import streamlit as st
from PyPDF2 import PdfReader
from docx import Document
from io import BytesIO
from pdf2image import convert_from_bytes
import pytesseract

def pdf_to_word(pdf_file, password=None):
    """Convert a PDF file to a Word file with optional decryption and OCR support."""
    reader = PdfReader(pdf_file)

    # Decrypt the PDF if it's encrypted
    if reader.is_encrypted:
        if password:
            try:
                reader.decrypt(password)
            except Exception as e:
                raise ValueError("Failed to decrypt the PDF. Check the password.") from e
        else:
            raise ValueError("The PDF is encrypted. Please provide a password.")

    document = Document()

    # Extract text from each page
    for page in reader.pages:
        if page.extract_text():  # Use PyPDF2 for text extraction
            text = page.extract_text()
            document.add_paragraph(text)
        else:
            # Convert the page to an image and use OCR
            pdf_bytes = pdf_file.read()
            images = convert_from_bytes(pdf_bytes)
            for image in images:
                text = pytesseract.image_to_string(image)
                document.add_paragraph(text)

    word_file = BytesIO()
    document.save(word_file)
    word_file.seek(0)
    return word_file

# Streamlit app configuration
st.set_page_config(page_title="PDF to Word Converter", page_icon="🖋", layout="centered")

# App header
st.title("PDF to Word Converter")
st.write("Upload a PDF file, and we will convert it into a Word document for you.")

# File uploader
uploaded_file = st.file_uploader("Choose a PDF file", type="pdf")
password = st.text_input("Enter password (if the PDF is encrypted):", type="password")

if uploaded_file is not None:
    with st.spinner("Converting PDF to Word..."):
        try:
            word_file = pdf_to_word(uploaded_file, password)
            st.success("Conversion successful!")
            st.download_button(
                label="Download Word file",
                data=word_file,
                file_name="converted.docx",
                mime="application/vnd.openxmlformats-officedocument.wordprocessingml.document"
            )
        except ValueError as ve:
            st.error(str(ve