src/utils.py

import string as st
from dateutil import parser
import logging
import sys
import re
from config.settings import COUNTRY_CODES

def setup_logger(name=__name__):
    """Sets up a logger with standard formatting."""
    logger = logging.getLogger(name)
    if not logger.handlers:
        handler = logging.StreamHandler(sys.stdout)
        formatter = logging.Formatter('%(asctime)s - %(name)s - %(levelname)s - %(message)s')
        handler.setFormatter(formatter)
        logger.addHandler(handler)
        logger.setLevel(logging.INFO)
    return logger

logger = setup_logger(__name__)

def parse_date(date_obj, airline="iraqi"):
    """Parses a date object or string based on airline format requirements."""
    try:
        date_str = date_obj.isoformat() if hasattr(date_obj, 'isoformat') else str(date_obj)
        date = parser.parse(date_str, yearfirst=True).date()
        
        if airline.lower() == "iraqi":
            # Iraqi format (previously Flydubai): DDMMMYY (e.g., 13NOV84)
            return date.strftime('%d%b%y').upper()
        else:
            # Default, Fly Dubai and Fly Baghdad format: DD/MM/YYYY (e.g., 12/1/2023)
            return date.strftime('%d/%m/%Y')
            
    except (ValueError, TypeError) as e:
        logger.debug(f"Date parsing failed for {date_obj}: {e}")
        return str(date_obj)

def clean_string(text):
    """Removes non-alphanumeric characters and converts to uppercase."""
    if not text:
        return ""
    return ''.join(i for i in text if i.isalnum()).upper()

def clean_name_field(text):
    """
    Cleans name/surname fields from MRZ.
    Handles separators (<<, <) and fixes OCR errors where filler '<' are read as 'K'.
    This version is safer and targets only trailing junk 'K's.
    """
    if not text:
        return ""
    
    text = text.upper()
    
    # Standard MRZ separator between surname and names
    text = text.replace("<<", " ")
    
    # Find the last non-'K' character's index
    last_good_char_idx = -1
    for i in range(len(text) - 1, -1, -1):
        if text[i] != 'K':
            last_good_char_idx = i
            break
            
    # If the string was all 'K's, it's empty.
    if last_good_char_idx == -1:
        return ""
        
    # Calculate how many 'K's are at the end
    trailing_k_count = len(text) - 1 - last_good_char_idx
    
    # If there are 2 or more trailing 'K's, they are junk fillers. Trim them.
    if trailing_k_count >= 2:
        text = text[:last_good_char_idx + 1]
        
    # Now, any remaining single '<' characters are separators.
    text = text.replace("<", " ")
    
    # Remove any numbers from the name
    text = ''.join(char for char in text if not char.isdigit())
    
    return text.strip()

def clean_mrz_line(line: str) -> str:
    """Fix bad spacing or bad OCR for MRZ lines."""
    if not line:
        return ""
    
    line = line.upper().replace(" ", "")
    
    # Remove accidental characters except allowed
    allowed = set(st.ascii_uppercase + st.digits + "<")
    line = "".join([c for c in line if c in allowed])

    # Ensure 44 length (standard TD3 MRZ length)
    # Note: TD1/TD2 might be different lengths (30 or 36), but this logic enforces 44.
    # We will keep existing logic for consistency but be aware of other formats.
    if len(line) < 44:
        line += "<" * (44 - len(line))
    return line[:44]

def get_country_name(country_code):
    """Resolves 3-letter country code to full name."""
    country_code = str(country_code).upper()
    for c in COUNTRY_CODES:
        if c['alpha-3'] == country_code:
            return c['name'].upper()
    return country_code

def get_sex(code):
    """Standardizes sex code."""
    code = str(code).upper() if code else ''
    if code in ['M', 'F']:
        return code
    if code == '0':
        return 'M' # Fallback based on existing logic
    return code

def parse_barcode_data(barcode_data):
    """
    Parse PDF417 barcode data from passport.
    Handles various formats from different countries.
    """
    try:
        lines = barcode_data.split('\n')
        if len(lines) < 2:
            return None
        
        # Extract names from first line
        name_line = lines[0].replace('@','').replace('<',' ').strip()
        surname, given_names = name_line.split(' ', 1) if ' ' in name_line else (name_line, '')
        
        # Parse remaining fields (simplified format)
        if len(lines[1]) >= 25:
            passport_number = lines[1][:9].strip()
            nationality = lines[1][9:12].strip()
            dob = parse_date(lines[1][12:18])
            sex = lines[1][18]
            expiry = parse_date(lines[1][19:25])
            
            return {
                "surname": clean_name_field(surname),
                "given_names": clean_name_field(given_names),
                "passport_number": passport_number,
                "nationality": nationality,
                "sex": sex,
                "date_of_birth": dob,
                "expiration_date": expiry,
                "barcode_data": barcode_data
            }
        return None
    except Exception as e:
        logger.error(f"Error parsing barcode data: {e}")
        return None