3', 'MARKO', 'GODDARD', 'ALEKSANDER', 'FEELIP', 
-            'DENIS', 'TOME', 'MILA', 'VELJKO', 'DAMJAN', 'DULE', 'CONRAD', 
-            'ALEXANDER', 'VEJKO3'
-        ]
-        
-        # Reindex with expected order, keeping any additional chatters at the end
-        final_df = pivoted.reindex(expected_chatters + [x for x in pivoted.index if x not in expected_chatters])
-        
-        # Fill empty cells with 'OFF'
+        # Use the order of chatters as they appear in the input file
+        input_order = grouped['CHATTER'].drop_duplicates().tolist()
+        final_df = pivoted.reindex(input_order)
         final_df = final_df.fillna('OFF')
-        
-        # Reset index and rename the index column to 'CHATTER'
         final_df = final_df.reset_index()
         final_df = final_df.rename(columns={'index': 'CHATTER'})
-        
         return final_df
     except Exception as e:
         return pd.DataFrame({"Error": [str(e)]})
 
 def process_file_b_to_a(input_file):
     try:
-        # Read the Excel file
         input_df = pd.read_excel(input_file.name, header=0)
-        
-        # Get the date columns (all columns except the first one which contains chatter names)
         date_columns = input_df.columns[1:].tolist()
-        
-        # Melt the dataframe to long format
         df_long = input_df.melt(
-            id_vars=[input_df.columns[0]],  # First column (Chatter names)
+            id_vars=[input_df.columns[0]],
             var_name='DATE',
             value_name='MODEL'
         )
-        
-        # Clean up the data
-        df_long = df_long[df_long['MODEL'].notna()]  # Remove empty cells
-        df_long = df_long[df_long['MODEL'] != '']    # Remove empty strings
-        df_long = df_long[df_long['MODEL'] != 'OFF'] # Remove 'OFF' entries
-        
-        # Split comma-separated models into separate rows
+        df_long = df_long[df_long['MODEL'].notna()]
+        df_long = df_long[df_long['MODEL'] != '']
+        df_long = df_long[df_long['MODEL'] != 'OFF']
         df_long['MODEL'] = df_long['MODEL'].str.split(', ')
         df_long = df_long.explode('MODEL')
-        
-        # Group by model and date, collect all chatters
         grouped = df_long.groupby(['MODEL', 'DATE'])[input_df.columns[0]].apply(
             lambda x: ', '.join(sorted(x))
         ).reset_index()
-        
-        # Pivot to get models as rows and dates as columns
         pivoted = grouped.pivot(
             index='MODEL',
             columns='DATE',
             values=input_df.columns[0]
         )
-        
-        # Reorder columns to match original date order
         pivoted = pivoted[date_columns]
-        
-        # Sort models alphabetically
-        final_df = pivoted.sort_index()
-        
-        # Fill empty cells with 'OFF'
+        # Use the order of models as they appear in the input file
+        input_order = grouped['MODEL'].drop_duplicates().tolist()
+        final_df = pivoted.reindex(input_order)
         final_df = final_df.fillna('OFF')
-        
-        # Reset index to make MODEL a column
         final_df = final_df.reset_index()
-        
+        final_df = final_df.rename(columns={'index': 'MODEL'})
         return final_df
     except Exception as e:
         return pd.DataFrame({"Error": [str(e)]})