Update app.py

app.py

@@ -1,41 +1,83 @@
 import io
+import os
+import tempfile
 from numbers_parser import Document
 from openpyxl import Workbook
 import gradio as gr
 import pandas as pd
+from typing import Optional, Union
 
-def numbers_to_xlsx(numbers_file):
-  """
-  Reads a Numbers file and converts it to an xlsx file for download.
+def numbers_to_xlsx(numbers_file) -> Union[str, bytes]:
+    """
+    Efficiently converts a Numbers file to XLSX format with optimized memory usage.
 
-  Args:
-    numbers_file: The uploaded Numbers file object.
+    Args:
+        numbers_file: The uploaded Numbers file object.
 
-  Returns:
-    bytes: The contents of the converted xlsx file.
-  """
-  try:
-    # Read the Numbers file
-    doc = Document(numbers_file.name)
-    sheets = doc.sheets
-    tables = sheets[0].tables
-    data = tables[0].rows(values_only=True)
-    df = pd.DataFrame(data[1:], columns=data[0])
-    xls_path = './new.xlsx'
-    df.to_excel(xls_path, index=False)
+    Returns:
+        Union[str, bytes]: Path to the converted xlsx file, or error message if conversion fails.
+    """
+    if not numbers_file:
+        return "Please upload a Numbers file"
+    
+    # Create a temporary directory with context management
+    with tempfile.TemporaryDirectory() as temp_dir:
+        output_path = os.path.join(temp_dir, "converted.xlsx")
+        
+        try:
+            # Read the Numbers file efficiently
+            doc = Document(numbers_file.name)
+            
+            # Validate document structure
+            if not doc.sheets or not doc.sheets[0].tables:
+                return "Invalid Numbers file: No data tables found"
+            
+            # Get the first table's data efficiently
+            table = doc.sheets[0].tables[0]
+            
+            # Extract headers and data in one pass
+            rows = list(table.rows(values_only=True))
+            if not rows:
+                return "No data found in the table"
+            
+            headers = rows[0]
+            data = rows[1:]
+            
+            # Use pandas optimized DataFrame construction
+            df = pd.DataFrame(data, columns=headers)
+            
+            # Optimize Excel writing
+            with pd.ExcelWriter(
+                output_path,
+                engine='openpyxl',
+                mode='w',
+                engine_kwargs={'options': {'strings_to_urls': False}}
+            ) as writer:
+                df.to_excel(
+                    writer,
+                    index=False,
+                    sheet_name='Sheet1',
+                    freeze_panes=(1,0)  # Freeze header row
+                )
+            
+            # Read the file into memory before the temp directory is cleaned up
+            with open(output_path, 'rb') as f:
+                return f.read()
+                
+        except Exception as e:
+            return f"Error converting file: {str(e)}"
 
-    return  xls_path
-  except Exception as e:
-    return f"Error converting file: {e}"
-  
 # Define the Gradio interface
 interface = gr.Interface(
     fn=numbers_to_xlsx,
-    inputs="file",
-    outputs=gr.File(label="XLSX file"),
+    inputs=gr.File(label="Numbers File", file_types=[".numbers"]),
+    outputs=gr.File(label="XLSX file", file_types=[".xlsx"]),
     title="Numbers to XLSX Converter",
     description="Convert your Numbers files to Excel format easily and download the result.",
+    examples=None,
+    cache_examples=False
 )
 
 # Launch the Gradio app
-interface.launch(share=True)
+if __name__ == "__main__":
+    interface.launch()