question_generator.py

import random
import csv
import os
import logging
import hashlib
import json
import re
from typing import List, Dict
from datetime import datetime
from mistralai.client import MistralClient
from mistralai.models.chat_completion import ChatMessage

# Set up logging
logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(levelname)s - %(message)s')

# Get the Mistral API key from environment variables
api_key = os.environ.get("MISTRAL_API_KEY")
if not api_key:
    logging.error("MISTRAL_API_KEY environment variable is not set.")
    raise ValueError("MISTRAL_API_KEY environment variable is not set.")

model = "mistral-large-latest"

# Initialize Mistral client
client = MistralClient(api_key=api_key)

# ... (previous functions remain the same)

def extract_json_from_markdown(markdown_text: str) -> str:
    """Extract JSON content from Markdown-formatted text."""
    json_match = re.search(r'```json\s*(.*?)\s*```', markdown_text, re.DOTALL)
    if json_match:
        return json_match.group(1)
    else:
        raise ValueError("No JSON content found in the Markdown text")

def generate_microbiology_question() -> Dict[str, str]:
    """Generate a microbiology question."""
    # ... (previous code remains the same)
    
    try:
        chat_response = client.chat(
            model=model,
            messages=[
                ChatMessage(role="system", content="You are a medical educator creating unique microbiology questions for the NBME exam. Ensure each question is distinct from previously generated ones and follows the specified template."),
                ChatMessage(role="user", content=prompt)
            ]
        )
        
        response_content = chat_response.choices[0].message.content
        logging.info(f"Received response from Mistral API: {response_content[:100]}...")  # Log first 100 characters
        
        # Extract JSON from Markdown if necessary
        try:
            json_content = extract_json_from_markdown(response_content)
        except ValueError:
            json_content = response_content  # If not in Markdown, use the original content
        
        # Parse the JSON response
        question_data = json.loads(json_content)
        
        # Validate the structure of the parsed JSON
        required_keys = ["question", "options", "correct_answer", "explanation", "medical_reasoning"]
        if not all(key in question_data for key in required_keys):
            raise ValueError("Response is missing required keys")
        
        if not all(key in question_data["options"] for key in ["A", "B", "C", "D", "E"]):
            raise ValueError("Response is missing required option keys")
        
        # Save the question hash
        question_hash = hash_question(question_data['question'])
        if question_hash not in generated_questions:
            generated_questions.add(question_hash)
            save_generated_question(question_hash)
        
        return question_data
    
    except json.JSONDecodeError as e:
        logging.error(f"Failed to parse JSON response: {e}")
        logging.error(f"Response content: {response_content}")
        raise
    except ValueError as e:
        logging.error(f"Invalid response structure: {e}")
        logging.error(f"Response content: {response_content}")
        raise
    except Exception as e:
        logging.error(f"An unexpected error occurred: {e}")
        raise

# Example usage
if __name__ == "__main__":
    question = generate_microbiology_question()
    print(json.dumps(question, indent=2))