Peter Norvig's Lisp in Python implemented in Go.

I didn't try to be Scheme spec compliant, I just wanted to see how hard this would be in a different programming language. I certainly wrote more Golang than Norvig wrote Python.

The fun is doing 3 languages all at once. Bearing in mind that I'm not a Lisp programmer, I got to write an interpreter for a Lisp-style language in Go, while understanding the original interpreter in Python. Golang's type system was probably the major barrier.

This set of Go types seems to work. I'm not sure it's any better than using a list of interface{} as the List type.

type Node interface {
	isNode()
}
type List []Node
func (l List) isNode() { }

type I int
func (i I) isNode() { }
type F float64
func (f F) isNode() { }
type S string
func (s S) isNode() { }
type Fn func(Node)(Node)
func (fn Fn) isNode() { }
type B bool
func (b B) isNode() { }
type Pr struct {
    body Node
    params Node
    env *Environment
}
func (pr Pr) isNode() { }

It took me a few minutes, and a good bit of googling, but the Golang interface means that I didn't have to do anything dodgy to get a Node type that could have a []Node as one of its sub-types.

Norvig's Python code takes good advantage of Python magic like __call__() methods to make executing a lambda expression look exactly the same as executing a built-in primitive like '*'. Very clever.

Beyond just typing in simple forms to see if the pieces work, I decided to quit when it could do two things:

1. Calculate factorial
2. Execute a self-replicating expression.

It can run two variations on factorial:

(define fact (lambda (x) (if (== x 1) 1 (* x (fact (- x 1))))))
(fact 5)
(define fact2 (lambda (y x) (if (== x 1) 1 (* x (y y (- x 1))))))
(fact2 fact2 5)

And it runs a self-replicating program:

((lambda (y) (cons y y)) (lambda (y) (cons y y)))

This isn't exactly the same as the "canonical" Scheme self-replicating program, which has list where I have cons. I'm not sure I understand the subtle differences between lists made with the two function.