Related
I'm trying to make my own union function and realizing how much I dislike LISP. The goal is to give the function two lists and it will return a set theoretic union of the two. My attempted solution has grown increasingly complex with the same result: NIL. I can't change that from being the result no matter what I do.
I was thinking of building a separate list in my "removeDuplicates" function below, but then idk how I'd return that with recursion. I think what's happening is my "removeDuplicates" function eventually returns an empty list (as intended) but then an empty list is return at every level of the stack when the recursion unfurls (starts returning values up the stack) but I could be wrong. I've always had trouble understanding recursion in detail. The code is below.
(defun rember (A LAT)
(cond
((null LAT) ())
((EQ (car LAT) A) (cdr LAT))
(T (cons (car LAT)(rember A (cdr LAT))))
)
)
(defun my_member (A LAT)
(cond
((null LAT) nil)
((EQ (car LAT) A) T)
(T (my_member A (cdr LAT)))
)
)
(defun removeDuplicates (L)
(cond
((null L) '())
((my_member (car L) (cdr L)) (rember (car L) L) (removeDuplicates (cdr L)))
(T (removeDuplicates (cdr L)))
)
)
(defun my_union (A B)
(setq together(append A B))
(removeDuplicates together)
)
I'm aware most people are not a fan of this format of LISP code, but I prefer it. It allows me to see how parentheses line up better than if you just put all the closing parentheses together at the end of functions and condition blocks.
If I run (my_union '(a b) '(b c)) for example, the result is NIL.
When you call removeDuplicates recursively in the last condition, you're not combining the result with the car of the list, so you're discarding that element.
You're also not using the result of rember.
(defun removeDuplicates (L)
(cond
((null L) '())
((my_member (car L) (cdr L))
(cons (car L)
(removeDuplicates
(rember (car L) (cdr L))
))
)
(T (cons (car L) (removeDuplicates (cdr L))))
)
)
Here's a simple, obvious, union function:
(defun union/tfb (&rest lists)
;; compute the union of any number of lists, implicitly using EQL
(labels ((union/spread (l1 ls)
;; UNION/SPREAD just exists to avoid the impedance
;; mismatch in argument convention
(if (null ls)
l1
(let ((result l1))
(destructuring-bind (l2 . more) ls
(dolist (e l2 (union/spread result more))
(pushnew e result)))))))
(union/spread (first lists) (rest lists))))
I think this is reasonably natural CL, although of course the whole point of using a language like CL is avoiding endless wheel-reinvention like this.
So the rules of the game perhaps say you're not allowed to use PUSHNEW: well, you can easily can replace it with a conditional involving MEMBER:
(defun union/tfb (&rest lists)
;; compute the union of any number of lists, implicitly using EQL
(labels ((union/spread (l1 ls)
;; UNION/SPREAD just exists to avoid the impedance
;; mismatch in argument convention
(if (null ls)
l1
(let ((result l1))
(destructuring-bind (l2 . more) ls
(dolist (e l2 (union/spread result more))
;; Really use PUSHNEW for this
(unless (member e result)
(setf result (cons e result)))))))))
(union/spread (first lists) (rest lists))))
And perhaps you are also not allowed to use MEMBER: well you can easily write a predicate which does what you need:
(defun union/tfb (&rest lists)
;; compute the union of any number of lists, implicitly using EQL
(labels ((union/spread (l1 ls)
;; UNION/SPREAD just exists to avoid the impedance
;; mismatch in argument convention
(if (null ls)
l1
(let ((result l1))
(destructuring-bind (l2 . more) ls
(dolist (e l2 (union/spread result more))
;; Really use MEMBER for this, and in fact
;; PUSHNEW
(unless (found-in-p e result)
(setf result (cons e result))))))))
(found-in-p (e list)
;; is e found in LIST? This exists only because we're not
;; meant to use MEMBER
(cond ((null list) nil)
((eql e (first list)) t)
(t (found-in-p e (rest list))))))
(union/spread (first lists) (rest lists))))
If you want the result to be a set with unique elements even if the first list is not you can trivially do that (note CL's UNION does not promise this, and you can get the same result with the earlier version of UNION/TFB by (union/tfb '() ...)):
(defun union/tfb (&rest lists)
;; compute the union of any number of lists, implicitly using EQL
(labels ((union/spread (l1 ls)
;; UNION/SPREAD just exists to avoid the impedance
;; mismatch in argument convention
(if (null ls)
