I don't see why not. Feel free to make a pull request!

I think making _add_dataclass_options public is not the correct thing to do.

It is only one step in the process of initializing the dataclass. It configures the ArgumentParser assuming that the output will be passed into the dataclass for the final stage of initialization. This currently only means that the default values don't get set to the expected values after calling parse_args but later updates might bring other behaviors that don't make sense when it is public function.

in some of my code in order to be able to continue using normal argparse.ArgumentParser instances and use a dataclass to define only a subset of my script arguments.

I might not be fully understanding your use case but it seems like parse_known_args might work for what you are doing.

Example

import argparse

from dataclasses import dataclass
from argparse_dataclass import parse_known_args

@dataclass
class DataclassArgs:
    name: str
    clear: bool


def main():
    dc_args, remaining = parse_known_args(DataclassArgs)

    parser = argparse.ArgumentParser()
    parser.add_argument("--foo", action="store_true")
    args = parser.parse_args(remaining)

    print(dc_args)
    print(args)

if __name__ == "__main__":
    main()

usage

$ python ex_argparse_dataclass.py --name "Fred" --foo
DataclassArgs(name='Fred', clear=False)
Namespace(foo=True)