test_parser.py

#!/usr/bin/env python3
"""
Test script for the PDF parser to verify functionality.
"""

import json
import sys
from pathlib import Path
from pdf_parser import PDFParser


def test_basic_parsing():
    """Test basic PDF parsing functionality."""
    pdf_file = "fund_factsheet.pdf"
    
    if not Path(pdf_file).exists():
        print(f"Error: {pdf_file} not found in current directory")
        return False
    
    try:
        print(f"Testing PDF parsing with {pdf_file}...")
        
        with PDFParser(pdf_file) as parser:
            extracted_data = parser.parse_pdf()
            
        # Basic validation
        assert "document_info" in extracted_data
        assert "pages" in extracted_data
        assert len(extracted_data["pages"]) > 0
        
        # Check structure of first page
        first_page = extracted_data["pages"][0]
        assert "page_number" in first_page
        assert "content" in first_page
        
        print(f"✓ Successfully parsed {len(extracted_data['pages'])} pages")
        print(f"✓ Found {len(first_page['content'])} content items on first page")
        
        # Count content types
        content_types = {}
        for page in extracted_data["pages"]:
            for item in page["content"]:
                content_type = item.get("type", "unknown")
                content_types[content_type] = content_types.get(content_type, 0) + 1
        
        print("Content type summary:")
        for content_type, count in content_types.items():
            print(f"  - {content_type}: {count}")
        
        # Save test output
        output_file = "test_output.json"
        with open(output_file, 'w', encoding='utf-8') as f:
            json.dump(extracted_data, f, indent=2, ensure_ascii=False)
        
        print(f"✓ Test output saved to {output_file}")
        return True
        
    except Exception as e:
        print(f"✗ Test failed with error: {e}")
        import traceback
        traceback.print_exc()
        return False


def main():
    """Run the test."""
    print("PDF Parser Test Suite")
    print("=" * 50)
    
    success = test_basic_parsing()
    
    if success:
        print("\n✓ All tests passed!")
        return 0
    else:
        print("\n✗ Tests failed!")
        return 1


if __name__ == "__main__":
    sys.exit(main())