l1
(let ((result l1))
(destructuring-bind (l2 . more) ls
(dolist (e l2 (union/spread result more))
;; Really use MEMBER for this, and in fact
;; PUSHNEW
(unless (found-in-p e result)
(setf result (cons e result))))))))
(found-in-p (e list)
;; is e found in LIST? This exists only because we're not
;; meant to use MEMBER
(cond ((null list) nil)
((eql e (first list)) t)
(t (found-in-p e (rest list))))))
(union/spread '() lists)))
Finally if the rules prevent you using iterative constructs and assignment you can do that too:
(defun union/tfb (&rest lists)
;; compute the union of any number of lists, implicitly using EQL
(labels ((union/spread (l1 ls)
;; UNION/SPREAD just exists to avoid the impedance
;; mismatch in argument convention
(if (null ls)
l1
(union/loop l1 (first ls) (rest ls))))
(union/loop (result l more)
;; UNION/LOOP is just an iteration
(if (null l)
(union/spread result more)
(destructuring-bind (e . remainder) l
(union/loop (if (found-in-p e result)
result
(cons e result))
remainder more))))
(found-in-p (e list)
;; is e found in LIST? This exists only because we're not
;; meant to use MEMBER
(cond ((null list) nil)
((eql e (first list)) t)
(t (found-in-p e (rest list))))))
(union/spread '() lists)))
The final result of all these changes is something which is, perhaps, very pure, but is not natural CL at all: something like it might be more natural in Scheme (albeit not gratuitously replacing MEMBER with a home-grown predicate like this).
One way to test your Common Lisp code is to ask your interpreter to TRACE functions:
(trace removeDuplicates my_member rember)
To avoid having too many traces, use small examples.
First, let's try with an empty list; this is an example from the REPL ("read eval print loop"), tested with SBCL, while in the "SO" package (StackOverflow); the trace is printed a bit indented, a is numbered according to the depth of the recursion. Here the call is not recursive and terminates right away:
SO> (removeduplicates nil)
0: (SO::REMOVEDUPLICATES NIL)
0: REMOVEDUPLICATES returned NIL
NIL
This works, let's try an example with a singleton list, where there is obviously no duplicate:
SO> (removeduplicates '(1))
0: (SO::REMOVEDUPLICATES (1))
1: (SO::MY_MEMBER 1 NIL)
1: MY_MEMBER returned NIL
1: (SO::REMOVEDUPLICATES NIL)
1: REMOVEDUPLICATES returned NIL
0: REMOVEDUPLICATES returned NIL
NIL
removeDuplicate calls my_member, which correctly returns nil, followed by a recursive call to removeDuplicates with nil, which correctly returns nil. There is however a problem because then, the outermost call returns nil too, which is incorrect.
Looking at the trace, we have to look back at the code to find a place where my_member is called, followed by a recursive call to removeDuplicates. There is only one place wher my_member is called, as a test to the second clause in the cond;
Since the result is nil for that test, the next clause is tried, in that case the default case:
(cond
...
;; this is the call to my_member (= nil)
((my_member (car L) (cdr L)) ...)
;; this is the recursive call
(t (removeDuplicates (cdr L))))
The value of the cond is the one given by the last (removeDuplicates (cdr L)), which just does not retain the existing elements in front of L. If you were mutating a sequence, you could just recurse down the subsequence and ignore the previous elements: in that case the caller would still hold a reference to the original sequence, which would get its element removed by a side-effect of your functions. But here you are following a strictly immutable approach, and you have to recontruct a list as a return value.
In other words, removeDuplicates is expressed as: return a new list which contains the same elements as the original list, but without duplicates.
So you have to add (car L) in front of (removeDuplicates (cdr L)).
(defun removeDuplicates (L)
(cond
((null L) '())
((my_member (car L) (cdr L)) (rember (car L) L) (removeDuplicates (cdr L)))
(T (cons (car L)
(removeDuplicates (rest L))))))
Let's test:
SO> (removeduplicates '())
0: (SO::REMOVEDUPLICATES NIL)
0: REMOVEDUPLICATES returned NIL
NIL
SO> (removeduplicates '(1))
0: (SO::REMOVEDUPLICATES (1))
1: (SO::MY_MEMBER 1 NIL)
1: MY_MEMBER returned NIL
1: (SO::REMOVEDUPLICATES NIL)
1: REMOVEDUPLICATES returned NIL
0: REMOVEDUPLICATES returned (1)
(1)
You can test with a longer list (without duplicates), the result is correct, but the trace is longer.
