Update app.py

app.py

@@ -3,13 +3,34 @@ import os
 import tempfile
 from numbers_parser import Document
 from openpyxl import Workbook
+from openpyxl.styles import PatternFill, Font, Alignment
 import gradio as gr
 import pandas as pd
 from pathlib import Path
 
+def get_excel_color(numbers_color):
+    """Convert Numbers color to Excel color format"""
+    if not numbers_color:
+        return None
+    # Convert RGB values to hex color
+    r = int(numbers_color.red * 255)
+    g = int(numbers_color.green * 255)
+    b = int(numbers_color.blue * 255)
+    return f"{r:02x}{g:02x}{b:02x}"
+
+def apply_cell_formatting(cell, cell_data):
+    """Apply formatting from Numbers cell to Excel cell"""
+    if hasattr(cell_data, 'background_color') and cell_data.background_color:
+        color = get_excel_color(cell_data.background_color)
+        if color:
+            cell.fill = PatternFill(start_color=color, end_color=color, fill_type="solid")
+    
+    if hasattr(cell_data, 'font_bold'):
+        cell.font = Font(bold=cell_data.font_bold)
+
 def numbers_to_xlsx(numbers_file):
     """
-    Efficiently converts a Numbers file to XLSX format with optimized memory usage.
+    Converts a Numbers file to XLSX format preserving colors and formatting.
 
     Args:
         numbers_file: The uploaded Numbers file object from Gradio.
@@ -21,12 +42,11 @@ def numbers_to_xlsx(numbers_file):
         return None
     
     try:
-        # Create output directory if it doesn't exist
         output_dir = "outputs"
         os.makedirs(output_dir, exist_ok=True)
         output_path = os.path.join(output_dir, "converted.xlsx")
         
-        # Read the Numbers file efficiently using the temporary path
+        # Read the Numbers file
         doc = Document(numbers_file.name)
         
         # Get all sheets and validate
@@ -35,39 +55,52 @@ def numbers_to_xlsx(numbers_file):
             print("No sheets found in the document")
             return None
             
-        # Process each sheet
-        writer = pd.ExcelWriter(output_path, engine='openpyxl')
+        # Create a new workbook
+        wb = Workbook()
+        wb.remove(wb.active)  # Remove default sheet
         
         for sheet_index, sheet in enumerate(sheets):
             if not sheet.tables:
                 continue
                 
-            # Get the first table in the sheet
+            # Create new sheet
+            ws = wb.create_sheet(title=f"Sheet{sheet_index + 1}")
             table = sheet.tables[0]
             
-            # Extract data
-            rows = list(table.rows(values_only=True))
+            # Get all rows with their formatting
+            rows = list(table.rows())
             if not rows:
                 continue
-                
-            # Split headers and data
-            headers = rows[0] if rows else []
-            data = rows[1:] if len(rows) > 1 else []
             
-            # Create DataFrame and write to Excel
-            df = pd.DataFrame(data, columns=headers)
-            sheet_name = f"Sheet{sheet_index + 1}"
+            # Write data and apply formatting
+            for row_idx, row in enumerate(rows, 1):
+                for col_idx, cell_data in enumerate(row, 1):
+                    # Write cell value
+                    cell = ws.cell(row=row_idx, column=col_idx)
+                    cell.value = cell_data.value if hasattr(cell_data, 'value') else cell_data
+                    
+                    # Apply formatting if available
+                    if hasattr(cell_data, 'background_color') or hasattr(cell_data, 'font_bold'):
+                        apply_cell_formatting(cell, cell_data)
             
-            df.to_excel(
-                writer,
-                sheet_name=sheet_name,
-                index=False
-            )
+            # Adjust column widths
+            for column in ws.columns:
+                max_length = 0
+                column_letter = column[0].column_letter
+                for cell in column:
+                    try:
+                        if len(str(cell.value)) > max_length:
+                            max_length = len(str(cell.value))
+                    except:
+                        pass
+                adjusted_width = (max_length + 2)
+                ws.column_dimensions[column_letter].width = adjusted_width
             
             # Freeze the header row
-            writer.sheets[sheet_name].freeze_panes = 'A2'
+            ws.freeze_panes = 'A2'
         
-        writer.close()
+        # Save the workbook
+        wb.save(output_path)
         
         # Verify the file was created and has content
         if os.path.exists(output_path) and os.path.getsize(output_path) > 0: