How to understand this Python functional style code by Peter Norvig?

Question

While reading Peter Norvig's Python IAQ, I came across this code snippet:

def _if(test):
    return lambda alternative: \
               lambda result: \
                   [delay(result), delay(alternative)][not not test]()
def delay(f):
    if callable(f): return f
    else: return lambda: f
fact = lambda n: _if (n <= 1) (1) (lambda: n * fact(n-1))
fact(100)

I searched this in the internet and this code appeared in several forums but it seems that those who commented on it all understand how it works.

I am quite new to functional programming concepts. I know that if test is evaluated to True, delay(alternative) will be selected. But in fact, if test is true, result is returned. This seems counter-intuitive to me.

Answer 1

Let's see:

• _if(True) is called, and immediately returns a lambda with the alternative parameter
• The returned lambda is called with alternative set to 1 and returns the result lambda
• The result lambda is called with result set to lambda: n * fact(n-1)
• not not True evaluates to 1 (This example is from python 2.4 era!), which indexes the second list item, which is delay(alternative)
• alternative was set to 1 earlier
• delay(1) is called, which returns lambda: 1
• lambda: 1 is called, it returns 1.

TL/DR: 1 is the alternative.

 

Named functions version:

def _if(test):
    def then_closure(expr_if_true):
        def else_closure(expr_if_false):
            if test:
                delayed = delay(expr_if_true)
            else:
                delayed = delay(expr_if_false)
            return delayed()
        return else_closure
    return then_closure