Now, let's add duplicates:
SO> (removeduplicates '(1 2 2 1))
0: (SO::REMOVEDUPLICATES (1 2 2 1))
1: (SO::MY_MEMBER 1 (2 2 1))
2: (SO::MY_MEMBER 1 (2 1))
3: (SO::MY_MEMBER 1 (1))
3: MY_MEMBER returned T
2: MY_MEMBER returned T
1: MY_MEMBER returned T
1: (SO::REMBER 1 (1 2 2 1))
1: REMBER returned (2 2 1)
1: (SO::REMOVEDUPLICATES (2 2 1))
2: (SO::MY_MEMBER 2 (2 1))
2: MY_MEMBER returned T
2: (SO::REMBER 2 (2 2 1))
2: REMBER returned (2 1)
2: (SO::REMOVEDUPLICATES (2 1))
3: (SO::MY_MEMBER 2 (1))
4: (SO::MY_MEMBER 2 NIL)
4: MY_MEMBER returned NIL
3: MY_MEMBER returned NIL
3: (SO::REMOVEDUPLICATES (1))
4: (SO::MY_MEMBER 1 NIL)
4: MY_MEMBER returned NIL
4: (SO::REMOVEDUPLICATES NIL)
4: REMOVEDUPLICATES returned NIL
3: REMOVEDUPLICATES returned (1)
2: REMOVEDUPLICATES returned (2 1)
1: REMOVEDUPLICATES returned (2 1)
0: REMOVEDUPLICATES returned (2 1)
(2 1)
The result is correct (order does not matter).
So far, our tests are good.
You might not have identified the other problem in that function, namely that all calls to rember are useless, and frankly this is not necessarily easy to spot with the trace. But looking at the code, it should be clear if you write code to have little side-effects that the following clause calls (rember ...) for nothing:
((my_member (car L) (cdr L)) (rember (car L) L) (removeDuplicates (cdr L)))
A cond clause has for syntax (TEST . BODY), where BODY is a sequence of expressions that evaluates like a PROGN: the value of a PROGN is the value of its last clause, all intermediate clauses are only used for their side-effects. For example:
(progn
(print "I am here")
(* 10 3))
Here above, the call to PRINT returns a value, but it is discarded: the value of the enclosing PROGN is 30.
In your code, rember does no side-effect, and its return value is discarded. Just remove it:
(defun removeDuplicates (L)
(cond
((null L) '())
((my_member (car L) (cdr L))
(removeDuplicates (cdr L)))
(T (cons (first L)
(removeDuplicates (rest L))))))
I would write the same code as follows, personally:
(defun remove-duplicate-elements (list)
(when list
(let ((head (first list))
(tail (remove-duplicate-elements (rest list))))
(if (member head tail) tail (cons head tail)))))
Here is a remove-dupes that removes duplicates from a list in O(n) time using a hash table. It supports a custom equality function (which must be eq, eql, equal or `equalp) and a custom test function, so that any aspect of an item can be treated as the key.
(defun remove-dupes (list &key (test #'eql) (key #'identity))
(let ((hash (make-hash-table :test test)))
(loop for item in list
for item-key = (funcall key item)
for seen = (gethash item-key hash)
unless seen collect item and
do (setf (gethash item-key hash) t))))
For instance, suppose we have the assoc list ((a . 1) (a . 2) (b . 3) (c . 4) (b . 4)). We'd like to remove duplicates by car:
[1]> (remove-dupes '((a . 1) (a . 2) (b . 3) (c . 4) (b . 4)) :key #'car)
((A . 1) (B . 3) (C . 4))
Only the leftmost A, B and C entries are reported; the duplicates are suppressed. Now let's do it by cdr:
[2]> (remove-dupes '((a . 1) (a . 2) (b . 3) (c . 4) (b . 4)) :key #'cdr)
((A . 1) (A . 2) (B . 3) (C . 4))
The (b . 4) got culled due to the duplicated 4 value.
But, why do all this, when Common Lisp provides a remove-duplicates function (not to mention union).
remove-duplicates is more general than what I have here: it handles sequences, rather than just lists, so it works on vectors and strings. It has more keyword parameters.
