app.py

from pywebio import start_server
from pywebio.input import input
from pywebio.output import (
    put_text, put_success, put_error, put_file,
    put_markdown, put_buttons
)

import requests, os, shutil, uuid, subprocess, threading, time, re
from urllib.parse import urlparse, unquote

# ============ AUTO PINGER =============
PING_URL = "https://sanch1tx-onedrive.hf.space"
def auto_pinger():
    while True:
        try:
            requests.get(PING_URL, timeout=10)
        except:
            pass
        time.sleep(300)

threading.Thread(target=auto_pinger, daemon=True).start()

# ============ FILENAME EXTRACTION ============
def extract_filename(response, fallback_url):
    cd = response.headers.get("content-disposition")
    if cd:
        m = re.search(r'filename="?([^"]+)"?', cd)
        if m:
            return m.group(1)
    path = urlparse(fallback_url).path
    name = os.path.basename(path)
    return unquote(name) if name else "document.docx"

# ============ DOWNLOAD FROM ONEDRIVE ============
def download_file_from_onedrive(shared_url, workdir):
    dl = shared_url + ("&download=1" if "?" in shared_url else "?download=1")
    r = requests.get(dl, stream=True)
    r.raise_for_status()
    filename = extract_filename(r, shared_url)
    if not filename.lower().endswith(".docx"):
        filename += ".docx"
    path = os.path.join(workdir, filename)
    with open(path, "wb") as f:
        for chunk in r.iter_content(8192):
            if chunk:
                f.write(chunk)
    return path, filename

# ============ DOCX → PDF USING LIBREOFFICE CLI ============
def convert_docx_to_pdf(docx_path, workdir):
    cmd = [
        "soffice",
        "--headless",
        "--convert-to", "pdf:writer_pdf_Export",
        docx_path,
        "--outdir", workdir
    ]
    subprocess.run(cmd, check=True)
    name_pdf = os.path.basename(docx_path).replace(".docx", ".pdf")
    return os.path.join(workdir, name_pdf), name_pdf

# ============ PYWEBIO APP ============
def app():
    put_markdown("## OneDrive DOCX → PDF Converter (LibreOffice)")

    url = input("Paste OneDrive DOCX link")
    workdir = f"temp_{uuid.uuid4()}"
    os.makedirs(workdir, exist_ok=True)

    try:
        put_text("Downloading file…")
        docx_path, docx_name = download_file_from_onedrive(url, workdir)
        put_success(f"Downloaded: {docx_name}")

        put_text("Converting to PDF…")
        pdf_path, pdf_name = convert_docx_to_pdf(docx_path, workdir)

        def download_click(btn_val):
            with open(pdf_path, "rb") as f:
                put_file(pdf_name, f.read())

        put_buttons(["Download PDF"], onclick=download_click)
        put_success("PDF ready! Click the button above.")

    except Exception as e:
        put_error(str(e))

    finally:
        # Clean up after giving user time to download
        def cleanup():
            time.sleep(30)
            shutil.rmtree(workdir, ignore_errors=True)
        threading.Thread(target=cleanup, daemon=True).start()

    put_markdown("---")
    put_markdown(
        "<center><small>Developed by: <b>Sanchit Vishwakarma</b></small></center>",
        sanitize=False
    )

if __name__ == "__main__":
    start_server(app, host="0.0.0.0", port=7860)