dtyper 1.2.1

⌨️dtyper: Call typer commands or make dataclasses from them ⌨

⌨️dtyper: Call typer commands, or make a dataclass from them ⌨️

typer is a famously easy and useful system for writing Python CLIs but it has two issues.

You cannot quite call the functions it creates directly.

And as you add more and more functionality into your CLI, there is no obvious way to break up the code sitting in one file.

---

dtyper is a drop-in replacement for typer, so you can write:

import dtyper as typer

It adds just two members, and overrides a third:

• dtyper.dataclass is a decorator that takes an existing typer command and makes a dataclass from it.

• dtyper.function is a decorator that takes a new typer command and returns a callable function with the correct defaults.

• dtyper.Typeris identical to typer.Typer, except that the command() decorator method wraps its functions with dtyper.function above so they can be called from regular Python code. You can think of it as as a fix to a bug in typer.Typer.command, if you like. :-)

dtyper.function filled a need several people had mentioned online, but I think developing with dtyper.dataclass is the way to go, particularly if you expect the code to grow medium-sized or beyond.

Installation

pip install dtyper

Examples

dtyper.dataclass: simple

@command(help='test')
def get_keys(
    bucket: str = Argument(
        ..., help='The bucket to use'
    ),

    keys: str = Argument(
        'keys', help='The keys to download'
    ),

    pid: Optional[int] = Option(
        None, help='pid'
    ),
):
    return GetKeys(**locals())()

@dtyper.dataclass(get_keys)
class GetKeys:
    site = 'https://www.some-websijt.nl'

    def __post_init(self):
        self.pid = self.pid or os.getpid()

    def __call__(self):
        return self.url, self.keys, self.pid

    def url(self):
       return f'{self.site}/{self.url}/{self.pid}'

dtyper.dataclass: A pattern for larger CLIs

# In interface.py

@command(help='test')
def compute_everything(
    bucket: str = Argument(
        ..., help='The bucket to use'
    ),
    # dozens of parameters here
):
    d = dict(locals())

    from .compute import ComputeEverything

    return ComputeEverything(**d)()

# In compute.py

from .interface import compute_everything

@dtyper.dataclass(compute_everything)
class ComputeEverything:
    def __call__(self):
       if self.huge_thing() and self.etc():
          self.more_stuff()

       # Dozens of methods here