lambdazen

Syntax changes for python lambdas.


License
MIT
Install
pip install lambdazen==0.1.7

Documentation

A Better Lambda

PyPI version Build Status Supports Python 2.6 - 3.5

What is this?

A better python lambda syntax for your anonymous function needs.

Write a = (x) > x instead of a = lambda x: x. See below for syntax caveats.

Get started immediately: pip install lambdazen

from lambdazen import zen
def otherfunc(*args):
    print sum(args)

# The zen decorator allows you to define lambdas with a better syntax
@zen
def example():
    example.epic = (x, y, z) > otherfunc(x, y, z)

    # Multiline lambdas are a tuple or list of statements
    # The assignment operator inside is << instead of =
    # The last statement is the return value
    example.multiline = (x, y, z) > (
        s << otherfunc(x, y, z),
        s
    )

# Call function so the lambdas are bound to function attributes
example()

example.epic(1,2,3)
>>> 6

example.multiline(1,2,3)
>>> 6

Caveats

  • better lambdas can only be defined in a function with the @zen attribute
  • any other code in this function will be executed, it's best to use the function as a container of lambdas

How does it work

Read the story

TLDR; Runtime in-memory source rewriting and recompilation

Additional Examples

from lambdazen import zen

# Lambdas don't need to be bound to the function
@zen
def normalizeString(nS):
    transforms = [
        (s) > s.strip(),
        (s) > s.lower(),
        (s) > s.replace(' ', '_')]

    apply_all = (transforms_list, s) > (
        is_done << (len(transforms_list) == 0),
        current_transform << transforms_list[0] if not is_done else None,
        remaining_transforms << transforms_list[1:] if not is_done else None,
        current_transform(apply_all(remaining_transforms, s)) if not is_done else s)

    return apply_all(transforms, nS)

normalizeString("Abraham Lincoln")
>>> "abraham_lincoln"