I have the following Racket code:
#lang racket
(define (for-each proc items)
(cond ((not (null? items))
(proc (car items))
(for-each proc (cdr items)))))
(for-each (lambda (x) (newline) (display x))
(list 57 321 88))
Because the condition isn't satisfied after 88 is printed, and the Racket's documentation says:
If none of the question-expressions evaluates to true, cond’s value is the answer-expression of the else clause. If there is no else, cond reports an error.
So the expected result is an error, however I get the correct output:
57
321
88
Can somebody tell me why? (ps: I'm using DrRacket 6.2)
Changing the indentation a little:
(define (for-each proc items)
(cond
[(not (null? items)) (proc (car items))
(for-each proc (cdr items))]))
A call (for-each f (list 1 2 3))
will call the function and bind proc to f and items to '(1 2 3).
The test (not (null? items)) is true, since '(1 2 3) is non-empty.
The right hand side is therefore evaluated. The right hand side does two things:
1.) (proc (car items)) calls f with the first element in the list (here 1) and then 2). calls (for-each proc (cdr items)). Here (cdr items) becomes '(2 3).
Eventually for-each will be called with (for-each f '()).
At that point (not (null? items)) will evaluate to #f and the cond will try the next clause. Since there are no more clauses cond will return void.
The documentation says:
If no cond-clauses are present, the result is #<void>.
Note: The cond in the teaching languages will return an error. The constructs in the teaching language generally reports errors, where there is a chance of shooting ones own foot.
I don't know where you got that information. To quote the spec:
The last test-expr in a cond can be replaced by else. In terms of evaluation, else serves as a synonym for #t, but it clarifies that the last clause is meant to catch all remaining cases. If else is not used, then it is possible that no test-exprs produce a true value; in that case, the result of the cond expression is #void.
I want to ask why this function doesn't work...
(defun nenum(ls)
(cond
((null ls) nil)
((listp car(ls)) (nenum (rest ls)))
((numberp car(ls)) (nenum (rest ls)))
(t (cons (car ls) (nenum (rest ls))))))
Example: (nenum '(l 1 i (b) (5) s -2 p)) --> (l i s p)
Thank you!
Looking at the predicate you have in one of your cond terms:
(listp car (ls))
Thus apply the function listp with the two arguments car and the result of calling the function ls with no arguments. car and ls both need to be free variables and listp needs to be a different function than the one defined in CLHS since it only takes one argument.
Perhaps you have though you were writing Algol? An Algol function call look like operator(operand) but not CL. CL is a LISP dialect and we have this form on our function calls:
(operand operator)
If we nest we do the same:
(operand (operand operator))
You got it right in the alternative (cons (car ls) (nenum (rest ls)))
Replace car(ls) with (car ls).
Here's a much easier way to write that function:
(defun nenum (list)
(remove-if (lambda (item)
(or (listp item)
(numberp item)))
list))
Note that NIL doesn't need its own test because listp covers it.
There's no need to write a function like this from scratch. Common Lisp already provides remove-if, and you can give it a predicate that matches numbers and non-atoms:
CL-USER> (remove-if #'(lambda (x)
(or (numberp x)
(not (atom x))))
'(l 1 i (b) (5) s -2 p))
;=> (L I S P)
Or, to make it even clearer that you're keeping non-numeric atoms, you can use remove-if-not with a predicate that checks for numeric atoms:
CL-USER> (remove-if-not #'(lambda (x)
(and (atom x)
(not (numberp x))))
'(l 1 i (b) (5) s -2 p))
;=> (L I S P)
Note that the empty list, which is often written as (), is just the symbol nil. As such, it too is a non-numeric atom. If you'd want to keep other symbols, e.g.,
CL-USER> (remove-if-not #'(lambda (x)
(and (atom x)
(not (numberp x))))
'(li (b) -1 (5) sp))
;=> (LI SP)
then you'll probably want to keep nil as well:
CL-USER> (remove-if-not #'(lambda (x)
(and (atom x)
(not (numberp x))))
'(van (b) () (5) a))
;=> (VAN NIL A)
I have a lisp homework I am having a hard time with it.
I have to write a function that perform a union operation. The function takes 2 inputs, either in the form of either atom or list and unions every element, preserving the order and stripping off all levels of parenthesis.
