This is a good use case for define-modify-macro
(which has also been described in what is to append as push is to cons, in Lisp?, but the present case is simpler). First, write your bounded sum as a function. This is pretty straightforward; it takes val
and delta
and returns 1.0
if their sum is greater than 1.0
, and their sum otherwise. Based on the pseudo code and Lisp code you posted, this could be:
(defun sum-bounded (val delta)
(if (>= (+ val delta) 1.0)
1.0
(+ val delta)))
Actually, for just computing this value, you can use:
(defun sum-bounded (val delta)
(min 1.0 (+ val delta)))
Now you use define-modify-macro
to define a macro incf-bounded
:
(define-modify-macro incf-bounded (delta) sum-bounded)
The macro takes a place as its first argument and delta as a second. It safely retrieves the value from the place, computes sum-bounded
with that value and delta, and then stores the result back into the place. “Safely” here means that it avoids possible problems with multiple evaluation, as Lars Brinkhoff's wisely warns against. Then you just use it:
(let ((x .5))
(incf-bounded x .3)
(print x) ; prints 0.8
(incf-bounded x .3)
(print x)) ; prints 1.0 (not 1.1)
For more complicated cases where the place that would be modified isn't naturally the first argument to the macro that you want, you'd need to write your own macro and use get-setf-expansion
, but this is explained in more detail in
Code all together for easy copy & paste
(defun sum-bounded (val delta)
"Returns the lesser of 1.0 or the sum of val and delta."
(min 1.0 (+ val delta)))
(define-modify-macro incf-bounded (delta) sum-bounded
"(incf-bounded place delta) computes the sum of the value of the
place and delta, and assigns the lesser of 1.0 and the sum of the value
and delta to place.")
(defun demo ()
(let ((x .5))
(incf-bounded x .3)
(print x) ; prints 0.8
(incf-bounded x .3)
(print x))) ; prints 1.0 (not 1.1)
define-modify-macro
, which is what you'd probably want to use here. – Wightelse
line beelse val := val + delta
? – Wight