git_cleaner.py: Recursively run git gc on all your repositories

#!/usr/bin/env python3
import os
import subprocess
import argparse
import sys

def get_dir_stats(path):
    """Returns a tuple of (total_size_bytes, total_file_count)."""
    total_size = 0
    file_count = 0
    for dirpath, _, filenames in os.walk(path):
        for f in filenames:
            fp = os.path.join(dirpath, f)
            if not os.path.islink(fp):
                total_size += os.path.getsize(fp)
                file_count += 1
    return total_size, file_count

def format_size(bytes_size):
    """Convert bytes to a human-readable string."""
    for unit in ['B', 'KB', 'MB', 'GB']:
        if abs(bytes_size) < 1024:
            return f"{bytes_size:.2f} {unit}"
        bytes_size /= 1024
    return f"{bytes_size:.2f} TB"

def main():
    parser = argparse.ArgumentParser(
        description="Recursively run 'git gc', reporting space and files saved."
    )
    parser.add_argument(
        "target_dir",
        nargs="?",
        default=".",
        help="Directory to search (default: '.')"
    )

    args, unknown_args = parser.parse_known_args()
    target_path = os.path.abspath(args.target_dir)

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

    total_saved_bytes = 0
    total_files_removed = 0
    repos_processed = 0

    print(f"--- Scanning: {target_path} ---")

    for root, dirs, _ in os.walk(target_path):
        if ".git" in dirs:
            repos_processed += 1
            git_dir = os.path.join(root, ".git")

            # 1. Capture stats before
            size_before, files_before = get_dir_stats(git_dir)

            print(f"\n[{repos_processed}] Optimizing: {root}")

            try:
                # 2. Run git gc
                # Note: We use -C to run git in the specific directory
                cmd = ["git", "-C", root, "gc"] + unknown_args
                subprocess.run(cmd, check=True, capture_output=True)

                # 3. Capture stats after
                size_after, files_after = get_dir_stats(git_dir)

                saved_size = size_before - size_after
                removed_files = files_before - files_after

                total_saved_bytes += max(0, saved_size)
                total_files_removed += max(0, removed_files)

                print(f"    Files: {files_before} -> {files_after} ({removed_files} removed)")
                print(f"    Size:  {format_size(size_before)} -> {format_size(size_after)} ({format_size(saved_size)} saved)")

            except subprocess.CalledProcessError as e:
                print(f"    [!] Error: {e.stderr.decode().strip()}")

            # Efficiency: don't walk into the .git folder we just processed
            dirs.remove(".git")

    # Final Summary Table
    print("\n" + "="*45)
    print(f"{'FINAL SUMMARY':^45}")
    print("-" * 45)
    print(f" Repositories Processed : {repos_processed}")
    print(f" Total Space Reclaimed  : {format_size(total_saved_bytes)}")
    print(f" Total Files Removed    : {total_files_removed}")
    print("="*45)

if __name__ == "__main__":
    main()

[...]

[433] Optimizing: /home/uli/dev/FlareDNS
    Files: 73 -> 33 (40 removed)
    Size:  70.14 KB -> 47.03 KB (23.10 KB saved)

=============================================
                FINAL SUMMARY
---------------------------------------------
 Repositories Processed : 433
 Total Space Reclaimed  : 238.14 MB
 Total Files Removed    : 21612
=============================================