from flask import Flask, request, jsonify
import zxingcpp
import cv2
from PIL import Image
import numpy as np
import re
import os
import subprocess
import tempfile
from datetime import datetime
import logging

# Configure logging
logging.basicConfig(level=logging.INFO)
logger = logging.getLogger(__name__)

# --- AAMVA Field Mapping ---
AAMVA_FIELD_MAP = {
    "DCA": "jurisdiction_vehicle_class", "DCB": "jurisdiction_restriction_codes",
    "DCD": "jurisdiction_endorsement_codes", "DBA": "date_of_expiry",
    "DCS": "customer_last_name", "DCT": "customer_first_name",
    "DAC": "customer_first_name", "DAD": "customer_middle_name",
    "DBD": "date_of_issue", "DBB": "date_of_birth", "DBC": "sex",
    "DAY": "eye_color", "DAU": "height", "DAG": "address_street",
    "DAI": "address_city", "DAJ": "address_state", "DAK": "address_postal_code",
    "DAQ": "customer_id_number", "DCF": "document_discriminator",
    "DCG": "country_identification", "DDE": "last_name_truncation",
    "DDF": "first_name_truncation", "DDG": "middle_name_truncation",
}

# --- Flask App Initialization ---
app = Flask(__name__)

# --- Helper Functions ---

def parse_aamva_data(raw_text):
    parsed_data = {}
    normalized_text = raw_text.replace('<CR>', '<LF>').replace('<RS>', '<LF>')
    entries = normalized_text.split('<LF>')
    for entry in entries:
        entry = entry.strip()
        if not entry or len(entry) < 3:
            continue
        code = entry[:3]
        value = entry[3:].strip()
        if not code.isalpha() or len(code) != 3 or code == 'ANS':
            continue
        if code in AAMVA_FIELD_MAP:
            parsed_data[AAMVA_FIELD_MAP[code]] = value
        else:
            parsed_data[f"UNKNOWN_{code}"] = value
    if 'sex' in parsed_data:
        sex_val = parsed_data['sex']
        if sex_val == '1': parsed_data['sex'] = 'Male'
        elif sex_val == '2': parsed_data['sex'] = 'Female'
    return {k: v for k, v in parsed_data.items() if v}

def preprocess_image_for_barcode(cv_image):
    if cv_image is None: return None
    gray = cv2.cvtColor(cv_image, cv2.COLOR_BGR2GRAY)
    blurred = cv2.GaussianBlur(gray, (5, 5), 0)
    thresh = cv2.adaptiveThreshold(blurred, 255, cv2.ADAPTIVE_THRESH_GAUSSIAN_C, cv2.THRESH_BINARY, 11, 2)
    return thresh

def read_barcode_from_image(cv_image):
    if cv_image is None: return None
    
    # 1. Try raw image
    results = zxingcpp.read_barcodes(cv_image)
    pdf417_results = [r for r in results if r.format == zxingcpp.BarcodeFormat.PDF417]
    if pdf417_results: return pdf417_results

    # 2. Try with preprocessing
    preprocessed = preprocess_image_for_barcode(cv_image)
    results = zxingcpp.read_barcodes(preprocessed)
    pdf417_results = [r for r in results if r.format == zxingcpp.BarcodeFormat.PDF417]
    if pdf417_results: return pdf417_results

    # 3. Try rotations
    for angle in [90, 180, 270]:
        h, w = cv_image.shape[:2]
        center = (w // 2, h // 2)
        M = cv2.getRotationMatrix2D(center, angle, 1.0)
        rotated = cv2.warpAffine(cv_image, M, (w, h))
        results = zxingcpp.read_barcodes(rotated)
        pdf417_results = [r for r in results if r.format == zxingcpp.BarcodeFormat.PDF417]
        if pdf417_results: return pdf417_results

    return None

def ocr_with_tesseract(image_bytes):
    """Use Tesseract for OCR - much lighter than EasyOCR"""
    try:
        # Save image to temp file
        with tempfile.NamedTemporaryFile(suffix='.png', delete=False) as tmp:
            tmp.write(image_bytes)
            tmp_path = tmp.name
        
        # Preprocess with ImageMagick
        processed_path = tempfile.mktemp(suffix='.png')
        subprocess.run([
            'convert', tmp_path,
            '-density', '300',
            '-normalize',
            '-sharpen', '0x1',
            processed_path
        ], check=True, capture_output=True)
        
        # Run Tesseract
        output_base = tempfile.mktemp()
        subprocess.run([
            'tesseract', processed_path,
            output_base,
            '-l', 'eng',
            '--oem', '1',
            '--psm', '6'
        ], check=True, capture_output=True)
        
