app.py

import gradio as gr
from Kanun_Patrika_Scraper_For_HFSpaces import LegalCaseScraper
import os
import shutil
import zipfile

def run_scraper(mudda_type, nepali_year, progress=gr.Progress()):
    """
    Run the scraper with the given inputs and update progress.
    Returns status message and file paths for download.
    """
    try:
        # Initialize scraper
        scraper = LegalCaseScraper(output_db="legal_cases.db", html_folder="scraped_html")
        
        # Validate inputs
        if not mudda_type or not nepali_year:
            return "Error: Please select a mudda type and enter a Nepali year.", None, None
        
        # Run scraper
        progress(0.1, desc="Starting scraper...")
        scraper.run_scraper(mudda_type=mudda_type, sal=nepali_year, use_saved=True)
        
        # Check for files
        db_path = "legal_cases.db"
        html_zip_path = "scraped_html.zip"
        status = f"Scraping completed for mudda_type: {mudda_type}, year: {nepali_year}. Data saved to SQLite database.\n"
        
        # Verify database file
        if os.path.exists(db_path):
            status += f"Database file found at: {db_path}\n"
        else:
            status += "Warning: Database file (legal_cases.db) not found.\n"
        
        # Create zip of scraped_html folder
        if os.path.exists("scraped_html") and any(os.scandir("scraped_html")):
            shutil.make_archive("scraped_html", 'zip', "scraped_html")
            if os.path.exists(html_zip_path):
                status += f"HTML files zipped at: {html_zip_path}"
            else:
                status += "Warning: Failed to create scraped_html.zip."
        else:
            status += "Warning: No HTML files found in scraped_html folder."
        
        progress(1.0, desc="Scraping completed!")
        return (
            status,
            db_path if os.path.exists(db_path) else None,
            html_zip_path if os.path.exists(html_zip_path) else None
        )
    
    except Exception as e:
        return f"Error: {str(e)}", None, None
    
    finally:
        scraper.close()

# Define Gradio interface using Blocks
with gr.Blocks(title="Nepal Kanoon Patrika Scraper") as demo:
    gr.Markdown("# Nepal Kanoon Patrika Scraper")
    gr.Markdown("Scrape legal case data from Nepal Kanoon Patrika website. Select a mudda type and enter a Nepali year to begin.")
    
    with gr.Row():
        mudda_type = gr.Dropdown(
            choices=[
                "दुनियाबादी देवानी",
                "सरकारबादी देवानी",
                "दुनियावादी फौजदारी",
                "सरकारवादी फौजदारी",
                "रिट",
                "निवेदन",
                "विविध"
            ],
            label="Mudda Type",
            info="Select the type of legal case"
        )
        nepali_year = gr.Textbox(label="Nepali Year", placeholder="e.g., २०७३", max_lines=1)
    
    run_button = gr.Button("Run Scraper")
    output = gr.Textbox(label="Status", interactive=False)
    db_download = gr.File(label="Download SQLite Database")
    html_download = gr.File(label="Download Scraped HTML (Zipped)")
    
    run_button.click(
        fn=run_scraper,
        inputs=[mudda_type, nepali_year],
        outputs=[output, db_download, html_download]
    )

# Launch the interface
demo.launch()