生成Haskell表达式的逐步减少

Question

是否有一个系统可以逐步减少Haskell表达式？目的是提供延迟评估的教程示例。如果它是用Haskell编写的，那会很好，但这不是必需的。

例如，假设我们有：

myCycle xs = xs `myAppend` myCycle xs

[] `myAppend` ys = ys
(x:xs) `myAppend` ys = x: (xs `myAppend` ys)

myTake 0 _  = []
myTake _ [] = []
myTake n (x:xs) = x: (myTake (n-1) xs)

> myTake 3 $ myCycle [1, 2]的评估应该显示这样的序列。 （如果我在这方面犯了任何错误，请纠正我。）

>  myTake 3 $ myCycle [1, 2]
=> myTake 3 ([1, 2] `myAppend` myCycle [1, 2])
=> myTake 3 (1:([2] `myAppend` myCycle [1, 2]))
=> 1: myTake 2 ([2] `myAppend` myCycle [1, 2]) -- Not showing 3-1 => 2
=> 1: myTake 2 (2:([] `myAppend` myCycle [1, 2]))
=> 1:2: myTake 1 ([] `myAppend` myCycle [1, 2])
=> 1:2: myTake 1 (myCycle [1, 2])
=> 1:2: myTake 1 ([1, 2] `myAppend` myCycle [1, 2])
=> 1:2: myTake 1 (1:([2] `myAppend` myCycle [1, 2]))
=> 1:2:1: myTake 0 ([2] `myAppend` myCycle [1, 2])
=> 1:2:1:[] = [1, 2, 1]

所有这一切的要点是说明懒惰的评估如何使无限列表成为可能。