why is racket code returning void : Sudoku - racket

(define (solver row col board boolboard) ;solve the game
(if (equal? row 9)
board
(if (equal? (get-value boolboard row col) #t)
(begin
(for ([ite '(1 2 3 4 5 6 7 8 9)]) ;true
(when (equal? (check-available row col ite board) #t) ;should be careful here
(begin
(let ([new-board (set-value board row col ite)])
(if (equal? col 8)
(solver (+ 1 row) 0 new-board boolboard)
(solver row (+ 1 col) new-board boolboard)))
)
)
))
(if (and (equal? col 8) (equal? row 8)) ;false
board
(if (equal? col 8)
(solver (+ 1 row) 0 board boolboard)
(solver row (+ 1 col) board boolboard)
)))
)
)
This is the soduku solver. I've checked that the solver finally reached (if (equal? row 9) so it gets the board. But the function still returns #. I dont know why it is not returning the board.
Anyone could help me with that?

A general rule for fixing these types of bugs:
If it's returning something you don't expect (here, #<void>), then look for the cases in your function that could return that. You determined that the base case returns what you expect, but you have to check the other cases too.
The next case has a begin with a for inside of it. The for form always returns #<void> no matter what the body expression returned on any iteration.
So there's your case where it's returning #<void>. The for form evaluates the body of the loop on each iteration, but ignores the return values.
Because of this, for is pretty much never what you want. Normally forms like for/list, for/or, and for/fold are much more useful. In this case, for/first looks like it could be what you meant.
If you translate it blindly,
(for/first ([ite '(1 2 3 4 5 6 7 8 9)])
(when (equal? (check-available row col ite board) #t)
(let ([new-board (set-value board row col ite)])
(if (equal? col 8)
(solver (+ 1 row) 0 new-board boolboard)
(solver row (+ 1 col) new-board boolboard)))))
Then this still doesn't work because it still returns #<void> a lot of the time, since the when form returns #<void> when the condition is false. The for/first form isn't magic. It doesn't see that the when form fails or use that information to go to the next clause. To make for/first check the condition and go on to the next clause if it's false, use a #:when clause after the [ite ...].
(for/first ([ite '(1 2 3 4 5 6 7 8 9)]
#:when (equal? (check-available row col ite board) #t))
(let ([new-board (set-value board row col ite)])
(if (equal? col 8)
(solver (+ 1 row) 0 new-board boolboard)
(solver row (+ 1 col) new-board boolboard))))
The for/first form will evaluate the body for the first iteration where the #:when condition is true, and return that.
However, if the #:when condition is never true for any iteration, the for/first form will return false, so that will cause your solver function to return false.
But that means that the recursive calls within the for/first form could return false. If that happens, what should it do? You probably want to go to the next iteration of the loop. For that, you can use for/or, which is similar to for/first, except that it goes on to the next iteration if the body returns false.
(for/or ([ite '(1 2 3 4 5 6 7 8 9)]
#:when (equal? (check-available row col ite board) #t))
(let ([new-board (set-value board row col ite)])
(if (equal? col 8)
(solver (+ 1 row) 0 new-board boolboard)
(solver row (+ 1 col) new-board boolboard))))
That way, if filling in one number doesn't end up yielding a solution and it returns false, it can try the next number.
This solves your question about it returning #<void>. If there are more problems with your code, you should ask another question.

Related

How do I add each element into a list?

So Im' trying to print out a list of numbers but instead of printing all the numbers only the last number is put into the list. I know that means that the other numbers are not being inserted but I dont know how to fix this problem. I tried moving my cons around the function but nothign changes.
I'm new to racket so if Im not understanding a concept, please let me know.
should look like:
(gen-list 1 4)
output:
'(1 2 3 4)
(define (gen-list start end)
(if (> start end)
'()
(cond ((<= start end)
(gen-list start (- end 1))
)))
(cons end '()))
I see you would like to use recursion, so I would suggest to use printf to understand what really happen.
The idea is quite simple, you have start and end, what you want to have is the following:
Step 1: start=1 end=4
Step 2: start=2 end=4
Step 3: start=3 end=4
Step 4: start=4 end=4
Step 5: start=5 end=4, here we stop the recursion
So you need to increment start at each step of the recursion when the condition is met (<= start end).
(define (gen-list start end)
(printf "CALL ~A ~A~%" start end)
(cond
((<= start end)
(cons start
(gen-list (+ start 1) end)))
(else '())))
(gen-list 1 4)
Will return:
CALL 1 4
CALL 2 4
CALL 3 4
CALL 4 4
CALL 5 4
'(1 2 3 4)
EDIT: if you concern is to write programs, you could also use built-in functions, which are ready for use:
> (range 1 5)
(1 2 3 4)
> (sequence->list (in-inclusive-range 1 4))
'(1 2 3 4)
> (build-list 4 (lambda (x) x))
'(0 1 2 3)
https://docs.racket-lang.org/reference/sequences.html

How can I write a MEMBER function using the DO macro in Common Lisp?

