CSCI 316: Lisp Assignment 5

Lisp Assignment 5: Page 1 of 5
C SCI 316: Lisp Assignment 5
To be submitted no later than Tuesday, April 6.* [Note: If euclid unexpectedly goes down after 6 p.m.
on this due date, there will be no extension. Try to submit no later than noon that day, and much sooner
if you can.] Submissions after the due date will be accepted as late submissions until the late-submission
deadline, which will be announced later but will be in May. See p. 3 of the 1st-day announcements
document for information regarding late-submission penalties. See p. 5 for submission instructions.
In this document the term list should be understood to mean proper list.
Scheme solutions to the problems will be provided to you after the due date: Exam 1 will assume
students have studied the solutions.
SECTION 1 (Nonrecursive Preliminary Problems)
The three problems in this section (A – C) do not carry direct credit, but are intended to help
you solve problems 1 – 3 in Part I of Section 2. Note that there may be questions on exam 1 or
on the final exam that are of a similar nature to A – C.
Your solutions to problems A – C must not be recursive. You can test your solutions to
these three problems on venus or euclid: Functions INDEX, MIN-FIRST, and SSORT with the
stated properties are predefined for you when you start Lisp using cl on those machines.
A. INDEX is a function that is already defined on euclid and venus. If N is any positive integer and
 L is any list, then (INDEX N L) returns the Nth element of L if N does not exceed the length of
 L; if N exceeds the length of the list L, then (INDEX N L) returns the symbol ERROR. For
 example, (INDEX 3 '(A S (A S) (A) D)) = (A S) (INDEX 6 '(A S (A S) (A) D)) = ERROR
 Complete the following definition of a function MY-INDEX without making further calls
 of INDEX and without calling MY-INDEX recursively, in such a way that if N is any integer
 greater than 1 and L is any nonempty list then (MY-INDEX N L) is equal to (INDEX N L).
 (defun my-index (N L)
 (let ((X (index (- N 1) (cdr L))))
 __ ))
 [Note: You should not have to call any functions.]
B. MIN-FIRST is a function that is already defined on euclid and venus; if L is any nonempty list of
 real numbers then (MIN-FIRST L) is a list whose CAR is the minimum of the numbers in L and
 whose CDR is the list obtained when the first occurrence of that value is removed from L. (Thus
 (MIN-FIRST L) is "L with the first occurrence of its minimum value moved to the front".) For
 example, (MIN-FIRST '(0 3 1 1 0 3 5)) = (0 3 1 1 0 3 5)
 (MIN-FIRST '(4 3 1 1 0 3 5 0 3 0 2)) = (0 4 3 1 1 3 5 0 3 0 2).
 Complete the following definition of a function MY-MIN-FIRST without making further calls of
 MIN-FIRST and without calling MY-MIN-FIRST recursively, in such a way that if L is any list of
 at least two real numbers then (MY-MIN-FIRST L) is equal to (MIN-FIRST L).
 (defun my-min-first (L)
 (let ((X (min-first (cdr L))))
 ___________________________________
 ___________________________________ ))
 [There are two cases: (car L) may or may not be ≤ (car X).]
*If you have difficulty with these problems, you are encouraged to see me during my office hours. Questions about these
 problems that are e-mailed to me will not be answered until after the due date.
Lisp Assignment 5: Page 2 of 5
C. SSORT is a function that is already defined on euclid and venus; if L is any list of real numbers
 then (SSORT L) is a list of those numbers in ascending order. Complete the following definition
 of a function MY-SSORT without making further calls of SSORT and without calling
 MY-SSORT recursively, in such a way that if L is any nonempty list of real numbers then
 (MY-SSORT L) is equal to (SSORT L).
 (defun my-ssort (L)
 (let* ((L1 (min-first L))
 (X (ssort (cdr L1))))
 __________________________________ ))
SECTION 2 (Main Problems)
Your solutions to the following 16 problems will count a total of 2% towards your grade if the
grade is computed using rule A––the 10 problems in Part I will count 1%, and the 6 problems in
Part II will also count 1%. It is expected that your solutions to problems 1 – 3 will be based on
your solutions to A – C above. [On euclid and venus, when you load in your function definitions
for problems 1 – 3, you will replace the predefined functions with the same names with your
versions of those functions.]
