test_json_analyzer.py

import asyncio
import logging
import sys
from pathlib import Path

# Configure logging
logging.basicConfig(
    level=logging.INFO,
    format="%(asctime)s - %(name)s - %(levelname)s - %(message)s",
)
logger = logging.getLogger(__name__)

async def test_json_analyzer():
    """Test the JsonAnalyzer with the problematic file."""
    try:
        # Import the JsonAnalyzer
        from src.analyzers.json_analyzer import JsonAnalyzer
        
        # Create an instance of the analyzer
        analyzer = JsonAnalyzer()
        
        # Path to the problematic JSON file
        problematic_file = Path(r"C:\Users\jonah\Documents\Cline\datasets\tategallery\tategallery_artists\a\abakanowicz-magdalena-10093.json")
        
        if not problematic_file.exists():
            logger.error(f"File not found: {problematic_file}")
            return 1
            
        logger.info(f"Testing file: {problematic_file}")
        
        # Try to detect if the file is JSON
        is_json = await analyzer.detect(problematic_file)
        logger.info(f"File detected as JSON: {is_json}")
        
        if is_json:
            # Try to analyze the file
            schema = await analyzer.analyze(problematic_file)
            logger.info(f"Successfully analyzed file and extracted schema: {schema.id}")
            logger.info(f"Found {len(schema.tables)} tables in the schema")
            logger.info("JSON analyzer is working correctly!")
            return 0
        else:
            logger.error("File not detected as JSON")
            return 1
            
    except Exception as e:
        logger.error(f"Error testing JSON analyzer: {str(e)}", exc_info=True)
        return 1

if __name__ == "__main__":
    exit_code = asyncio.run(test_json_analyzer())
    sys.exit(exit_code)