""" Image Processor Utilities. This module handles image processing for structure diagrams and scanned documents. Design Note: Image processing here is limited to: - Basic image validation - Path management - Potential OCR preparation (TODO) Actual chemical structure recognition would require external services (e.g., OSRA, ChemDraw API) and is out of scope for this implementation. """ from typing import Optional, Tuple from pathlib import Path import base64 class ImageProcessor: """ Image processor for structure diagrams and scanned documents. Primary responsibilities: - Validate image files - Prepare images for LLM vision APIs (if supported) - Encode images for embedding """ SUPPORTED_FORMATS = {'.png', '.jpg', '.jpeg', '.gif', '.webp', '.bmp'} def __init__(self): """Initialize the image processor.""" self._pil_available = self._check_pil() def _check_pil(self) -> bool: """Check if PIL/Pillow is available.""" try: from PIL import Image return True except ImportError: return False def validate_image(self, file_path: str) -> bool: """ Validate that a file is a readable image. Args: file_path: Path to the image file Returns: True if valid image, False otherwise """ path = Path(file_path) # Check extension if path.suffix.lower() not in self.SUPPORTED_FORMATS: return False # Check file exists if not path.exists(): return False # Try to open with PIL if available if self._pil_available: try: from PIL import Image with Image.open(file_path) as img: img.verify() return True except Exception: return False return True def get_image_info(self, file_path: str) -> Optional[dict]: """ Get basic information about an image. Args: file_path: Path to the image file Returns: Dictionary with image info, or None if failed """ if not self._pil_available: return {"path": file_path, "status": "PIL not available"} try: from PIL import Image with Image.open(file_path) as img: return { "path": file_path, "format": img.format, "mode": img.mode, "size": img.size, # (width, height) "width": img.size[0], "height": img.size[1], } except Exception as e: return {"path": file_path, "error": str(e)} def encode_base64(self, file_path: str) -> Optional[str]: """ Encode an image as base64 string. Useful for embedding in HTML or sending to vision APIs. Args: file_path: Path to the image file Returns: Base64 encoded string, or None if failed """ try: with open(file_path, "rb") as f: image_data = f.read() return base64.b64encode(image_data).decode('utf-8') except Exception as e: print(f"Error encoding image: {e}") return None def get_data_uri(self, file_path: str) -> Optional[str]: """ Get a data URI for embedding an image directly in HTML. Args: file_path: Path to the image file Returns: Data URI string, or None if failed """ path = Path(file_path) suffix = path.suffix.lower() # Map extension to MIME type mime_types = { '.png': 'image/png', '.jpg': 'image/jpeg', '.jpeg': 'image/jpeg', '.gif': 'image/gif', '.webp': 'image/webp', '.bmp': 'image/bmp', } mime_type = mime_types.get(suffix, 'image/png') base64_data = self.encode_base64(file_path) if base64_data: return f"data:{mime_type};base64,{base64_data}" return None def resize_for_report( self, file_path: str, max_width: int = 400, max_height: int = 300 ) -> Optional[str]: """ Resize an image for report embedding. Creates a temporary resized copy suitable for report generation. Args: file_path: Path to the original image max_width: Maximum width in pixels max_height: Maximum height in pixels Returns: Path to resized image, or original path if resizing fails """ if not self._pil_available: return file_path try: from PIL import Image import tempfile with Image.open(file_path) as img: # Calculate new size maintaining aspect ratio img.thumbnail((max_width, max_height), Image.Resampling.LANCZOS) # Save to temp file suffix = Path(file_path).suffix with tempfile.NamedTemporaryFile( suffix=suffix, delete=False ) as tmp: img.save(tmp.name) return tmp.name except Exception as e: print(f"Error resizing image: {e}") return file_path def prepare_for_llm(self, file_path: str) -> Optional[dict]: """ Prepare an image for LLM vision API submission. Returns a dictionary suitable for vision model APIs. Args: file_path: Path to the image file Returns: Dictionary with image data for API submission """ if not self.validate_image(file_path): return None base64_data = self.encode_base64(file_path) if not base64_data: return None path = Path(file_path) mime_types = { '.png': 'image/png', '.jpg': 'image/jpeg', '.jpeg': 'image/jpeg', '.gif': 'image/gif', '.webp': 'image/webp', } mime_type = mime_types.get(path.suffix.lower(), 'image/png') return { "type": "image", "source": { "type": "base64", "media_type": mime_type, "data": base64_data, } }