Last active
April 18, 2024 22:13
-
-
Save stuaxo/f1beb981dc4845921c31fcb4e16f4821 to your computer and use it in GitHub Desktop.
List filenames and content ready for input to a LLM - optionally copy to the clipboard with a prompt.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
#!/usr/bin/env python3 | |
# Usage: python dirtogpt.py [--dir /path/to/directory] [--glob *.py] [--prompt [Custom prompt]] [--copy] [--exclude *.pyc] | |
import argparse | |
import pathlib | |
import fnmatch | |
try: | |
import pyperclip | |
except ImportError: | |
pyperclip = None | |
def append_file_content(output, path): | |
try: | |
content = path.read_text() | |
except UnicodeDecodeError: | |
#content = "Skipped (binary file)\n\n" | |
content = "" | |
except Exception as ex: | |
content = f"Skipped (error reading file: {ex})\n\n" | |
output += f"#:{path}:\n{content}\n\n" | |
return output | |
def dirtogpt(output, directory, glob, exclude): | |
file_count = 0 | |
p = pathlib.Path(directory) | |
for child in p.iterdir(): | |
if child.is_dir(): | |
output, sub_file_count = dirtogpt(output, child, glob, exclude) | |
file_count += sub_file_count | |
for child in p.glob(glob): | |
if child.is_file() and not fnmatch.fnmatch(str(child), exclude): | |
output = append_file_content(output, child) | |
file_count += 1 | |
return output, file_count | |
if __name__ == "__main__": | |
parser = argparse.ArgumentParser(description='Process some files.') | |
parser.add_argument('--dir', type=str, help='Directory to process', default=".") | |
parser.add_argument('--exclude', type=str, help='Glob pattern to exclude', default="") | |
parser.add_argument('--glob', type=str, help='Glob pattern to match', default="*") | |
parser.add_argument('--prompt', nargs='?', const="Filenames followed by file content-:", default=None, help='Display a prompt before the files') | |
parser.add_argument('--copy', action='store_true', help='Copy output to the clipboard instead of stdout') | |
args = parser.parse_args() | |
output = "" | |
if args.prompt is not None: | |
output += args.prompt + "\n" | |
output, file_count = dirtogpt(output, args.dir, args.glob, args.exclude) | |
token_count = len(output.split()) | |
if args.copy: | |
if pyperclip: | |
pyperclip.copy(output) | |
print(f"Copied {file_count} files, {len(output)} bytes, and approximately {token_count} tokens to the clipboard.") | |
else: | |
print("The --copy option requires the 'pyperclip' module. Please install it to use this functionality.") | |
else: | |
print(output) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment