afire 1.0.0

python fire cli with type hint support

Python Afire

Fork from python-fire.

Python AFire is a library for automatically generating command line interfaces (CLIs) from absolutely any Python object.

• Python Afire is a simple way to create a CLI in Python. [1]
• Python Afire is a helpful tool for developing and debugging Python code. [2]
• Python Afire helps with exploring existing code or turning other people's code into a CLI. [3]
• Python Afire makes transitioning between Bash and Python easier. [4]
• Python Afire makes using a Python REPL easier by setting up the REPL with the modules and variables you'll need already imported and created. [5]
• Python Afire fully support type hint.

Installation

To install Python Afire with pip, run: pip install afire

To install Python Afire with conda, run: conda install afire -c conda-forge

To install Python Afire from source, first clone the repository and then run: poetry install

Basic Usage

You can call Fire on any Python object:
functions, classes, modules, objects, dictionaries, lists, tuples, etc. They all work!

Here's an example of calling Fire on a function with type hint, it will automatically recognize and convert types according to your hype hint.

import afire
from pathlib import Path

def hello(path: Path):
  assert isinstance(path, Path)

if __name__ == '__main__':
  afire.Fire(hello)

Then, from the command line, you can run:

python hello.py --path=/root  # No error
python hello.py --help  # Shows usage information.

Here's an example of calling Fire on a function with nested type hint.

import afire
from typing import Dict, Union, Set

def test(a: Union[Dict[str, int], Set[bytes]]):
  # check types
  assert isinstance(a, (Dict, Set))

  # check types in dict or set
  if isinstance(a, Dict):
    for k, v in a.items():
      assert isinstance(k, str)
      assert isinstance(v, int)
  else:
    for i in a:
      assert isinstance(i, bytes)
  print(a)

if __name__ == '__main__':
  afire.Fire(test)

Then, from the command line, you can run:

# dict type
python test.py --a='{1: 2}'  # {'1': 2}
# or use position arg
python test.py '{1: 2}'  # {'1': 2}

# set type
python test.py --a='{a, b, c}'  # {b'a', b'b', b'c'}

Type conversion rules

Currently support input types:

type	example
str	"a"
int	1
bytes	b"a"
List	[x, y, z]
Dict	[x: y]
Set	{x, y, z}
Tuple	(x, y, z)

Rule

	str	int	bytes	<- input
str	*	*	*
int	can be converted to int	*	x
bytes	*(utf8)	*(length 8, big order)	*
datetime/date	format: YYYY-MM-DD-HH:MM:SS	x	x
	format: YYYY-MM-DD HH:MM:SS
	format: YYYY/MM/DD HH:MM:SS
	format: YYYYMMDDHHMMSS
	format: YYYY/MM/DD
	format: YYYY-MM-DD
any type with one parameter	if support	if support	if support	if support
e.g. Path, float
^ type hint

*: any kind of input will convert

x: not support to convert

License

Licensed under the Apache 2.0 License.

Disclaimer

This is not an official Google product.