from __future__ import annotations

import ast
from pathlib import Path


def find_never_imported_files(package_dir):
    """Find Python files that are never imported by other files."""
    package_path = Path(package_dir)
    python_files = list(package_path.rglob("*.py"))
    
    # Track which modules are imported
    imported_modules = set()
    
    for file_path in python_files:
        try:
            with open(file_path, encoding='utf-8') as f:
                tree = ast.parse(f.read())
        except (SyntaxError, UnicodeDecodeError):
            continue
            
        for node in ast.walk(tree):
            if isinstance(node, ast.Import):
                for alias in node.names:
                    imported_modules.add(alias.name)
                    
            elif isinstance(node, ast.ImportFrom):
                if node.module:
                    imported_modules.add(node.module)
                    # Also add individual imports
                    for alias in node.names:
                        if node.module:
                            imported_modules.add(f"{node.module}.{alias.name}")
    
    # Find files that are never imported
    never_imported = []
    
    for file_path in python_files:
        # Skip __init__.py files as they're special
        if file_path.name == "__init__.py":
            continue
            
        # Convert file path to potential module names
        relative_path = file_path.relative_to(package_path)
        module_path_parts = relative_path.with_suffix('').parts
        
        # Generate possible import names for this file
        possible_names = set()
        
        # Direct module name
        if len(module_path_parts) == 1:
            possible_names.add(module_path_parts[0])
        else:
            # Package.module format
            possible_names.add('.'.join(module_path_parts))
            # Just the module name
            possible_names.add(module_path_parts[-1])
        
        # Check if any possible name was imported
        if not any(name in imported_modules for name in possible_names):
            # Double-check by looking for any partial matches
            file_stem = file_path.stem
            if not any(file_stem in imp for imp in imported_modules):
                never_imported.append(file_path)
    
    return never_imported

# Usage
unused_files = find_never_imported_files("mteb")
print("Files that are never imported:")
for file in unused_files:
    print(f"  {file}")