import json import os import logging logger = logging.getLogger(__name__) def save_results(results): """Save analysis results to file.""" try: output_dir = 'results' if not os.path.exists(output_dir): os.makedirs(output_dir) # Convert numpy types to Python native types processed_results = [] for result in results: try: # Handle question_id conversion if hasattr(result['question_id'], 'item'): question_id = result['question_id'].item() else: question_id = int(result['question_id']) # Handle misconception_ids conversion misconception_ids = [] for mid in result['misconception_ids']: try: if hasattr(mid, 'item'): # numpy type misconception_ids.append(mid.item()) else: misconception_ids.append(int(mid)) except (ValueError, TypeError): logger.warning(f"Could not convert misconception ID: {mid}, skipping") # Extract misconception names from details misconception_names = [] if 'misconception_details' in result: for detail in result['misconception_details']: misconception_names.append(detail['name']) processed_result = { 'question_id': question_id, 'question_text': result.get('question_text', ''), 'misconception_ids': misconception_ids, 'misconception_names': misconception_names, 'misconception_details': result.get('misconception_details', []) } processed_results.append(processed_result) except Exception as item_error: logger.error(f"Error processing result item: {str(item_error)}") continue output_file = os.path.join(output_dir, 'misconceptions.json') with open(output_file, 'w') as f: json.dump(processed_results, f, indent=4) logger.info(f"Saved {len(processed_results)} results to {output_file}") return True except Exception as e: logger.error(f"Error saving results: {str(e)}") return False