recursively creates git objects (trees, blobs) mirroring a directory in disk

create-object.py

import os, stat
from contextlib import contextmanager
from subprocess import run

@contextmanager
def fd_context(fd: int):
    try:
        yield fd
    finally:
        os.close(fd)

run_git = lambda *args, **kwargs: \
    run(['git', *args], **kwargs, check=True, capture_output=True) \
        .stdout.decode('ascii').strip()

def write_object(fd: int) -> tuple[str, str, str]:
    ''' recursively mirrors a file (or directory) as a git object, returning a (mode, type, object_id) tuple '''
    st = os.fstat(fd)
    if stat.S_ISREG(st.st_mode):
        return '100644', 'blob', run_git('hash-object', '-w', '--no-filters', '--stdin', stdin=fd)
    if stat.S_ISDIR(st.st_mode):
        tree_desc = bytearray()
        for fname in map(os.fsencode, os.listdir(fd)):
            with fd_context(os.open(fname, os.O_RDONLY, dir_fd=fd)) as cfd:
                entry_head = ' '.join(write_object(cfd)) + '\t'
            tree_desc += entry_head.encode('ascii') + fname + b'\0'
        return '040000', 'tree', run_git('mktree', '-z', input=tree_desc)
    raise AssertionError(f'type {stat.S_IFMT(st.st_mode)} unsupported')

# FIXME: add support for symlinks (right now it errors out)
# FIXME: preserve mode
# FIXME: improve performance by using batch mode
# FIXME: ignore ENOENT errors when opening child (TOCTOU)