I am trying to make a function which works like MEMBER function in Common Lisp.
So, I want to make this function work like this :
(my-member 2 '(1 4 5 5 3 2 5 6 9))
=> (2 5 6 9)
This is exactly same as how MEMBER function works.;
(member 2 '(1 4 5 5 3 2 5 6 9))
=> (2 5 6 9)
The condition is that I should use 'DO' macro to make this function.
This is my code that I have written to make this function :
(defun my-member (item x)
"This function works like MEMBER function."
(do ((z x (rest z))
(e (first x) (first z)))
(:when (equal item (first z))
(return z))))
But it doesn't work..
(my-member 2 '(3 4 5 2 1 1))
-> (3 4 5 2 1 1)
What should I do to solve this problem?
Here is the correct way to use do:
(do ((var 0 (1+ var))
(lst '() (cons var lst)))
((= var 5) lst)))
; ==> (4 3 2 1 0)
So both var and lst are variables initialised to 0 and () and after each iteration the variable is set to (1+ var) and (cons var lst).
What determines when it should stop is (= var 5) becoming not nil and when that happens the result of the whole do form is lst. This is the second part of the do and the last this has since I do not supply a body.
You can make an equivalent of the member function with using only one variable and a second part with an end condition and what should be the result of the do. Good luck!

Racket function that returns all numbers whose sum is <= given number?

I'm taking an intro to computer science course and one question needs me to write a function that takes a list of numbers and a number and returns the numbers in the list whose sum is less than the given number. I've written the function signature, definition, and check-expects, but I'm stuck. The function needs to assume intermediate student with lambda. I don't want any direct answers here; just help so that I can reach the answer myself.
I know it needs to use recursion. Perhaps a helper function would be needed.
;; sum-up-to: lon, number -> lon
;; consumes a list of numbers and a number and
;; returns the numbers in the list whose sum is
;; less than or equal to the given number
(define the-numbers (list 1 2 3 4 5 6 7 8 9))
(check-expect (sum-up-to the-numbers 7) (list 1 2 3))
(check-expect (sum-up-to the-numbers 18) (list 1 2 3 4 5))
(check-expect (sum-up-to the-numbers 45) the-numbers)
This problem can be simplified if we sort the list first and if we define a helper function that keeps track of the accumulated sum. Here's a skeleton, fill-in the blanks with the missing expressions and you'll have the solution:
(define (sum-up-to lst n)
(helper <???> n 0)) ; sort the input list, pass it to the helper
(define (helper lst n sum)
(cond (<???> '()) ; if the list is empty, end the recursion
((> <???> n) '()) ; also end recursion if sum + current element > n
(else
(cons <???> ; otherwise cons current element
(helper <???> ; advance recursion over list
n
(+ <???> <???>)))))) ; update sum
Following recursive method keeps adding numbers from the list sequentially to an initially empty outlist, till the sum is reached:
(define the-numbers (list 1 2 3 4 5 6 7 8 9))
(define (f lst sum)
(let loop ((lst lst)
(ol '()))
(if (or (..ENTER CONDITION FOR EMPTY LIST..)
(..ENTER CONDITION WHEN SUM IS REACHED..)
(..ENTER HOW TO PUT THE NEW LIST OUT..)
(loop (..ENTER ARGUMENTS TO BE SENT TO NEXT LOOP..)
))))
(f the-numbers 7)
(f the-numbers 18)
(f the-numbers 45)
Output:
'(1 2 3)
'(1 2 3 4 5)
'(1 2 3 4 5 6 7 8 9)

