install.py

#!/usr/bin/env python3
"""
Installation script for PDF Parser dependencies.
This script helps install the required packages with proper error handling.
"""

import subprocess
import sys
import os
from pathlib import Path


def run_command(command, description):
    """Run a command and handle errors."""
    print(f"\n{description}...")
    try:
        result = subprocess.run(command, shell=True, check=True, 
                              capture_output=True, text=True)
        print(f"✓ {description} completed successfully")
        return True
    except subprocess.CalledProcessError as e:
        print(f"✗ {description} failed:")
        print(f"Error: {e}")
        if e.stdout:
            print(f"Output: {e.stdout}")
        if e.stderr:
            print(f"Error output: {e.stderr}")
        return False


def check_python_version():
    """Check if Python version is compatible."""
    version = sys.version_info
    if version.major < 3 or (version.major == 3 and version.minor < 8):
        print(f"✗ Python {version.major}.{version.minor} is not supported.")
        print("Please use Python 3.8 or higher.")
        return False
    
    print(f"✓ Python {version.major}.{version.minor}.{version.micro} is compatible")
    return True


def check_java():
    """Check if Java is available (required for tabula-py)."""
    try:
        result = subprocess.run(["java", "-version"], capture_output=True, text=True)
        if result.returncode == 0:
            print("✓ Java is available")
            return True
    except FileNotFoundError:
        pass
    
    print("⚠ Java not found. This may cause issues with table extraction.")
    print("Consider installing Java for full functionality.")
    return False


def install_packages():
    """Install required packages."""
    requirements_file = Path("requirements.txt")
    
    if not requirements_file.exists():
        print("✗ requirements.txt not found")
        return False
    
    # Upgrade pip first
    if not run_command(f"{sys.executable} -m pip install --upgrade pip", 
                      "Upgrading pip"):
        print("⚠ Pip upgrade failed, continuing anyway...")
    
    # Install packages
    return run_command(f"{sys.executable} -m pip install -r requirements.txt", 
                      "Installing required packages")


def test_imports():
    """Test if critical packages can be imported."""
    packages = [
        ("fitz", "PyMuPDF"),
        ("pdfplumber", "pdfplumber"),
        ("pandas", "pandas"),
        ("camelot", "camelot-py")
    ]
    
    print("\nTesting package imports...")
    all_good = True
    
    for package, display_name in packages:
        try:
            __import__(package)
            print(f"✓ {display_name} imported successfully")
        except ImportError as e:
            print(f"✗ {display_name} import failed: {e}")
            all_good = False
    
    return all_good


def main():
    """Main installation function."""
    print("PDF Parser Installation Script")
    print("=" * 50)
    
    # Check Python version
    if not check_python_version():
        return 1
    
    # Check Java (optional but recommended)
    check_java()
    
    # Install packages
    print(f"\nInstalling packages from requirements.txt...")
    if not install_packages():
        print("\n✗ Package installation failed!")
        return 1
    
    # Test imports
    if not test_imports():
        print("\n⚠ Some packages failed to import. Check the error messages above.")
        print("You may need to install additional system dependencies.")
        return 1
    
    print("\n" + "=" * 50)
    print("✓ Installation completed successfully!")
    print("\nYou can now run the PDF parser:")
    print("  python pdf_parser.py your_file.pdf")
    print("\nOr run the test:")
    print("  python test_parser.py")
    
    return 0


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