This document will detail some of the more functional aspects of Common Lisp explained in the book “Land of Lisp”.
The reduce function in Common Lisp works like you would expect, it
takes a list and applies the function to the first two items, then
takes the result and applies it to the next item… So on and so
forth until we finally have a single value, our result.
(reduce #'+ '(1 2 3 4))10
But we have a few special things we can do with reduce. For
example, we can specify the initial value rather than have the first
item in the list be used instead, this is done with the
:initial-value keyword argument:
(reduce (lambda (x y) (if (> x y) x y)) '(1 2 3 4) :initial-value 5)5
This is desirable when the initial candidate is unfit as a result since it is used as the first “result” regardless of the code in the reduce function.
Another thing about reduce is that it is generic, so it works on
arrays, lists, and strings interchangeably. For example, here is a
reduce of a string:
(reduce (lambda (x y)
(concatenate 'string
x (prin1-to-string (char-int y)) " ")) "Pichu"
:initial-value "")80 105 99 104 117
That is inefficient though. And it has a trailing space… You would want to use map for this task. Speaking of which:
Common Lisp map is very similar to mapcar, except it is generic,
and accepts one more argument: An argument to tell it what kind of
data to return.
So, to write a more efficient version of the reduce above, we can:
(string-trim "()" (prin1-to-string (map 'list (lambda (x)
(char-int x)) "Goodra")))71 111 111 100 114 97