Env Scanner

env_scanner.py

import os
import re
from pathlib import Path
from collections import defaultdict

def scan_env_usage():
    # Load current .env variables
    env_vars = set()
    if os.path.exists('.env'):
        with open('.env', 'r') as f:
            for line in f:
                line = line.strip()
                if line and not line.startswith('#'):
                    key = line.split('=')[0].strip()
                    env_vars.add(key)
    
    # Initialize results
    usage_count = defaultdict(int)
    file_references = defaultdict(list)
    
    # Define patterns to search for
    patterns = [
        r'process\.env\.([A-Z_][A-Z0-9_]*)',  # JavaScript/Node
        r'os\.environ\.get\([\'"]([A-Z_][A-Z0-9_]*)[\'"]',  # Python
        r'\$\{([A-Z_][A-Z0-9_]*)\}',  # Shell/Docker
        r'env\([\'"]([A-Z_][A-Z0-9_]*)[\'"]',  # PHP/Laravel
        r'ENV\[[\'"]([A-Z_][A-Z0-9_]*)[\'"]'  # Ruby
    ]
    
    # File extensions to scan
    extensions = {'.js', '.jsx', '.ts', '.tsx', '.py', '.php', '.rb', '.env', '.yaml', '.yml', '.sh', '.Dockerfile'}
    
    def scan_file(file_path):
        with open(file_path, 'r', encoding='utf-8', errors='ignore') as f:
            content = f.read()
            for pattern in patterns:
                matches = re.findall(pattern, content)
                for var_name in matches:
                    usage_count[var_name] += 1
                    file_references[var_name].append(str(file_path))
    
    # Scan all files recursively
    for path in Path('.').rglob('*'):
        if path.suffix in extensions and 'node_modules' not in str(path) and 'venv' not in str(path) and 'vendor' not in str(path):
            try:
                scan_file(path)
            except Exception as e:
                print(f"Error scanning {path}: {e}")
    
    # Generate report
    print("\nEnvironment Variable Usage Report")
    print("=" * 50)
    
    # Variables in .env but not used in code
    unused_vars = env_vars - set(usage_count.keys())
    if unused_vars:
        print("\nUnused variables in .env:")
        for var in sorted(unused_vars):
            print(f"- {var}")
    
    # Variables used in code but not in .env
    missing_vars = set(usage_count.keys()) - env_vars
    if missing_vars:
        print("\nVariables used in code but missing from .env:")
        for var in sorted(missing_vars):
            print(f"- {var} (used in {usage_count[var]} places)")
            for ref in file_references[var][:3]:  # Show first 3 references
                print(f"  → {ref}")
    
    # Most frequently used variables
    print("\nMost frequently used variables:")
    for var, count in sorted(usage_count.items(), key=lambda x: x[1], reverse=True)[:10]:
        print(f"- {var}: {count} references")

if __name__ == '__main__':
    scan_env_usage()