solventum-image-metadata/backend/app/processors/utils.py

"""Utility functions for backup, logging, and file operations."""

import shutil
import logging
from pathlib import Path
from datetime import datetime
from typing import Optional
from .config import Config

# Setup logging
logging.basicConfig(
    level=logging.INFO,
    format='%(asctime)s - %(name)s - %(levelname)s - %(message)s'
)
logger = logging.getLogger(__name__)

def create_backup(file_path: str) -> Optional[Path]:
    """
    Create a backup of the file before modification.

    Args:
        file_path: Path to the file to backup

    Returns:
        Path to the backup file, or None if backup failed
    """
    try:
        source = Path(file_path)
        if not source.exists():
            logger.error(f"File not found for backup: {file_path}")
            return None

        # Create backup filename with timestamp
        timestamp = datetime.now().strftime("%Y%m%d_%H%M%S")
        backup_filename = f"{source.stem}_{timestamp}{source.suffix}"
        backup_path = Config.BACKUP_DIR / backup_filename

        # Ensure backup directory exists
        Config.BACKUP_DIR.mkdir(parents=True, exist_ok=True)

        # Copy file
        shutil.copy2(source, backup_path)
        logger.info(f"Backup created: {backup_path}")

        return backup_path

    except Exception as e:
        logger.error(f"Failed to create backup for {file_path}: {e}")
        return None

def get_logger(name: str) -> logging.Logger:
    """
    Get a logger instance.

    Args:
        name: Logger name

    Returns:
        Logger instance
    """
    return logging.getLogger(name)

def format_metadata_comparison(old_metadata: dict, new_metadata: dict) -> str:
    """
    Format metadata comparison for display.

    Args:
        old_metadata: Old metadata dictionary
        new_metadata: New metadata dictionary

    Returns:
        Formatted comparison string
    """
    lines = ["\n" + "="*60]
    lines.append("METADATA COMPARISON")
    lines.append("="*60)

    all_keys = set(old_metadata.keys()) | set(new_metadata.keys())

    for key in sorted(all_keys):
        old_value = old_metadata.get(key, "N/A")
        new_value = new_metadata.get(key, "N/A")

        lines.append(f"\n{key.upper()}:")
        lines.append(f"  Old: {old_value}")
        lines.append(f"  New: {new_value}")

        if old_value != new_value:
            lines.append("  [CHANGED]")

    lines.append("="*60 + "\n")
    return "\n".join(lines)

def sanitize_metadata_value(value: str, max_length: int = 500) -> str:
    """
    Sanitize and truncate metadata value.

    Args:
        value: Metadata value
        max_length: Maximum length

    Returns:
        Sanitized value
    """
    if not value:
        return ""

    # Remove control characters and excessive whitespace
    value = ' '.join(value.split())

    # Truncate if too long
    if len(value) > max_length:
        value = value[:max_length-3] + "..."

    return value.strip()

def validate_file_path(file_path: str) -> bool:
    """
    Validate file path exists and is accessible.

    Args:
        file_path: Path to validate

    Returns:
        True if valid
    """
    try:
        path = Path(file_path)
        return path.exists() and path.is_file()
    except Exception:
        return False

def get_file_size_mb(file_path: str) -> float:
    """
    Get file size in MB.

    Args:
        file_path: Path to file

    Returns:
        File size in MB
    """
    try:
        size_bytes = Path(file_path).stat().st_size
        return size_bytes / (1024 * 1024)
    except Exception:
        return 0.0

def create_report_entry(file_path: str, file_type: str, old_metadata: dict,
                       new_metadata: dict, status: str) -> dict:
    """
    Create a report entry for CSV export.

    Args:
        file_path: Path to file
        file_type: Type of file
        old_metadata: Old metadata
        new_metadata: New metadata
        status: Processing status (success/failed)

    Returns:
        Dictionary with report data
    """
    return {
        'timestamp': datetime.now().isoformat(),
        'file_path': file_path,
        'file_type': file_type,
        'old_title': old_metadata.get('title', 'N/A'),
        'new_title': new_metadata.get('title', 'N/A'),
        'old_subject': old_metadata.get('subject', 'N/A'),
        'new_subject': new_metadata.get('subject', 'N/A'),
        'old_keywords': old_metadata.get('keywords', 'N/A'),
        'new_keywords': new_metadata.get('keywords', 'N/A'),
        'status': status
    }