manimator/inputs/processor.py

import logging
from pathlib import Path
from typing import Union
from .pdf_parser import PDFParser
from .url_parser import URLParser

logger = logging.getLogger(__name__)

class InputProcessor:
    """
    Main entry point for processing different types of inputs.
    """
    
    @staticmethod
    def process(input_type: str, input_data: str) -> str:
        """
        Process input based on type.
        
        Args:
            input_type: 'text', 'pdf', or 'url'
            input_data: The actual text, file path, or URL
            
        Returns:
            Extracted text content
        """
        logger.info(f"Processing input type: {input_type}")
        
        if input_type == 'text':
            return input_data
            
        elif input_type == 'pdf':
            # Check if input_data is a file path
            is_path = False
            try:
                # Only check if it looks like a path (not too long)
                if len(str(input_data)) < 256 and Path(input_data).exists():
                    is_path = True
            except Exception:
                pass
                
            if is_path:
                return PDFParser.parse(input_data)
            
            # Try to decode as base64
            import base64
            import tempfile
            import os
            
            try:
                # Create a temporary file
                with tempfile.NamedTemporaryFile(delete=False, suffix=".pdf") as tmp:
                    # If it's bytes, write directly. If str, decode.
                    if isinstance(input_data, bytes):
                        tmp.write(input_data)
                    else:
                        # Handle potential header "data:application/pdf;base64,"
                        if "," in input_data:
                            input_data = input_data.split(",")[1]
                        tmp.write(base64.b64decode(input_data))
                    tmp_path = tmp.name
                
                logger.info(f"Saved base64 PDF to temporary file: {tmp_path}")
                try:
                    text = PDFParser.parse(tmp_path)
                    return text
                finally:
                    # Cleanup temp file
                    try:
                        os.unlink(tmp_path)
                    except Exception as e:
                        logger.warning(f"Failed to delete temp PDF file: {e}")
                        
            except Exception as e:
                logger.error(f"Failed to process PDF input: {e}")
                raise ValueError(f"Invalid PDF input: {e}")
            
        elif input_type == 'url':
            return URLParser.parse(input_data)
            
        else:
            raise ValueError(f"Unsupported input type: {input_type}")