fondat/fondat

A foundation for Python resource-oriented applications

fondat

A foundation for resource-oriented backend applications.

Introduction

Fondat is a foundation for resource-oriented backend applications in Python.

Install

pip install fondat

Develop

poetry install
poetry run pre-commit install

Test

poetry run pytest

fondat's Issues

TKinter issue on latest version

On version 4.0a7 there's an from tkinter import N in validation module.
That import forces python to be configured for Tk.

8/22/2022, 1:32:06 PM	from fondat.validation import validate_arguments	*redacted*
8/22/2022, 1:32:06 PM	File "/opt/venv/lib/python3.10/site-packages/fondat/validation.py", line 5, in <module>	*redacted*
8/22/2022, 1:32:06 PM	import fondat.types	*redacted*
8/22/2022, 1:32:06 PM	File "/opt/venv/lib/python3.10/site-packages/fondat/types.py", line 11, in <module>	*redacted*
8/22/2022, 1:32:06 PM	from tkinter import N	*redacted*
8/22/2022, 1:32:06 PM	File "/usr/local/lib/python3.10/tkinter/__init__.py", line 37, in <module>	*redacted*
8/22/2022, 1:32:06 PM	import _tkinter # If this fails your Python may not be configured for Tk     *redacted*

load_datacls(dict of dicts)

I ran into fondat while looking for a way to create Resource APIs out of JSON Schema.

I wonder if it might actually be more efficient for me to define all my data classes as as pure YAML document, then create a 'load_datacls' wrapper that calls make_datacls

def make_datacls(
    cls_name: str,
    fields: Iterable[tuple[str, type] | tuple[str, type, dataclasses.Field]],
    init: bool = True,
    **kwargs,
) -> type:

The equivalent YAML would be something like:

cls_name:
   init: false
   fields:
     name: type
     name2: 
      type: int
      field: "default=5"
      description: this is a description
      optional: false
   keywords:
    namespace: "{'add_one': lambda self: self.x + 1}"

The reason, to be honest, is that I'm not sure we will always fondat (or even Python!) going forward, so I'd rather store data about our schemas as, well, data -- so we can generically reuse it with minimal parsing effort.

Trio support?

Alas, I ended up building my package around trio rather than asyncio. I realize this may be a heavy lift given your aiosqlite dependency, but is a trio option anywhere on your roadmap?

https://trio.readthedocs.io/en/stable/index.html

No documentation

Hello,
I have came across this interesting library, sadly I can't find any documentation on Google and Pypi documentation url is down (github 404). Can someone guide me to documentation/example usage?