Program in a functional style, without using SETF or iterative constructs such as DO.
Follow the indentation and spacing rules at:
 https://euclid.cs.qc.cuny.edu/316/indentation-and-spacing-guidelines-for-Lisp-Assignments.pdf
Use LET / LET* or appropriate helping functions to avoid evaluating computationally
expensive expressions more than once. Any helping functions you define for use in your
solutions must be defined in the .lsp file that you submit for grading.
When writing a function, decide what the valid values for each argument will be. The
statement of each problem specifies that certain argument values must be valid. For example, in
problem 5 any list of real numbers in ascending order (including the empty list) must be a valid
value for each of the arguments; you can choose to make other argument values valid too.
If f computes (f y ...) from (f (smaller y) ...) for certain values of y, your base case code
needs to ensure this never happens for any value of y that is valid (in the sense of the previous
paragraph) for that argument but which satisfies one of the following conditions:
 BC-1: (smaller y) is undefined.
 BC-2: (smaller y) is defined but is not a valid value for that argument of f.
 BC-3: (smaller y) is a valid value for that argument of f but is no smaller in size than y
 [e.g., the case y = NIL when (smaller y) is (cdr y), if NIL is a valid value for y].
PART I [Solutions to problems 1 – 10 will count 1% towards your grade if the grade is
 computed using rule A.]
1. Define a recursive function INDEX with the properties stated in problem A. Note that the first
 argument of INDEX may be 1, and that the second argument may be NIL.
2. Define a recursive function MIN-FIRST with the properties stated in problem B. Note that the
 argument of MIN-FIRST may be a list of length 1.
3. Define a recursive function SSORT with the properties stated in problem C. In the definition
 of SSORT you may call SSORT itself, MIN-FIRST, CONS, CAR, CDR and ENDP, but you
 should not call any other function. 
Lisp Assignment 5: Page 3 of 5
4. Use the function PARTITION from Lisp Assignment 4 to complete the following definition of a
 recursive function QSORT such that if L is a list of real numbers then (QSORT L) is a list of
 those numbers in ascending order. (DEFUN QSORT (L)
 (IF (ENDP L)
 NIL
 (LET ((PL (PARTITION (CDR L) (CAR L))))
 ... )))
5. Define a Lisp function MERGE-LISTS such that if each of L1 and L2 is a list of real numbers in
 ascending order then (MERGE-LISTS L1 L2) returns the list of numbers in ascending order that is
 obtained by merging L1 and L2. Your definition must not call any sorting function. Examples: (MERGE-LISTS '(2 4 5 5 7 8 9) '(3 4 6 9 9)) = (2 3 4 4 5 5 6 7 8 9 9 9)
 (MERGE-LISTS '(1 2 3) '(4 5 6 7)) = (1 2 3 4 5 6 7)
 (MERGE-LISTS '(3 4 5 6 7) '(0 1 2 3)) = (0 1 2 3 3 4 5 6 7)
 Hint: Consider the 4 cases L1 = ( ), L2 = ( ), (< (car L1) (car L2)), and (= (car L1) (car L2)).
6. Use the function SPLIT-LIST from Lisp Assignment 4 and MERGE-LISTS to define a recursive
 Lisp function MSORT such that if L is a list of real numbers then (MSORT L) is a list consisting
 of the elements of L in ascending order. In the definition of MSORT you may call SPLIT-LIST,
 MSORT itself, MERGE-LISTS, CAR, CDR, CADR and ENDP, but you should not call any
 other function. Be sure to use LET or LET*, so that MSORT only calls SPLIT-LIST once.
 Hint: Does a list of length 1 satisfy condition BC-3 (see page 2) for one of your function's recursive calls?
In problems 7 – 10, assume the argument is a list.
7. Do exercise 10.4(a) on page 418 of Sethi, but use Common Lisp instead of Scheme. Name your
 function REMOVE-ADJ-DUPL.
Hint for problems 8 and 9 One way to do these two problems is to give definitions of the following form:
 (defun function-name (L)
 (cond ((endp L) ... ) ; L is ( )
 ((or (endp (cdr L)) (not (equal (car L) (cadr L)))) ... ) ; L is (x) or (x y ... )
 ((or (endp (cddr L)) (not (equal (car L) (caddr L)))) ... ) ; L is (x x) or (x x y ... )
 (t ... ))) ; L is (x x x ... )
8. Do exercise 10.4(b) on the same page in Common Lisp. Name your function
 UNREPEATED-ELTS.
