import streamlit as st
import subprocess
import json
import os
import uuid

st.title("Advanced Scrapy Scraper")

url = st.text_input("Start URL", "https://example.com")
run = st.button("Run Scrape")

if run and url:
    output_file = f"output_{uuid.uuid4().hex}.json"  # unique file per run

    with st.spinner("Scraping..."):
        cmd = [
            "scrapy", "crawl", "advanced",
            "-a", f"start_url={url}",
            "-o", output_file,
            "--nolog"
        ]

        result = subprocess.run(
            cmd,
            cwd=".",
            capture_output=True,
            text=True
        )

    # Check if Scrapy failed
    if result.returncode != 0:
        st.error("Scrapy failed")
        st.text(result.stderr)
    else:
        if os.path.exists(output_file):
            try:
                with open(output_file, "r") as f:
                    data = json.load(f)

                st.success("Scrape complete")
                st.json(data)

                st.download_button(
                    "Download JSON",
                    json.dumps(data, indent=2),
                    "results.json",
                    "application/json"
                )

            except json.JSONDecodeError:
                st.error("Output file is not valid JSON")

        else:
            st.error("No output file generated")