Accumulator / Stateful / async functions

[dberardo]

I am evaluating the usage of this library to perform simple simulations based on different type of functions.

Before I start contributing to the code myself, i have a couple of questions:

• Can asynchronous functions be seamlessly implemented using the library (e.g. i assume that the google.function is doing exactly that, right, as in calling an external API and returning a value)
• is this library already handling exceptions generating within the workflow?
• Would it be possible to implement rolling statistics (like rolling means, etc.) And also various Stateful operators like Integrals, etc. with this library,? Where would be the best place to state store? Should it be a "virtual cell" or is there a place in memory for that?

And also: Can input arguments be given in a more flexible way? Can we use dictionaries/kwargs, where order of arguments doesn't metter, and. Can we also pass null values to excel-like functions?

[dberardo-com]

also: would you endorse usage of caching libraries to speed up the computation of custom formulas ? or would it inerhently clash with the nature of the schedula library ?

example:

from cachetools import cached

shared_cache = dict()

@cached(
    cache=shared_cache,
    key=lambda a, b, c=None: "a" if c is None else "b"
)
def foo(a, b, c=None):
    print("foo")

import formulas
FUNCTIONS = formulas.get_functions()
FUNCTIONS['MYFUNC'] = foo
func = formulas.Parser().ast('=MYFUNC(1, 2)')[1].compile()