iter_path.py

def list_files_at_revision(path: str, revision: str) -> List[str]:

    import subprocess

    # Check if the path is a file or a directory

    # We can use the `git cat-file` command to check if the path is a file

    # The `-e` flag indicates that we want to check if the object exists

    # If the object exists, then the command will return 0, otherwise it will return a non-zero value

    result = subprocess.run(["git", "cat-file", "-e", f"{revision}:{path}"])

    if result.returncode == 0:

        # The path is a file, so we can use the `git cat-file` command to retrieve its contents

        # The `-p` flag indicates that we want to pretty-print the object's contents

        result = subprocess.run(["git", "cat-file", "-p", f"{revision}:{path}"], capture_output=True)

        output = result.stdout

        # The output is the contents of the file, so we can return it as a list with a single element

        return [output.decode("utf-8")]

    else:

        # The path is a directory, so we can use the `git ls-tree` command to list its contents

        # The `-r` flag indicates that we want to list the files recursively

        # The `-t` flag indicates that we want to include the file's type (e.g. "blob" for regular files)

        # The `-z` flag indicates that we want the output to be separated by null characters, rather than newlines

        # This allows us to use the `split()` method to split the output into individual lines, even if the filenames contain newlines

        result = subprocess.run(["git", "ls-tree", "-r", "-t", "-z", revision, path], capture_output=True)

        output = result.stdout

        # Split the output into individual lines

        lines = output.split(b"\0")

        # The first line is a header, so we can ignore it

        files = []

        for line in lines[1:]:

            # Each line has the following format:

            # <mode> SP <type> SP <sha1> TAB <file>

            # We want to extract the file name, which is the last part of the line

            # So we can split the line on the TAB character and take the last part

            file = line.split(b"\t")[-1].decode("utf-8")

            files.append(file)

        return files