src/streamlit_app.py

import streamlit as st
from docx import Document
import pdfplumber
import pytesseract
from PIL import Image
import fitz  # PyMuPDF
import io

st.title("📄 Image & PDF → Word Converter")

uploaded_files = st.file_uploader(
    "Upload PDF or Image files",
    type=["pdf", "jpg", "jpeg", "png"],
    accept_multiple_files=True
)

if uploaded_files:
    for uploaded in uploaded_files:
        file_name = uploaded.name
        ext = file_name.split(".")[-1].lower()

        st.write(f"### Processing: {file_name}")

        text = ""

        # ----------------- PDF -----------------
        if ext == "pdf":
            pdf = pdfplumber.open(uploaded)
            for page in pdf.pages:
                text += page.extract_text() or ""
            pdf.close()

        # -------------- Images -----------------
        else:
            img = Image.open(uploaded)
            text = pytesseract.image_to_string(img)

        # ----------------- Create Word -----------------
        doc = Document()
        doc.add_heading(f"Converted from {file_name}", level=1)
        doc.add_paragraph(text)

        # Save to in-memory buffer
        buffer = io.BytesIO()
        doc.save(buffer)
        buffer.seek(0)

        st.download_button(
            label=f"Download Word file for {file_name}",
            data=buffer,
            file_name=file_name.replace(ext, "docx"),
            mime="application/vnd.openxmlformats-officedocument.wordprocessingml.document"
        )