Writing the Foo Function In LISP With the following Specification

I am struggling to find the right approach to solve the following function
(FOO #'– '(1 2 3 4 5))
=> ((–1 2 3 4 5) (1 –2 3 4 5) (1 2 –3 4 5) (1 2 3 –4 5) (1 2 3 4 –5))
The first Parameter to the foo function is supposed to be a function "-" that has to be applied to each element returning a list of list as shown above. I am not sure as to what approach I can take to create this function. I thought of recursion but not sure how I will preserve the list in each call and what kind of base criteria would I have. Any help would be appreciated. I cannot use loops as this is functional programming.
It's a pity you cannot use loop because this could be elegantly solved like so:
(defun foo (fctn lst)
(loop
for n from 0 below (length lst) ; outer
collect (loop
for elt in lst ; inner
for i from 0
collect (if (= i n) (funcall fctn elt) elt))))
So we've got an outer loop that increments n from 0 to (length lst) excluded, and an inner loop that will copy verbatim the list except for element n where fctn is applied:
CL-USER> (foo #'- '(1 2 3 4 5))
((-1 2 3 4 5) (1 -2 3 4 5) (1 2 -3 4 5) (1 2 3 -4 5) (1 2 3 4 -5))
Replacing loop by recursion means creating local functions by using labels that replace the inner and the outer loop, for example:
(defun foo (fctn lst)
(let ((len (length lst)))
(labels
((inner (lst n &optional (i 0))
(unless (= i len)
(cons (if (= i n) (funcall fctn (car lst)) (car lst))
(inner (cdr lst) n (1+ i)))))
(outer (&optional (i 0))
(unless (= i len)
(cons (inner lst i) (outer (1+ i))))))
(outer))))
Part of the implementation strategy that you choose here will depend on whether you want to support structure sharing or not. Some of the answers have provided solutions where you get completely new lists, which may be what you want. If you want to actually share some of the common structure, you can do that too, with a solution like this. (Note: I'm using first/rest/list* in preference to car/car/cons, since we're working with lists, not arbitrary trees.)
(defun foo (operation list)
(labels ((foo% (left right result)
(if (endp right)
(nreverse result)
(let* ((x (first right))
(ox (funcall operation x)))
(foo% (list* x left)
(rest right)
(list* (revappend left
(list* ox (rest right)))
result))))))
(foo% '() list '())))
The idea is to walk down list once, keeping track of the left side (in reverse) and the right side as we've gone through them, so we get as left and right:
() (1 2 3 4)
(1) (2 3 4)
(2 1) (3 4)
(3 2 1) (4)
(4 3 2 1) ()
At each step but the last, we take the the first element from the right side, apply the operation, and create a new list use revappend with the left, the result of the operation, and the rest of right. The results from all those operations are accumulated in result (in reverse order). At the end, we simply return result, reversed. We can check that this has the right result, along with observing the structure sharing:
CL-USER> (foo '- '(1 2 3 4 5))
((-1 2 3 4 5) (1 -2 3 4 5) (1 2 -3 4 5) (1 2 3 -4 5) (1 2 3 4 -5))
By setting *print-circle* to true, we can see the structure sharing:
CL-USER> (setf *print-circle* t)
T
CL-USER> (let ((l '(1 2 3 4 5)))
(list l (foo '- l)))
((1 . #1=(2 . #2=(3 . #3=(4 . #4=(5))))) ; input L
((-1 . #1#)
(1 -2 . #2#)
(1 2 -3 . #3#)
(1 2 3 -4 . #4#)
(1 2 3 4 -5)))
Each list in the output shares as much structure with the original input list as possible.
I find it easier, conceptually, to write some of these kind of functions recursively, using labels, but Common Lisp doesn't guarantee tail call optimization, so it's worth writing this iteratively, too. Here's one way that could be done:
(defun phoo (operation list)
(do ((left '())
(right list)
(result '()))
((endp right)
(nreverse result))
(let* ((x (pop right))
(ox (funcall operation x)))
(push (revappend left (list* ox right)) result)
(push x left))))
The base case of a recursion can be determined by asking yourself "When do I want to stop?".
As an example, when I want to compute the sum of an integer and all positive integers below it, I can do this recusively with a base case determined by answering "When do I want to stop?" with "When the value I might add in is zero.":
(defun sumdown (val)
(if (zerop val)
0
(+ (sumdown (1- val)) val)))
With regard to 'preserve the list in each call', rather than trying to preserve anything I would just build up a result as you go along. Using the 'sumdown' example, this can be done in various ways that are all fundamentally the same approach.
The approach is to have an auxiliary function with a result argument that lets you build up a result as you recurse, and a function that is intended for the user to call, which calls the auxiliary function:
(defun sumdown1-aux (val result)
(if (zerop val)
result
(sumdown1-aux (1- val) (+ val result))))
(defun sumdown1 (val)
(sumdown1-aux val 0))
You can combine the auxiliary function and the function intended to be called by the user by using optional arguments:
(defun sumdown2 (val &optional (result 0))
(if (zerop val)
result
(sumdown2 (1- val) (+ val result))))
You can hide the fact that an auxiliary function is being used by locally binding it within the function the user would call:
(defun sumdown3 (val)
(labels ((sumdown3-aux (val result)
(if (zerop val)
result
(sumdown3-aux (1- val) (+ val result)))))
(sumdown3-aux val 0)))
A recursive solution to your problem can be implemented by answering the question "When do I want to stop when I want to operate on every element of a list?" to determine the base case, and building up a result list-of-lists (instead of adding as in the example) as you recurse. Breaking the problem into smaller pieces will help - "Make a copy of the original list with the nth element replaced by the result of calling the function on that element" can be considered a subproblem, so you might want to write a function that does that first, then use that function to write a function that solves the whole problem. It will be easier if you are allowed to use functions like mapcar and substitute or substitute-if, but if you are not, then you can write equivalents yourself out of what you are allowed to use.

