app.py

import pdfplumber
import pandas as pd
from io import BytesIO
import re
import gradio as gr

def extract_data_from_pdf(pdf_file):
    data = []
    po_number = None

    # Save BytesIO to temporary file
    with open("temp.pdf", "wb") as f:
        f.write(pdf_file.getbuffer())

    with pdfplumber.open("temp.pdf") as pdf:
        for page in pdf.pages:
            text = page.extract_text()

            # Extract PO number
            if po_number is None:
                po_match = re.search(r"Purchase Order : (\w+)", text)
                po_number = po_match.group(1) if po_match else "N/A"

            # Regex pattern for row data
            row_pattern = re.compile(
                r"(\d+)\s+(\d+)\s+(\w+)\s+(\d{4}-\d{2}-\d{2})\s+([\d.]+)\s+([\d.]+)\s+([\d.]+)"
            )

            # Extract matching rows
            for match in row_pattern.finditer(text):
                (
                    pos,
                    item_code,
                    unit,
                    delivery_date,
                    quantity,
                    basic_price,
                    amount,
                ) = match.groups()

                sub_total_match = re.search(r"SUB TOTAL : ([\d.]+)", text)
                sub_total = sub_total_match.group(1) if sub_total_match else "0.0"

                data.append(
                    {
                        "Purchase Order": po_number,
                        "Pos.": pos,
                        "Item Code": item_code,
                        "Unit": unit,
                        "Delivery Date": delivery_date,
                        "Quantity": quantity,
                        "Basic Price": basic_price,
                        "Amount": amount,
                        "SUB TOTAL": sub_total,
                    }
                )

    # Convert data to DataFrame and save to Excel
    df = pd.DataFrame(data)
    output = BytesIO()
    with pd.ExcelWriter(output, engine="xlsxwriter") as writer:
        df.to_excel(writer, index=False, sheet_name="Extracted Data")
    output.seek(0)

    # Remove temporary PDF file
    import os
    os.remove("temp.pdf")

    return output

# Gradio Interface
iface = gr.Interface(
    fn=extract_data_from_pdf,
    inputs=gr.File(label="Upload PDF"),
    outputs=gr.File(label="Download Excel"),
    title="PDF Data Extractor",
    description="Extract structured data from a PDF and output it as an Excel file.",
)
iface.launch()