Pydian partials: have nicer ways to work with lists

[ericpan64]

Problem

Working on a case where the result of a get is a list[tuple] where I want to do something to each tuple in the list while keeping the list structure, so the apply= parameter is kind of useless because it forces me to apply to the list first (which makes sense)

Can get around this

Requested feature

Maybe a foreach function? E.g.

from typing import Any
from pydian import get
import pydian.partials as p

def swap_tup_order(t: tuple[Any, Any]) -> tuple:
    return (t[1], t[0])

data = { 'list_tups': [(1, 2), (3, 4), (5, 6)] }
print(get(data, 'list_tups[::-1]', apply=p.foreach(swap_tup_order))) # Expect: `[(6, 5), (4, 3), (2, 1)]`

Alternatives considered

• Maybe just have people use list comprehensions (more "pythonic"), though could get hard for nested cases

Additional context

• From working with OpenAI Whisper model -- audio chunks are passed as tuple of timestamps and corresponding text

[ericpan64]

Another partial thing: for p.keep provide more intuitive handling for indices. E.g. I want to keep the last two -- isn't a way to do this now (in Python: some_list[-2:]). Maybe p.index makes more sense`