Clone large repository in depth

gitclone.py

import os
import sys
import logging
from git import Repo, GitCommandError

def extract_repo_name(repo_url):
    """
    Extracts the repository name from the given Git URL.
    Supports URLs ending with '.git' and those without it.
    """
    repo_name = os.path.splitext(os.path.basename(repo_url))[0]
    return repo_name

def is_shallow(repo_path):
    """
    Checks if the repository at the given path is shallow.
    """
    shallow_file = os.path.join(repo_path, '.git', 'shallow')
    return os.path.exists(shallow_file)

def main():
    if len(sys.argv) != 2:
        print("Usage: python script.py <repository_url>")
        sys.exit(1)

    # Enable GitPython command tracing
    os.environ['GIT_PYTHON_TRACE'] = 'full'
    logging.basicConfig(level=logging.INFO)

    repo_url = sys.argv[1]
    clone_dir = extract_repo_name(repo_url)

    # Initial depth
    depth = 1

    try:
        # Clone the repository with the initial depth
        print(f"Cloning {repo_url} into {clone_dir} with depth {depth}...")
        repo = Repo.clone_from(repo_url, clone_dir, depth=depth)

        # Loop to incrementally fetch more history
        while is_shallow(clone_dir):
            depth *= 2
            print(f"Increasing depth to {depth}...")
            repo.git.fetch(depth=depth)

        print("Repository is fully cloned.")

    except GitCommandError as e:
        print(f"An error occurred: {e}")

if __name__ == "__main__":
    main()