Simple script to help prevent GitHub Copilot from creating duplicates in a larger project

filestruct.py

#!/usr/bin/env python3

# USAGE: python filestruct.py <path> > report.md
# Add to your .github/copilot-instructions.md

import os
import re
import sys


def print_directory_tree(root_dir):
    print(f"## Directory tree of '{root_dir}':\n```\n")
    for dirpath, dirnames, filenames in os.walk(root_dir):
        # Exclude __pycache__ directories
        dirnames[:] = [d for d in dirnames if d != '__pycache__']
        # Sort directories and files alphabetically
        dirnames.sort()
        filenames.sort()
        level = dirpath.replace(root_dir, '').count(os.sep)
        indent = ' ' * 4 * level
        print(f"{indent}{os.path.basename(dirpath)}/")
        subindent = ' ' * 4 * (level + 1)
        for f in filenames:
            print(f"{subindent}{f}")
    print("```")

def print_init_files_with_all(root_dir):
    print("\nContents of __init__.py files with __all__ defined:")
    for dirpath, _, filenames in os.walk(root_dir):
        if '__init__.py' in filenames:
            init_file = os.path.join(dirpath, '__init__.py')
            with open(init_file) as f:
                content = f.read()
                if re.search(r'^\s*__all__\s*=.*', content, re.MULTILINE):
                    print(f"\n## {init_file}")
                    print("```python\n" + content + "```")

if __name__ == '__main__':
    # Accept path as an argument; default to current directory
    if len(sys.argv) > 1:
        root_dir = sys.argv[1]
    else:
        root_dir = '.'

    if not os.path.isdir(root_dir):
        print(f"Error: '{root_dir}' is not a directory.")
        sys.exit(1)

    print_directory_tree(root_dir)
    print_init_files_with_all(root_dir)