"""Utility functions for image handling and processing."""

import base64
import re


def ensure_bytes_format(image_data):
    """
    Ensure image data is in the proper format for storage.
    File paths and URLs should remain as strings, only actual image data should be bytes.

    Args:
        image_data: Image data that could be bytes, base64 string, file path, or URL string

    Returns:
        Properly formatted data for database storage (bytes for actual image data, string for paths/URLs)
    """
    if image_data is None:
        return None

    # If it's already bytes, return as is (for actual image data)
    if isinstance(image_data, bytes):
        return image_data

    # If it's a string, determine if it's a file path, URL, or base64 data
    if isinstance(image_data, str):
        # Check if it's a data URL (base64 encoded image)
        if image_data.startswith('data:image/'):
            try:
                # Extract base64 part and decode to bytes
                base64_part = image_data.split(',')[1]
                return base64.b64decode(base64_part)
            except Exception:
                # If decoding fails, return the original string
                return image_data
        # Check if it looks like a file path
        # This regex matches common file path patterns like /tmp/..., ./path/, ../path/, C:\path\, etc.
        elif re.match(r'^[/\\]|[.][/\\]|[\w]:[/\\]|.*\.(jpg|jpeg|png|gif|webp|bmp|tiff|svg)$', image_data.lower()):
            # This appears to be a file path, return as string - don't convert to bytes
            return image_data
        # Check if it looks like a URL (has a scheme like http:// or https://)
        elif image_data.startswith(('http://', 'https://', 'ftp://', 'file://')):
            # This is a URL, return as string
            return image_data
        else:
            # Assume it's a string that should remain as is
            return image_data

    # For any other type, return as is
    return image_data