import poser, toolz

https://toolz.readthedocs.io/en/latest/composition.html

Be a poser

How is functional programming like legos. https://toolz.readthedocs.io/en/latest/composition.html#lego

poser is a API for lazy, (dys)functional python programming. It allows complex functions to be composed using fluent or symbollic interfaces.

pip install poser

dysfunctional programming === Functional programming with all the side effects.

New to [functional programming]? Functional programming uses declarative functions to compose complex operations on arguments. If you are familiar with python then [toolz][toolz] is a great starting point, [poser][poser] is a compact [API] for [toolz] and the python [standard library].

poser

poser API

from poser import λ, Λ, stars

λ is a composition that always returns a function. For example, below we create a list of numbers from 1 to a some value

>>> f = λ.range(1).list(); f
λ(<class 'list'>, functools.partial(<class 'range'>, 1))
>>> assert callable(f)
>>> f(5), f(10)
([1, 2, 3, 4], [1, 2, 3, 4, 5, 6, 7, 8, 9])

poser can use forward references to lazily import modules.

>>> g = λ.range(1).enumerate().dict()['pandas.Series']; g
λ(ForwardRef('pandas.Series'), <class 'dict'>, <class 'enumerate'>, functools.partial(<class 'range'>, 1))
>>> λ(9) + g + type + ...
<class 'pandas.core.series.Series'>

Λ is for object function composition where the function represents symbollic or chained methods

>>> (Λ*10+2)(1)
12
>>> assert (Λ*10+2)(1) == 1*10+2
>>> s = "A formatted string :{}: with a %s"
... (Λ.format('foo') % '% formatted').upper()(s)
'A FORMATTED STRING :FOO: WITH A % FORMATTED'

>>> assert (Λ.format('foo') % '% formatted').upper()(s)\
... == (s.format('foo') % '% formatted').upper()

The poser API expresses all of the symbols in the python data model. Refer to the tests for examples.

juxtaposition

>>> assert isinstance(λ[range, type, str](10), tuple)
>>> assert isinstance(λ[[range, type, str]](10), list)
>>> assert isinstance(λ[{range, type, str}](10), set)

Value and keyword functions can be supplied to juxtapose functions across dictionaries.

λ[{'a': range, type: str}](10)
{'a': range(0, 10), int: '10'}

Example recipes.

get = λ['functools.lru_cache']()(λ['requests.get'][λ[Λ.json()] ^ BaseException | Λ.text()]); get

read = λ**Λ.startswith('http') & get | λ.Path().read_text()