#!/usr/bin/env python
import re
import pyperclip
# pip install pyperclip

def redact_github_tokens(text):
    # Patterns for GitHub tokens (Personal Access Tokens, fine-tuned tokens, etc.)
    # Example formats:
    # - ghp_xxx... (Personal Access Token)
    # - github_pat_xxx... (Fine-grained)
    # - gho_xxx..., ghs_xxx..., ghr_xxx..., etc.
    token_patterns = [
        r'(gh[pousr]_[A-Za-z0-9]{30,})',
        r'(github_pat_[A-Za-z0-9_]{22,255})'
    ]
    # Combine all patterns
    combined_pattern = '|'.join(token_patterns)
    regex = re.compile(combined_pattern)

    def mask_token(match):
        token = match.group(0)
        underscore_indices = [i for i, c in enumerate(token) if c == '_']

        if len(underscore_indices) >= 3:
            prefix_end = underscore_indices[1] + 1  # preserve up to second underscore
        elif len(underscore_indices) == 1:
            prefix_end = 4  # preserve first 4 characters
        else:
            # default behavior: preserve up to last underscore
            prefix_end = token.rfind('_') + 1 if '_' in token else 4

        suffix_length = 8
        if len(token) <= prefix_end + suffix_length:
            return token  # too short to mask

        prefix = token[:prefix_end]
        suffix = token[-suffix_length:]
        masked_length = len(token) - len(prefix) - len(suffix)
        return prefix + '*' * masked_length + suffix

    # Substitute tokens with redacted version
    return regex.sub(mask_token, text)

def main():
    clipboard_content = pyperclip.paste()
    redacted_content = redact_github_tokens(clipboard_content)
    pyperclip.copy(redacted_content)
    print("Clipboard content scanned and redacted (if any GitHub tokens were found).")

if __name__ == "__main__":
    main()