The output for the function:
(my-union 'a 'b) ;; (a b)
(my-union 'a '(b)) ;; (a b)
(my-union '(a b) '(b c)) ;; (a b c)
(my-union '(((a))) '(b(c((d e))a))) ;; (a b c d e)
I am fairly new to lisp.
Here is what I have written so far and it works only for the third example:
(defun new-union (a b)
(if (not b)
a
(if (member (car b) a)
(new-union a (cdr b))
(new-union (append a (list (car b))) (cdr b)))))
Any help would be appreciated!
Since this is your first homework, and you are new to Lisp, here is a very simple top-down approach, not worrying about performance, and making good use of the tools CL offers:
In Common Lisp, there is already a function which removes duplicates: remove-duplicates. Using it with the :from-end keyword-argument will "preserve order". Now, imagine you had a function flatten, which flattens arbitrarily nested lists. Then the solution to your question would be:
(defun new-union (list1 list2)
(remove-duplicates (flatten (list list1 list2)) :from-end t))
This is how I would approach the problem when no further restrictions are given, and there is no real reason to worry much about performance. Use as much of the present toolbox as possible and do not reinvent wheels unless necessary.
If you approach the problem like this, it boils down to writing the flatten function, which I will leave as an exercise for you. It is not too hard, one easy option is to write a recursive function, approaching the problem like this:
If the first element of the list to be flattened is itself a list, append the flattened first element to the flattened rest. If the first element is not a list, just prepend it to the flattened rest of the list. If the input is not a list at all, just return it.
That should be a nice exercise for you, and can be done in just a few lines of code.
(If you want to be very correct, use a helper function to do the work and check in the wrapping function whether the argument really is a list. Otherwise, flatten will work on atoms, too, which may or may not be a problem for you.)
Now, assuming you have written flatten:
> (defun new-union (list1 list2)
(remove-duplicates (flatten (list list1 list2)) :from-end t))
NEW-UNION
> (new-union 'a 'b)
(A B)
> (new-union 'a '(b))
(A B)
> (new-union '(a b) '(b c))
(A B C)
> (new-union '(((a))) '(b (c ((d e)) a)))
(A B C D E)
One way to approach this is to separate your concerns. One is flattening; another is duplicates-removing; yet another is result-building.
Starting with empty list as your result, proceed to add into it the elements of the first list, skipping such elements that are already in the result.
Then do the same with the second list's elements, adding them to the same result list.
(defun my-union (a b &aux (res (list 1)) (p res))
(nadd-elts p a)
(nadd-elts p b)
(cdr res))
nadd-elts would add to the end of list, destructively updating its last cell (pointed to by p) using e.g. rplacd. An example is here.
To add elements, nadd-elts would emulate the flattening procedure, and add each leaf element into p after checking res for duplicates.
Working in functional style, without destructive update, the general approach stays the same: start with empty result list, add first list into it - without duplicates - then second.
(defun my-union (a b &aux res)
(setq res (add-into res a))
(setq res (add-into res b))
res)
Now we're left with implementing the add-into function.
(defun add-into (res a &aux r1 r2)
(cond
((atom a) .... )
(T (setq r1 (add-into res (car a)))
(setq r2 (............ (cdr a)))
r2)))
The above can be re-written without the auxiliary variables and without set primitives. Try to find out how... OK here's what I meant by that:
(defun my-union (a b) (add-into NIL (cons a b)))
(defun add-into (res a)
(cond
((atom a) .... )
(T (add-into (add-into res (car a))
(cdr a)))))
Unless you are not allowed to use hash table (for some reason I've encountered this as a requirement before), you could come up with an ordering function that will help you build the resulting set in the way, that you don't have to repeat the search over and over again.
Also, since nested lists are allowed your problem scales down to only removing duplicates in a tree (as you can simply append as many lists as you want before you start processing them.
Now, I'll try to show few examples of how you could do it:
;; Large difference between best and worst case.