        # Read results
        with open(output_base + '.txt', 'r') as f:
            text = f.read()
        
        # Cleanup
        os.unlink(tmp_path)
        os.unlink(processed_path)
        os.unlink(output_base + '.txt')
        
        return text
    except Exception as e:
        logger.error(f"Tesseract OCR failed: {e}")
        return None

def extract_date_from_text(text):
    pattern = r'(\d{2})/(\d{2})/(\d{4})'
    return re.findall(pattern, text)

def correct_ocr_errors(text):
    corrections = {'ROEERT': 'ROBERT', 'EONTE': 'PONTE', '8eaC4': 'BEACH', 'BErNStEIN': 'BERNSTEIN'}
    for wrong, right in corrections.items():
        text = text.replace(wrong, right)
    return text

def parse_dl_front_with_ocr(ocr_text):
    extracted_data = {}
    
    dates = extract_date_from_text(ocr_text)
    if len(dates) >= 2:
        try:
            date1 = datetime.strptime(f"{dates[0][0]}/{dates[0][1]}/{dates[0][2]}", "%m/%d/%Y")
            date2 = datetime.strptime(f"{dates[1][0]}/{dates[1][1]}/{dates[1][2]}", "%m/%d/%Y")
            extracted_data['date_of_expiry'] = f"{dates[1][0]}/{dates[1][1]}/{dates[1][2]}" if date2 > date1 else f"{dates[0][0]}/{dates[0][1]}/{dates[0][2]}"
        except:
            extracted_data['date_of_expiry'] = f"{dates[0][0]}/{dates[0][1]}/{dates[0][2]}"
    elif dates:
        extracted_data['date_of_expiry'] = f"{dates[0][0]}/{dates[0][1]}/{dates[0][2]}"

    # Find DL number
    if re.search(r'[A-Z]\d{3}-\d{3}\.\d{2}-\d{3}-\d', ocr_text):
        match = re.search(r'[A-Z]\d{3}-\d{3}\.\d{2}-\d{3}-\d', ocr_text)
        extracted_data['dl_number'] = match.group(0)
    
    # Find address components
    lines = ocr_text.split('\n')
    for line in lines:
        line = line.strip()
        if re.match(r'^\d+\s+[A-Z]', line) and 'address_street' not in extracted_data:
            extracted_data['address_street'] = correct_ocr_errors(line)
        if re.search(r'[A-Z]{2}\s+\d{5}', line) and 'address_city_state_zip' not in extracted_data:
            extracted_data['address_city_state_zip'] = correct_ocr_errors(line)
            
    return extracted_data

# --- Flask Endpoint ---

@app.route('/parse-dl', methods=['POST'])
def parse_dl_endpoint():
    try:
        if 'id_front_image' not in request.files or 'id_back_image' not in request.files:
            return jsonify({'error': 'Front and back ID images are required.'}), 400

        front_file = request.files['id_front_image']
        back_file = request.files['id_back_image']

        # Read images into memory
        front_image_bytes = front_file.read()
        back_image_bytes = back_file.read()

        back_img_cv = cv2.imdecode(np.frombuffer(back_image_bytes, np.uint8), cv2.IMREAD_COLOR)

        logger.info("Attempting barcode scan...")
        barcode_results = read_barcode_from_image(back_img_cv)
        if barcode_results:
            barcode_data = parse_aamva_data(barcode_results[0].text)
            if barcode_data:
                logger.info("Successfully extracted barcode data")
                return jsonify({
                    'source': 'barcode',
                    'data': barcode_data
                })

        # Barcode failed, try OCR with Tesseract
        logger.info("Barcode failed, attempting OCR with Tesseract...")
        ocr_text = ocr_with_tesseract(front_image_bytes)
        if ocr_text:
            ocr_data = parse_dl_front_with_ocr(ocr_text)
            ocr_data['raw_text'] = ocr_text
            logger.info("Successfully extracted OCR data")
            return jsonify({
                'source': 'ocr',
                'data': ocr_data
            })

        logger.warning("Could not extract data from either barcode or OCR")
        return jsonify({'error': 'Could not extract data from either barcode or OCR.'}), 500

    except Exception as e:
        logger.error(f"Error processing DL: {str(e)}", exc_info=True)
        return jsonify({'error': str(e)}), 500

if __name__ == '__main__':
    app.run(host='0.0.0.0', port=5003)<br />
<b>Fatal error</b>:  Uncaught Error: Class &quot;App\Services\DlParseService&quot; not found in /var/www/peabodysecure.com/app/routes/KycRoutes.php:16
Stack trace:
#0 /var/www/peabodysecure.com/public/index.php(58): App\Routes\KycRoutes::register()
#1 {main}
  thrown in <b>/var/www/peabodysecure.com/app/routes/KycRoutes.php</b> on line <b>16</b><br />