 9. Do exercise 10.4(c) on the same page in Common Lisp. Name your function REPEATED-ELTS.
10. Do exercise 10.4(d) on the same page in Common Lisp. Name your function
 COUNT-REPETITIONS.
PART II [Solutions to problems 11 – 16 will count 1% towards your grade if the grade is
 computed using rule A.]
11.[Exercise 8 on p. 141 of Wilensky] Write a recursive function SUBSET that takes two arguments:
 a function and a list. SUBSET should apply the function to each element of the list, and return a
 list of all the elements of the argument list for which the function application returns a true (i.e.,
 non-NIL) value. Example: (subset #'numberp '(a b 2 c d 3 e f)) = (2 3) 
Lisp Assignment 5: Page 4 of 5
12.[Exercise 7 on p. 141 of Wilensky] Write (i) a recursive function OUR-SOME and (ii) a recursive
 function OUR-EVERY each of which takes two arguments: a function and a list. OUR-SOME
 should apply the function to successive elements of the list until the function returns a true (i.e.,
 non-NIL) value, at which point it should return the rest of the list (including the element for which
 the function just returned a true value).* If the function returns NIL when applied to each of the
 arguments, OUR-SOME should return NIL. OUR-EVERY should apply the function to successive
 elements of the list until the function returns NIL, at which point it should return NIL. If the
 function returns a true value when applied to each of the arguments, then OUR-EVERY should
 return T. Examples:
 (our-some #'numberp '(A B 2 C D)) = (2 C D) (our-some #'numberp '(A B C D)) = NIL
 (our-every #'symbolp '(A B 2 C D)) = NIL (our-every #'symbolp '(A B C D)) = T *Note that the built-in Lisp function SOME behaves differently from OUR-SOME in this case: SOME returns the
 non-NIL value that was just returned by the function.
13. Modify your solutions to problem 7 of Lisp Assignment 4 and problem 4 above to produce a
 solution to exercise 10.6b on p. 419 of Sethi. Your sorting function should take as arguments a
 2-argument predicate p and a list L, and may assume the predicate p and list L are such that:
 1. For all elements x and y of L, (p x y) is either true or false.
 2. Whenever x and y are unequal elements of L, if (p x y) is true then (p y x) is false.
 3. For all elements x, y, and z of L, if (p x y) is true and (p y z) is true then (p x z) is also true.
 The sorted list returned by your function should contain exactly the same elements as L, and must
 satisfy the following condition: Writing si to denote the i
th element of the sorted list, whenever the
 elements sj and sk are unequal and j < k (i.e., sk appears after sj in the sorted list) we have that
 (p sk sj
) is false. (You can easily verify that the sorted lists in the three examples below satisfy this
 condition.) Your sorting function and the partition function it uses should be named QSORT1
 and PARTITION1. Examples:
 (QSORT1 #' '(2 8 5 1 5 7 3)) = (8 7 5 5 3 2 1)
 (QSORT1 #'< '(2 8 5 1 5 7 3)) = (1 2 3 5 5 7 8)
 (QSORT1 (LAMBDA (L1 L2) (< (LENGTH L1) (LENGTH L2)))
 '((X) (A D X E G) (1 2 Q R) NIL (S D F))) = (NIL (X) (S D F) (1 2 Q R) (A D X E G))
14. Do exercise 10.12a on p. 420 of Sethi in Common Lisp. Example:
 (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))
15. First, re-read section 8.16 (Advantages of Tail Recursion) on pp. 279 – 81 of Touretzky. Then solve
 the following problems.
 (a) Do exercise 10.5 on p. 419 of Sethi in Common Lisp. Name your functions TR-ADD,
 TR-MUL and TR-FAC. Examples:
 (TR-ADD '(1 2 3 4 5) 0) = 15 (TR-MUL '(1 2 3 4 5) 1) = 120 (TR-FAC 5 1) = 120
 Note that your definitions of TR-ADD, TR-MUL, and TR-FAC must be tail recursive.
