clifun

Because cli's should be fun(ctions) ;).

import clifun

def my_program(a: int, b: str = "not provided"):
  print(f"Running some code with: a={a}, b={b}")

if __name__ == "__main__":
  clifun.call(my_program)

That's all it takes. Clifun will inspect your function and collect the values it needs it from command line arguments, environment variables, or config files, and then call it. It does all of this using your type annotations, so it will fail if you have any arguments without a type annotation.

python examples/function --a 1

Running some code with: a=1, b=not provided

You can even run functions in any module without modifying the module at all

python clifun.py examples/module.py my_program --a 1

if you have the code checked out. Or if you pip installed

python -m clifun some_module.py function ...

Or if you have environment variables defined

export A=1
export B=hi
python example.py

again yields without you having to provide values

Basic(a=1, b='hi')

clifun also supports nested objects (or functions taking complex objects as inputs)

import datetime as dt
from dataclasses import dataclass
from typing import Optional


import clifun


@dataclass(frozen=True)
class Foo:
    a: dt.datetime
    b: Optional[str] = None


@dataclass(frozen=True)
class Bar:
    f: Foo
    c: int


if __name__ == "__main__":
    bar = clifun.call(Bar)
    print(bar)

You specify values for the fields in the nested class by referring to them with a their field name in the outer class

python examples/advanced.py --c 1 --f.a 2020-01-01 --f.b hi

Bar(f=Foo(a=datetime.datetime(2021, 1, 1, 0, 0), b='hi'), c=1)

You can also supply one or more json formatted config files. Provide the name(s) of these files as positional arguments. clifun will search them, last file first, for any keys fields that are not provided at the command line before searching the environment.

python examples/advanced.py --c 1 examples/foo.json

Bar(f=Foo(a=datetime.datetime(2021, 1, 1, 0, 0), b='str'), c=1)

clifun is inspired by clout, but I wanted to try being a bit more opinionated to make both the library and code using it simpler.