Update app.py

app.py

@@ -1,8 +1,24 @@
 import re
 import pandas as pd
+import pdfplumber
 import gradio as gr
 
 
+def extract_text_from_pdf(pdf_file):
+    """
+    Extracts text from an uploaded PDF file.
+    Args:
+        pdf_file: The uploaded PDF file.
+    Returns:
+        str: The extracted text from the PDF.
+    """
+    with pdfplumber.open(pdf_file.name) as pdf:
+        text = ""
+        for page in pdf.pages:
+            text += page.extract_text() + "\n"
+    return text
+
+
 def extract_po_data(text):
     """
     Extracts purchase order data from the text into structured rows with ITEM, DESCRIPTION, QTY, UNIT, UNIT PRICE, TOTAL PRICE.
@@ -92,35 +108,41 @@ def save_to_excel(df, output_path="extracted_po_data.xlsx"):
     return output_path
 
 
-def process_text_input(text):
+def process_pdf(file):
     """
-    Processes the raw text input, extracts data, and saves it to an Excel file.
+    Processes the uploaded PDF file, extracts data, and saves it to an Excel file.
     Args:
-        text (str): Raw text input.
+        file: The uploaded PDF file.
     Returns:
-        str: Path to the saved Excel file.
+        tuple: Path to the saved Excel file and a status message.
     """
-    df, status = extract_po_data(text)
-    if df is not None:
-        output_path = save_to_excel(df)
-        return output_path, status
-    return None, status
+    try:
+        # Extract text from the uploaded PDF
+        text = extract_text_from_pdf(file)
+        # Extract structured data
+        df, status = extract_po_data(text)
+        if df is not None:
+            output_path = save_to_excel(df)
+            return output_path, status
+        return None, status
+    except Exception as e:
+        return None, f"Error: {str(e)}"
 
 
 # Gradio Interface
 def create_interface():
     """
-    Creates a Gradio interface for processing PO data.
+    Creates a Gradio interface for processing PO data from PDF files.
     """
     interface = gr.Interface(
-        fn=process_text_input,
-        inputs=gr.Textbox(label="Paste Raw Text from PDF", lines=10, placeholder="Paste extracted text here..."),
+        fn=process_pdf,
+        inputs=gr.File(label="Upload PDF", file_types=[".pdf"]),
         outputs=[
             gr.File(label="Download Extracted Excel"),
             gr.Textbox(label="Status"),
         ],
         title="PO Data Extraction",
-        description="Paste the raw text from the PDF to extract purchase order data into an Excel file.",
+        description="Upload a purchase order PDF file to extract data into an Excel file.",
     )
     return interface
 
@@ -128,4 +150,4 @@ def create_interface():
 if __name__ == "__main__":
     # Run the Gradio app
     app = create_interface()
-    app.launch()
+    app.launch()