;; Lists containing all the same items will be processed
;; in square time
(defun union-naive (list &rest lists)
(when lists (setf list (append list lists)))
(let (result)
(labels ((%union-naive (tree)
(if (consp tree)
(progn
(%union-naive (car tree))
(when (cdr tree) (%union-naive (cdr tree))))
(unless (member tree result)
(setq result (cons tree result))))))
(%union-naive list) result)))
;; Perhaps the best solution, it is practically linear time
(defun union-hash (list &rest lists)
(when lists (setf list (append list lists)))
(let ((hash (make-hash-table)) result)
(labels ((%union-hash (tree)
(if (consp tree)
(progn
(%union-hash (car tree))
(when (cdr tree) (%union-hash (cdr tree))))
(setf (gethash tree hash) t))))
(%union-hash list))
(maphash
#'(lambda (a b)
(declare (ignore b))
(push a result)) hash)
result))
;; This will do the job in more time, then the
;; solution with the hash-map, but it requires
;; significantly less memory. Memory is, in fact
;; a more precious resource some times, but you
;; have to decide what algo to use based on the
;; data size
(defun union-flatten (list &rest lists)
(when lists (setf list (append list lists)))
(labels ((%flatten (tree)
(if (consp tree)
(if (cdr tree)
(nconc (%flatten (car tree))
(%flatten (cdr tree)))
(%flatten (car tree)))
(list tree))))
;; the code below is trying to do something
;; that you could've done using
;; (remove-duplicates (%flatten list))
;; however sorting and then removing duplicates
;; may prove to be more efficient
(reduce
#'(lambda (a b)
(cond
((atom a) (list a))
((eql (car a) b) b)
(t (cons b a))))
(sort (%flatten list)
#'(lambda (a b)
(string< (symbol-name a)
(symbol-name b)))))))
(union-naive '(((a))) '(b(c((d e))a)))
(union-hash '(((a))) '(b(c((d e))a)))
(union-flatten '(((a))) '(b(c((d e))a)))
Notice that the function I've used to order elements is not universal, but you would probably be able to come up with an alternative function for any sort of data. Any fast hashing function in general would do, I've used this one for simplicity.
How can I remove nested parentheses recursively in Common LISP Such as
(unnest '(a b c (d e) ((f) g))) => (a b c d e f g)
(unnest '(a b)) => (a b)
(unnest '(() ((((a)))) ())) => (a)
Thanks
Here's what I'd do:
(ql:quickload "alexandria")
(alexandria:flatten list)
That works mainly because I have Quicklisp installed already.
(defun flatten (l)
(cond ((null l) nil)
((atom l) (list l))
(t (loop for a in l appending (flatten a)))))
I realize this is an old thread, but it is one of the first that comes up when I google lisp flatten. The solution I discovered is similar to those discussed above, but the formatting is slightly different. I will explain it as if you are new to lisp, as I was when I first googled this question, so it's likely that others will be too.
(defun flatten (L)
"Converts a list to single level."
(if (null L)
nil
(if (atom (first L))
(cons (first L) (flatten (rest L)))
(append (flatten (first L)) (flatten (rest L))))))
For those new to lisp, this is a brief summary.
The following line declares a function called flatten with argument L.
(defun flatten (L)
The line below checks for an empty list.
(if (null L)
The next line returns nil because cons ATOM nil declares a list with one entry (ATOM). This is the base case of the recursion and lets the function know when to stop. The line after this checks to see if the first item in the list is an atom instead of another list.
(if (atom (first L))
Then, if it is, it uses recursion to create a flattened list of this atom combined with the rest of the flattened list that the function will generate. cons combines an atom with another list.
(cons (first L) (flatten (rest L)))
If it's not an atom, then we have to flatten on it, because it is another list that may have further lists inside of it.
(append (flatten (first L)) (flatten (rest L))))))
The append function will append the first list to the start of the second list.
Also note that every time you use a function in lisp, you have to surround it with parenthesis. This confused me at first.
You could define it like this for example:
(defun unnest (x)
(labels ((rec (x acc)
(cond ((null x) acc)
((atom x) (cons x acc))
(t (rec (car x) (rec (cdr x) acc))))))
(rec x nil)))
(defun flatten (l)
(cond ((null l) nil)
((atom (car l)) (cons (car l) (flatten (cdr l))))
(t (append (flatten (car l)) (flatten (cdr l))))))
Lisp has the function remove to remove things. Here I use a version REMOVE-IF that removes every item for which a predicate is true. I test if the thing is a parenthesis and remove it if true.
If you want to remove parentheses, see this function:
(defun unnest (thing)
(read-from-string
(concatenate
'string
"("
(remove-if (lambda (c)
(member c '(#\( #\))))
(princ-to-string thing))
")")))
Note, though, as Svante mentions, one does not usually 'remove' parentheses.
