utils/helpers.py

from datetime import datetime
import re

def calculate_age(birth_date_str):
    """Calculate age from a birth date string (YYYY-MM-DD)."""
    if not birth_date_str:
        return None
    try:
        birth_date = datetime.fromisoformat(birth_date_str)
        today = datetime.utcnow()
        age = today.year - birth_date.year - ((today.month, today.day) < (birth_date.month, birth_date.day))
        return age
    except (ValueError, TypeError):
        return None

def standardize_language(language):
    """Standardize language codes or text to a consistent format."""
    if not language:
        return "en"
    language = language.lower().strip()
    language_map = {
        "english": "en",
        "french": "fr",
        "spanish": "es",
        "german": "de"
    }
    return language_map.get(language, re.sub(r'[^a-z]', '', language)[:2])

def escape_latex_special_chars(text):
    """Escape special characters for LaTeX compatibility."""
    if not isinstance(text, str) or not text:
        return ""
    latex_special_chars = {
        '&': r'\&',
        '%': r'\%',
        '$': r'\$',
        '#': r'\#',
        '_': r'\_',
        '{': r'\{',
        '}': r'\}',
        '~': r'\textasciitilde{}',
        '^': r'\textasciicircum{}',
        '\\': r'\textbackslash{}'
    }
    return ''.join(latex_special_chars.get(c, c) for c in text)

def hyphenate_long_strings(text, max_length=30):
    """Insert hyphens into long strings to prevent LaTeX table overflow."""
    if not isinstance(text, str) or not text:
        return ""
    if len(text) <= max_length:
        return text
    # Insert a zero-width space (allowing LaTeX to break) every max_length characters
    result = []
    for i in range(0, len(text), max_length):
        chunk = text[i:i + max_length]
        result.append(chunk)
    return r"\-".join(result)

def format_timestamp(timestamp_str):
    """Format a timestamp string for display in LaTeX."""
    if not timestamp_str:
        return "N/A"
    try:
        # Handle ISO format timestamps
        dt = datetime.fromisoformat(timestamp_str.replace('Z', '+00:00'))
        return dt.strftime("%Y-%m-%d %H:%M")
    except (ValueError, TypeError):
        return str(timestamp_str)