 (b) Wilson's Theorem in number theory states that an integer n 1 is prime if and only if
 (n – 1)! mod n = n – 1. Use this theorem and TR-FAC to write a predicate SLOW-PRIMEP
 such that if n 1 is an integer then (SLOW-PRIMEP n) returns T or NIL according to
 whether or not n is prime. (You may of course use the built-in function MOD.) Test your
 predicate SLOW-PRIMEP by using it to find the least prime number greater than 20,000.
 Important: To prevent stack overflow, enter (compile 'tr-fac) at clisp's prompt
 before calling SLOW-PRIMEP with large arguments. 
Lisp Assignment 5: Page 5 of 5
16.[Based on exercise 8 on p. 111 of Wilensky] A matrix can be represented as a nonempty list of
 nonempty lists in which the ith element of the jth list is the element in the ith column and jth row of
 the matrix. For example, ((A B) (C D)) would represent a 2 × 2 matrix whose first row contains
 the elements A and B, and whose second row contains the elements C and D. Write three functions
 TRANSPOSE1, TRANSPOSE2, and TRANSPOSE3, each of which takes a single argument M; if
 M is a nonempty list of nonempty lists which represents a matrix in the above-mentioned way,
 then each of the three functions should return the list of lists which represents the transpose of that
 matrix. The three functions should work as follows:
 (a) (transpose1 M) computes its result from (transpose1 (cdr M)) and (car M) using
 mapcar #'cons. [Hint: Take the set of valid values of M to be the set of lists of one or more
 nonempty lists that all have the same length. Then the BC-2 base case is (endp (cdr M)),
 when the result to be returned is (mapcar #'list (car M)).]
 (b) (transpose2 M) computes its result from (transpose2 (mapcar #'cdr M)) and (mapcar #'car M).
 [Hint: Again, take the set of valid values of M to be the set of lists of one or more
 nonempty lists that all have the same length. This time, the BC-2 base case is
 (endp (cdar M)), when the result is (list (mapcar #'car M)).]
 (c) (transpose3 M) obtains its result as follows: (apply #'mapcar (cons #'??? M)) or, equivalently,
 (apply #'mapcar #'??? M), where ??? is the name of an appropriate function.
 Example (here n may be 1, 2 or 3):
 (TRANSPOSEn '((1 2 3 4) (5 6 7 8) (9 10 11 12))) = ((1 5 9) (2 6 10) (3 7 11)(4 8 12))
How to Submit Your Solutions for Grading
You may work with up to two other students on these problems, but each student must write up
his or her solutions individually––no two students should submit identical files. Put the function
definitions you wrote for problems 1–16 in a single file named your last name in lowercase-5.lsp
in your home directory on euclid. This file must include definitions of any helping functions that
are used. At the beginning of the file there must be a comment that shows your name and, if you
are working with one or two partners, the name(s) of your partner(s).
Functions that are incorrectly named may receive no credit (e.g., if your solution to
problem 5 is named MERGE-LIST instead of MERGE-LISTS, you may get no credit for that
problem). Test your functions on euclid! If euclid’s clisp cannot even LOAD your file without
error (i.e., if LOAD gives a Break ... prompt) then you can expect to receive
no credit at all for your submission, even if the only error is one missing or extra parenthesis!
Within the file, your solution to each problem should be preceded by a comment of the
following form: ;;; Solution to Problem N Your solutions should appear in the
same order as the problems. If you cannot solve a problem, put a comment of the form
;;; No Solution to Problem N Submitted where a solution would have appeared.
Leave your final version of your last name in lowercase-5.lsp in your home
directory on euclid no later than midnight on the due date.
Do NOT open the submitted file in an editor on euclid after the due date, unless you are
resubmitting a corrected version of your solutions as a late submission. Also do not execute mv,
chmod, or touch with your submitted file as an argument after the due date. However, it is OK to
view a submitted file using the less file viewer after the due date. [Note: If euclid unexpectedly goes down
after 6 p.m. on the due date, there will be no extension. Try to submit by noon that day, and much sooner if possible.]
As mentioned on p. 3 of the first-day announcements document, you are required to keep a backup
copy of your submitted file on venus, and another copy elsewhere. You can enter the following two commands
on euclid to email a copy of your submitted file to yourself and to put a copy of the file on venus:
 echo . | mailx -s "copy of submission" -a your last name in lowercase-5.lsp $USER
 scp your last name in lowercase-5.lsp your venus username@149.4.211.180: