Update app.py

app.py

@@ -3,87 +3,34 @@ import pandas as pd
 import gradio as gr
 import tempfile
 import re
-import os
 
-def extract_po_to_excel(pdf_file):
-    # Regular expressions for Toshiba PO fields
-    toshiba_item_pattern = re.compile(r'Pos\.\s*Item Code\s*Unit\s*Delivery Date\s*Quantity\s*Basic Price\s*Discount\s*Cur\.\s*Amount', re.IGNORECASE)
-    toshiba_data_pattern = re.compile(r'(\d+)\s+(\d+)\s+(\w+)\s+(\d{4}-\d{2}-\d{2})\s+([\d.]+)\s+([\d.]+)\s+([\d.]+)\s+(\w+)\s+([\d.]+)')
+def extract_po_text(pdf_file):
+    # Initialize list to store text data from each page
+    text_data = []
 
-    # Regular expressions for BHEL PO fields
-    bhel_data_pattern = re.compile(r'Sl No\s+Material Description\s+Unit\s+Quantity\s+Dely Qty\s+Dely Date\s+Unit Rate\s+Value', re.IGNORECASE)
-    bhel_data_line_pattern = re.compile(r'(\d+)\s+([A-Za-z0-9 ]+)\s+(\w+)\s+(\d+)\s+(\d+)\s+(\d{2}-\d{2}-\d{4})\s+([\d.]+)\s+([\d.]+)')
-
-    # Initialize lists to store extracted data
-    toshiba_data = []
-    bhel_data = []
-
-    # Load PDF
+    # Load PDF and extract text page by page
     with fitz.open(pdf_file.name) as pdf:
         for page_num in range(pdf.page_count):
             page = pdf[page_num]
             text = page.get_text("text")
+            text_data.append(f"Page {page_num + 1}:\n{text}\n")
 
-            # Check for Toshiba PO format and extract data
-            if toshiba_item_pattern.search(text):
-                matches = toshiba_data_pattern.findall(text)
-                for match in matches:
-                    pos, item_code, unit, delivery_date, quantity, basic_price, discount, currency, amount = match
-                    toshiba_data.append({
-                        "Position": pos,
-                        "Item Code": item_code,
-                        "Unit": unit,
-                        "Delivery Date": delivery_date,
-                        "Quantity": quantity,
-                        "Basic Price": basic_price,
-                        "Discount": discount,
-                        "Currency": currency,
-                        "Amount": amount
-                    })
-
-            # Check for BHEL PO format and extract data
-            elif bhel_data_pattern.search(text):
-                matches = bhel_data_line_pattern.findall(text)
-                for match in matches:
-                    sl_no, material_desc, unit, qty, dely_qty, dely_date, unit_rate, value = match
-                    bhel_data.append({
-                        "Sl No": sl_no,
-                        "Material Description": material_desc,
-                        "Unit": unit,
-                        "Quantity": qty,
-                        "Dely Qty": dely_qty,
-                        "Dely Date": dely_date,
-                        "Unit Rate": unit_rate,
-                        "Value": value
-                    })
-
-    # Check if any data was extracted
-    if not toshiba_data and not bhel_data:
-        return None, "No data extracted from the PDF. Please check the format of the uploaded document."
-
-    # Save data to Excel sheets
-    temp_file = tempfile.NamedTemporaryFile(delete=False, suffix=".xlsx")
-    with pd.ExcelWriter(temp_file.name) as writer:
-        if toshiba_data:
-            pd.DataFrame(toshiba_data).to_excel(writer, sheet_name="Toshiba PO", index=False)
-        if bhel_data:
-            pd.DataFrame(bhel_data).to_excel(writer, sheet_name="BHEL PO", index=False)
-    
-    return temp_file.name, None
+    # Combine all page texts into one for inspection
+    full_text = "\n".join(text_data)
+    return full_text
 
 def main(pdf_file):
-    excel_file_path, error_message = extract_po_to_excel(pdf_file)
-    if error_message:
-        return None, error_message
-    return excel_file_path, "Excel file created successfully."
+    # Extract and display raw text for debugging purposes
+    extracted_text = extract_po_text(pdf_file)
+    return None, "Raw text extracted from PDF:\n\n" + extracted_text
 
-# Gradio interface with separate outputs for the file and error message
+# Gradio interface to display raw text output
 interface = gr.Interface(
     fn=main,
     inputs=gr.File(label="Upload PO PDF"),
-    outputs=[gr.File(label="Download Excel File"), gr.Textbox(label="Message")],
-    title="PO PDF to Excel Converter",
-    description="Upload a PO PDF file to extract and download it as an Excel sheet with separate sheets for Toshiba and BHEL POs."
+    outputs=[gr.File(label="Download Excel File (will not work for now)"), gr.Textbox(label="Raw Text from PDF")],
+    title="PDF Text Extractor",
+    description="Upload a PDF file to view its raw text content for troubleshooting extraction issues."
 )
 
 if __name__ == "__main__":