Functions, Iterators,

and Generators

The core of functional programming is the use of pure functions to map values from
the input domain to the output range. A pure function has no side effects, a relatively
easy threshold for us to achieve in Python.

Avoiding side effects also means reducing our dependence on variable assignment
to maintain the state of our computations. We can't purge the assignment statement
from the Python language, but we can reduce our dependence on stateful objects.
This means we need to choose among the available Python built-in data structures
to select those that don't require stateful operations.

This chapter will present several Python features from a functional viewpoint,
as follows:

• Pure Functions, free of side effects


• Functions as objects that can be passed as arguments or returned as results


• The use of Python strings using object-oriented suffix notation and
  prefix notation


• Using tuples and namedtuples as a way to create stateless objects


• Using iterable collections as our primary design tool for
  functional programming

We'll look at generators and generator expressions, since these are ways to work with
collections of objects. As we noted in Chapter 2, Introducing Some Functional Features,
there are some boundary issues while trying to replace all generator expressions with
recursions. Python imposes a recursion limit, and doesn't automatically handle TCO:
we must optimize recursions manually using a generator expression.