app.py

import os
import logging
from flask import Flask, render_template, request, redirect, url_for, flash, session, send_from_directory
from utility.utils import process_image_pipeline
from dotenv import load_dotenv

# Load environment variables
load_dotenv()

# Configure logging
logging.basicConfig(level=logging.INFO, handlers=[logging.StreamHandler()])

app = Flask(__name__)
app.secret_key = os.getenv('SECRET_KEY', 'default_secret_key')
app.config['UPLOAD_FOLDER'] = 'static/uploads/'
app.config['RESULT_FOLDER'] = 'static/results/'

os.makedirs(app.config['UPLOAD_FOLDER'], exist_ok=True)
os.makedirs(app.config['RESULT_FOLDER'], exist_ok=True)

@app.template_filter('basename')
def basename_filter(path):
    return os.path.basename(path)

@app.route('/')
def index():
    uploaded_files = session.get('uploaded_files', [])
    return render_template('index.html', uploaded_files=uploaded_files)

@app.route('/upload', methods=['POST'])
def upload_file():
    logging.info("Request: /upload received")
    if 'files' not in request.files:
        logging.warning("Upload: No file part in request")
        flash('No file part')
        return redirect(request.url)

    files = request.files.getlist('files')  
    if not files or all(file.filename == '' for file in files):
        logging.warning("Upload: No files selected")
        flash('No selected files')
        return redirect(request.url)

    uploaded_files = []
    for file in files:
        if file:
            filename = file.filename
            file_path = os.path.join(app.config['UPLOAD_FOLDER'], filename)
            file.save(file_path)
            uploaded_files.append(filename)
            logging.info(f"Upload: Successfully saved {filename}")

    session['uploaded_files'] = uploaded_files  
    return process_file()

@app.route('/process', methods=['GET', 'POST'])
def process_file():
    logging.info("Request: /process started")
    uploaded_files = session.get('uploaded_files', [])
    if not uploaded_files:
        logging.warning("Process: No files in session")
        flash('No files selected for processing')
        return redirect(url_for('index'))

    file_paths = [os.path.join(app.config['UPLOAD_FOLDER'], f) for f in uploaded_files]
    
    try:
        logging.info(f"Process: Sending {len(file_paths)} files to pipeline")
        processed_data = process_image_pipeline(file_paths)
        
        # Format images for result.html
        processed_Img = {f: os.path.join(app.config['UPLOAD_FOLDER'], f) for f in uploaded_files}

        session['processed_data'] = processed_data
        session['processed_Img'] = processed_Img

        logging.info("Process: Pipeline completed successfully")
        flash('Data processed successfully')
        return redirect(url_for('result'))
    except Exception as e:
        logging.exception(f"Process: Critical failure: {e}")
        flash(f'Processing error: {str(e)}')
        return redirect(url_for('index'))

@app.route('/result')
def result():
    data = session.get('processed_data', {})
    Img = session.get('processed_Img', {})
    if not data:
        return redirect(url_for('index'))
    return render_template('result.html', data=data, Img=Img)

@app.route('/reset_upload')
def reset_upload():
    uploaded_files = session.get('uploaded_files', [])
    for f in uploaded_files:
        path = os.path.join(app.config['UPLOAD_FOLDER'], f)
        if os.path.exists(path):
            os.remove(path)
    
    session.pop('uploaded_files', None)
    session.pop('processed_data', None)
    session.pop('processed_Img', None)
    flash('System reset successful.')
    return redirect(url_for('index'))

if __name__ == '__main__':
    from utility.utils import get_ocr, get_ner
    logging.info("Core: Pre-initializing engines (this may take a minute)...")
    # Trigger lazy load at startup to avoid reloader issues and request timeouts
    try:
        get_ocr()
        get_ner()
        logging.info("Core: Engines pre-initialized successfully.")
    except Exception as e:
        logging.error(f"Core: Failed to pre-initialize engines: {e}")

    app.run(debug=True, use_reloader=False, port=int(os.getenv('PORT', 5000)))