calculate sum of multiples of 3 & 5 below 1000

I have written the following program to calculate the sum of all multiples of 3 & 5 below 1000 in scheme. However, it gives me an incorrect output.
Any help would be much appreciated.
(define (multiples)
(define (calc a sum ctr cir)
(cond (> a 1000) (sum)
(= ctr 7) (calc (+ a (list-ref cir 0)) (+ sum a) 0 (list 3 2 1 3 1 2 3))
(else (calc (+ a (list-ref cir ctr)) (+ sum a) (+ 1 ctr) (list 3 2 1 3 1 2 3)))))
(calc 0 0 0 (list 3 2 1 3 1 2 3)))
You can simply port imperative style solution to functional Scheme by using an accumulator(sum parameter) and a target parameter to test when to stop summing:
(define (multiples)
(define (multiples-iter num sum target)
(if (> num target)
sum
(multiples-iter (+ 1 num)
(if (or (zero? (mod num 3)) (zero? (mod num 5)))
(+ sum num)
sum)
target)))
(multiples-iter 0 0 1000))
Here's my (Racket-specific) solution, which doesn't involve lots of (or, for that matter, any) modulo calls, and is completely general (so that you don't need to construct the (3 2 1 3 1 2 3) list that the OP has):
(define (sum-of-multiples a b limit)
(define (sum-of-multiple x)
(for/fold ((sum 0))
((i (in-range 0 limit x)))
(+ sum i)))
(- (+ (sum-of-multiple a) (sum-of-multiple b))
(sum-of-multiple (lcm a b))))
Test run:
> (sum-of-multiples 3 5 1000)
233168
If you're using Racket, there's a very compact way to do what you ask, using looping constructs:
(for/fold ([sum 0])
([i (in-range 1 1000)]
#:when (or (zero? (modulo i 3)) (zero? (modulo i 5))))
(+ sum i))
=> 233168
One problem is that your code is missing a pair of parentheses around the cond clauses.
In the line (cond (> a 1000) (sum) the condition is just> while a and 1000 are interpreted as forms to be evaluated if > is true (which it is), and thus 1000 will be returned as the result.
Two other problem (masked by the first one) is that you are initializing ctr to 0 when it reaches 7, while it should be set to the next value, i.e. 1, and that you are including 1000 in the result.
The corrected version of your function is
(define (multiples)
(define (calc a sum ctr cir)
(cond ((>= a 1000) sum)
((= ctr 7) (calc (+ a (list-ref cir 0)) (+ sum a) 1 (list 3 2 1 3 1 2 3)))
(else (calc (+ a (list-ref cir ctr)) (+ sum a) (+ 1 ctr) (list 3 2 1 3 1 2 3)))))
(calc 0 0 0 (list 3 2 1 3 1 2 3)))
The same algorithm can also be defined as a non-recursive function like this:
(define (multiples)
(do ((cir (list 3 2 1 3 1 2 3))
(ctr 0 (+ ctr 1))
(a 0 (+ a (list-ref cir (modulo ctr 7))))
(sum 0 (+ sum a)))
((>= a 1000) sum)))
(require-extension (srfi 1))
(define (sum-mod-3-5 upto)
(define (%sum-mod-3-5 so-far generator-position steps)
(let ((next (car generator-position)))
(if (> (+ steps next) upto)
so-far
(%sum-mod-3-5 (+ so-far steps)
(cdr generator-position)
(+ steps next)))))
(%sum-mod-3-5 0 (circular-list 3 2 1 3 1 2 3) 0)) ; 233168
For this particular task, it will do on average half the operations then you would do if incrementing the counter by one, also, one less if condition to check.
Also, modulo (as being division in disguise, probably) is more expensive then summation.
EDIT: I'm not a pro on modular system in different dialects of Scheme. The SRFI-1 extension here is only required to make it easier to create a circular list. I couldn't find an analogue to Common Lisp (#0=(3 2 1 3 1 2 3) . #0#), but perhaps, someone more knowledgeable will correct this.
If you absolutely want to use the "repeating pattern" method, you could go about it something like this.
This uses recursion on the list of intervals rather than relying on list-ref and explicit indexing.
(define (mults limit)
(define steps '(3 2 1 3 1 2 3))
(define (mults-help a sum ls)
(cond ((>= a limit) sum)
((null? ls) (mults-help a sum steps))
(else (mults-help (+ a (car ls))
(+ a sum)
(cdr ls)))))
(mults-help 0 0 steps))