app.py

# libraries
from flask import Flask, render_template, request, redirect, url_for, flash, session, send_from_directory
import os
import logging
from utility.utils import extract_text_from_images, process_extracted_text, process_resume_data
from backup.backup import NER_Model
from paddleocr import PaddleOCR

# Configure logging
logging.basicConfig(
    level=logging.INFO,
    handlers=[
        logging.StreamHandler()  # Remove FileHandler and log only to the console
    ]
)

# Flask App
app = Flask(__name__)
app.secret_key = 'your_secret_key'

@app.template_filter('basename')
def basename_filter(path):
    return os.path.basename(path)
app.config['UPLOAD_FOLDER'] = 'uploads/'
app.config['RESULT_FOLDER'] = 'results/'

UPLOAD_FOLDER = 'static/uploads/'
RESULT_FOLDER = 'static/results/'
os.makedirs(UPLOAD_FOLDER, exist_ok=True)
os.makedirs(RESULT_FOLDER, exist_ok=True)

if not os.path.exists(app.config['UPLOAD_FOLDER']):
    os.makedirs(app.config['UPLOAD_FOLDER'], exist_ok=True)

if not os.path.exists(app.config['RESULT_FOLDER']):
    os.makedirs(app.config['RESULT_FOLDER'], exist_ok=True)


# Set the PaddleOCR home directory to a writable location
os.environ['PADDLEOCR_HOME'] = '/tmp/.paddleocr' 

# Check if PaddleOCR home directory is writable
if not os.path.exists('/tmp/.paddleocr'):
    os.makedirs('/tmp/.paddleocr', exist_ok=True)
    logging.info("Created PaddleOCR home directory.")
else:
    logging.info("PaddleOCR home directory exists.")

@app.route('/')
def index():
    uploaded_files = session.get('uploaded_files', [])
    logging.info(f"Accessed index page, uploaded files: {uploaded_files}")
    return render_template('index.html', uploaded_files=uploaded_files)

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

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

    uploaded_files = session.get('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)
            print(f"file path --->{file_path}")
            uploaded_files.append(filename)
            logging.info(f"Uploaded file: {filename} at {file_path}")

    session['uploaded_files'] = uploaded_files  
    flash('Files successfully uploaded')
    logging.info(f"Files successfully uploaded: {uploaded_files}")
    return process_file()

@app.route('/reset_upload')
def reset_upload():
    """Reset the uploaded files and processed data in the session and filesystem."""
    uploaded_files = session.get('uploaded_files', [])
    for filename in uploaded_files:
        file_path = os.path.join(app.config['UPLOAD_FOLDER'], filename)
        if os.path.exists(file_path):
            os.remove(file_path)
            logging.info(f"Removed file: {filename}")
        else:
            logging.warning(f"File not found for removal: {file_path}")

    session.pop('uploaded_files', None)
    session.pop('processed_data', None)
    session.pop('processed_Img', None)
    flash('Reset successful. All files removed.')
    logging.info("Session and upload folder reset.")
    return redirect(url_for('index'))

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

    file_paths = [os.path.join(app.config['UPLOAD_FOLDER'], filename) for filename in uploaded_files]
    logging.info(f"Processing files: {file_paths}")

    extracted_text = {}
    LLMdata = {}
    processed_Img = {}

    try:
        # Primary: Groq VLM Single Pass
        LLMdata, extracted_text, processed_Img = extract_text_from_images(file_paths)
        LLMdata['meta'] = "Primary: Groq VLM"

        # Check if extracted data is essentially empty
        is_empty = all(len(v) == 0 for k, v in LLMdata.items() if k != 'meta')
        
        if is_empty:
            logging.info("Groq VLM returned empty data. Attempting backup...")
            raise ValueError("Empty data from Groq VLM")

    except Exception as e:
        logging.exception(f"Primary processing failed or returned empty: {e}")
        flash('Primary processing failed, using backup model...')
        
        # If extraction failed but we have some text/images, use them
        if not extracted_text:
             flash('Critical failure: Could not extract text from images.')
             return redirect(url_for('index'))
        
        try:
            # Backup: NER Model on extracted text
            text_for_ner = ""
            for path, text in extracted_text.items():
                text_for_ner += str(text) + " "
            
            LLMdata = NER_Model(text_for_ner)
            LLMdata['meta'] = "Backup: Local NER"
            logging.info(f"NER model data: {LLMdata}")
        except Exception as backup_e:
            logging.exception(f"Backup processing failed: {backup_e}")
            flash('Processing failed completely.')
            return redirect(url_for('index'))

    # Common merge and finalization step
    try:
        cont_data = process_extracted_text(extracted_text)
        processed_data = process_resume_data(LLMdata, cont_data, extracted_text)
        
        session['processed_data'] = processed_data
        session['processed_Img'] = processed_Img
        
        logging.info(f"Final processed data: {processed_data}")
        return redirect(url_for('result'))
    except Exception as merge_e:
        logging.exception(f"Error during data merging: {merge_e}")
        flash('Error finalizing data extraction.')
        return redirect(url_for('index'))

    
@app.route('/result')
def result():
    processed_data = session.get('processed_data', {})
    processed_Img = session.get('processed_Img', {})
    logging.info(f"Displaying results: Data - {processed_data}, Images - {processed_Img}")
    return render_template('result.html', data=processed_data, Img=processed_Img)

@app.route('/uploads/<filename>')
def uploaded_file(filename):
    logging.info(f"Serving file: {filename}")
    return send_from_directory(app.config['UPLOAD_FOLDER'], filename)

if __name__ == '__main__':
    logging.info("Starting Flask app")
    app.run(debug=True)