Most of the answers have already mentioned a recursive solution to the Flatten problem. Using Common Lisp Object System's multiple dispatching you could solve the problem recursively by defining 3 methods for 3 possible scenarios:
(defmethod flatten ((tree null))
"Tree is empty list."
())
(defmethod flatten ((tree list))
"Tree is a list."
(append (flatten (car tree))
(flatten (cdr tree))))
(defmethod flatten (tree)
"Tree is something else (atom?)."
(list tree))
(flatten '(2 ((8) 2 (9 (d (s (((((a))))))))))) ; => (2 8 2 9 D S A)
Just leaving this here as I visited this question with the need of only flattening one level and later figure out for myself that (apply 'concatenate 'list ((1 2) (3 4) (5 6 7))) is a cleaner solution in that case.
This is a accumulator based approach. The local function %flatten keeps an accumulator of the tail (the right part of the list that's already been flattened). When the part remaining to be flattened (the left part of the list) is empty, it returns the tail. When the part to be flattened is a non-list, it returns that part prefixed onto the tail. When the part to be flattened is a list, it flattens the rest of the list (with the current tail), then uses that result as the tail for flattening the first part of the list.
(defun flatten (list)
(labels ((%flatten (list tail)
(cond
((null list) tail)
((atom list) (list* list tail))
(t (%flatten (first list)
(%flatten (rest list)
tail))))))
(%flatten list '())))
CL-USER> (flatten '((1 2) (3 4) ((5) 6) 7))
(1 2 3 4 5 6 7)
I know this question is really old but I noticed that nobody used the push/nreverse idiom, so I am uploading that here.
the function reverse-atomize takes out each "atom" and puts it into the output of the next call. At the end it produces a flattened list that is backwards, which is resolved with the nreverse function in the atomize function.
(defun reverse-atomize (tree output)
"Auxillary function for atomize"
(if (null tree)
output
(if (atom (car tree))
(reverse-atomize (cdr tree) (push (car tree) output))
(reverse-atomize (cdr tree) (nconc (reverse-atomize (car tree)
nil)
output)))))
(defun atomize (tree)
"Flattens a list into only the atoms in it"
(nreverse (reverse-atomize tree nil)))
So calling atomize '((a b) (c) d) looks like this:
(A B C D)
And if you were to call reverse-atomize with reverse-atomize '((a b) (c) d) this would occur:
(D C B A)
People like using functions like push, nreverse, and nconc because they use less RAM than their respective cons, reverse, and append functions. That being said the double recursive nature of reverse-atomize does come with it's own RAMifications.
This popular question only has recursive solutions (not counting Rainer's answer).
Let's have a loop version:
(defun flatten (tree &aux todo flat)
(check-type tree list)
(loop
(shiftf todo tree nil)
(unless todo (return flat))
(dolist (elt todo)
(if (listp elt)
(dolist (e elt)
(push e tree))
(push elt flat))))))
(defun unnest (somewhat)
(cond
((null somewhat) nil)
((atom somewhat) (list somewhat))
(t
(append (unnest (car somewhat)) (unnest (cdr somewhat))))))
I couldn't resist adding my two cents. While the CL spec does not require tail call optimization (TCO), many (most?) implementations have that feature.
So here's a tail recursive version that collects the leaf nodes of a tree into a flat list (which is one version of "removing parentheses"):
(defun flatten (tree &key (include-nil t))
(check-type tree list)
(labels ((%flatten (lst accum)
(if (null lst)
(nreverse accum)
(let ((elem (first lst)))
(if (atom elem)
(%flatten (cdr lst) (if (or elem include-nil)
(cons elem accum)
accum))
(%flatten (append elem (cdr lst)) accum))))))
(%flatten tree nil)))
It preserves null leaf nodes by default, with the option to remove them. It also preserves the left-to-right order of the tree's leaf nodes.
Note from Google lisp style guide about TCO:
You should favor iteration over recursion.
...most serious implementations (including SBCL and CCL) do implement proper tail calls, but with restrictions:
The (DECLARE (OPTIMIZE ...)) settings must favor SPEED enough and not favor DEBUG too much, for some compiler-dependent meanings of "enough" and "too much".
And this from SBCL docs:
... disabling tail-recursion optimization ... happens when the debug optimization quality is greater than 2.