IRIS Legacy

""" Attachment Helper - Shared utilities for attachment processing Provides: - File storage with unique tokens - Markdown conversion via markitdown - Public URL generation """ import os import base64 import hashlib import logging from datetime import datetime from typing import Dict, Optional from pathlib import Path logger = logging.getLogger(__name__) # Storage directory for attachments ATTACHMENTS_DIR = "/app/attachments" # Base URL for public download (configured via env or default) # Uses /iris/ prefix which nginx proxies to iris-app:8001/mcp/ # Load from environment variable (configured in .env) BASE_DOWNLOAD_URL = os.getenv("ATTACHMENTS_BASE_URL", "https://trustypa.brainaihub.tech/iris/attachments") def generate_token(content: bytes) -> str: """Generate unique token from content hash + timestamp""" content_hash = hashlib.sha256(content).hexdigest()[:16] timestamp = datetime.now().strftime("%Y%m%d_%H%M%S") return f"{timestamp}_{content_hash}" def save_attachment( content: bytes, filename: str, token: Optional[str] = None ) -> Dict[str, str]: """ Save attachment to filesystem with unique token. Args: content: Binary file content filename: Original filename token: Optional custom token (auto-generated if None) Returns: { "file_path": "/app/attachments/token_filename.pdf", "token": "20231124_abc123", "download_url": "https://.../download/token_filename.pdf", "download_link_html": "<a href='...' target='_blank'>filename.pdf</a>" } """ # Create attachments directory os.makedirs(ATTACHMENTS_DIR, exist_ok=True) # Generate token if not provided if not token: token = generate_token(content) # Build safe filename safe_filename = f"{token}_{filename}" file_path = os.path.join(ATTACHMENTS_DIR, safe_filename) # Save file with open(file_path, 'wb') as f: f.write(content) logger.info(f"💾 Saved attachment: {file_path} ({len(content)} bytes)") # Generate download URL download_url = f"{BASE_DOWNLOAD_URL}/download/{safe_filename}" # Generate HTML link that opens in new tab download_link_html = f'<a href="{download_url}" target="_blank">{filename}</a>' return { "file_path": file_path, "token": token, "download_url": download_url, "download_link_html": download_link_html } def convert_to_markdown(file_path: str) -> Optional[str]: """ Convert file to markdown using markitdown. Supports: PDF, DOCX, PPTX, XLSX, images, audio, HTML, CSV, JSON, XML, ZIP Args: file_path: Path to file Returns: Markdown string or None if conversion fails """ try: from markitdown import MarkItDown md_converter = MarkItDown() result = md_converter.convert(file_path) markdown_content = result.text_content logger.info(f"✅ Converted to markdown: {file_path} ({len(markdown_content)} chars)") return markdown_content except Exception as e: logger.error(f"❌ Failed to convert {file_path} to markdown: {e}") return None def process_attachment_for_analysis( content: bytes, filename: str, content_type: Optional[str] = None ) -> Dict: """ Complete attachment processing: save + convert to markdown. This is the main function to use for both PEC and Microsoft attachments. Args: content: Binary file content filename: Original filename content_type: MIME type (optional) Returns: { "filename": "document.pdf", "content_type": "application/pdf", "size": 123456, "file_path": "/app/attachments/20231124_abc123_document.pdf", "download_url": "https://.../download/20231124_abc123_document.pdf", "download_link_html": "<a href='...' target='_blank'>document.pdf</a>", "markdown_content": "# Document\n\nContent here...", "markdown_available": true } """ # Save file storage_info = save_attachment(content, filename) # Convert to markdown markdown_content = convert_to_markdown(storage_info["file_path"]) return { "filename": filename, "content_type": content_type or "application/octet-stream", "size": len(content), "file_path": storage_info["file_path"], "download_url": storage_info["download_url"], "download_link_html": storage_info["download_link_html"], "markdown_content": markdown_content, "markdown_available": markdown_content is not None }