The function "greaterthan", (< NUM1 NUM2), allows only for returning t/nil for comparing 2 values.
I would like to test (var1 > var2 < var3 < var4), is there any way to do that using only one function in lisp? If not, what is the best procedure?
The best procedure is not to bother: (and (< var2 var1) (< var2 var3) (< var3 var4)) is not harder to read that your ..>..<..<.. chain.
It makes sense to test for the ascending order:
(require 'cl)
(defun cl-< (&rest args)
(every '< args (cdr args))
These days I don't hesitate to (require 'cl) anymore, but if you do,
here is another variant:
(defun cl-< (arg &rest more-args)
(or (null more-args)
(and (< arg (first more-args))
(apply #'cl-< more-args))))
The following is a macro implementation for variadic <
(defmacro << (x y &rest args)
(if args
(if (or (symbolp y)
(numberp y))
`(and (< ,x ,y) (<< ,y ,#args))
(let ((ys (make-symbol "y")))
`(let (,ys)
(and (< ,x (setq ,ys ,y))
(<< ,ys ,#args)))))
`(< ,x ,y)))
for simple cases just expands to (and ...) chains
(<< x y z) ==> (and (< x y) (< y z))
where the expression is not a number and not a symbol expands to a more complex form to avoid multiple evaluations in presence of side effects
(<< (f x) (g y) (h z)) ==> (let ((gy)) (and (< (f x) (setq gy (g y)))
(< gy (h z))))
for example
(setq foo (list))
nil
(defun call (x) (push x foo) x)
call
(<< (call 1) (call 2) (call 5) (call 4) (call 0))
nil
foo
(4 5 2 1)
every function has been called once, except for 0 that didn't need to be called because of short circuiting (I'm not 100% sure if short circuiting is a really good idea or not... #'< in Common Lisp is a regular function with all arguments all evaluated exactly once in left-to-right order without short circuiting).
(defun << (arg1 arg2 arg3 arg4)
(when (and (< arg1 arg2) (< arg2 arg3) (< arg3 arg4)))
)
(<< 1 2 3 4)
Probably possible to extend with any amount of arguments, but such a general form would seem useful.
(defmacro << (&rest args)
(let ((first (car args))
(min (gensym))
(max (gensym))
(forms '(t)) iterator)
(setq args (reverse (cdr args))
iterator args)
`(let ((,min ,first) ,max)
,(or
(while iterator
(push `(setq ,min ,max) forms)
(push `(< ,min ,max) forms)
(push `(setq ,max ,(car iterator)) forms)
(setq iterator (cdr iterator))) `(and ,#forms)))))
(macroexpand '(<< 10 20 30 (+ 30 3) (* 10 4)))
(let ((G99730 10) G99731)
(and (setq G99731 20)
(< G99730 G99731)
(setq G99730 G99731)
(setq G99731 30)
(< G99730 G99731)
(setq G99730 G99731)
(setq G99731 (+ 30 3))
(< G99730 G99731)
(setq G99730 G99731)
(setq G99731 (* 10 4))
(< G99730 G99731)
(setq G99730 G99731) t))
This is the idea similar to 6502's, but it may create less code, in a less trivial situation, but it will create more code in a trivial situation.
Related
I was thinking about a cond with a twist
(let ((a 0))
(let* ((result nil))
(tagbody
(let ((b1 (+ 0 a)))
(when (eq b1 1)
(print "1")
(setf result b1)
(go finish)))
(let ((b2 (+ 0 a)))
(when (eq b2 2)
(print "2")
(setf result b2)
(go finish)))
(when T
(print "else")
(setf result a))
(format t "=== ~A~%" a)
finish)
result))
where when test-form is wrapped in let. On one hand this seems to fit into a problem I am working on, but also seems overcomplicated. Can it be simplified with a macro? What would be the best way to simplify it if I had lots of test-forms?
Part of the problem in trying to do it that way is restricting the let blocks to only one test-form and its body.
But I wonder if I am going down the wrong path. Playing with an imaginary variant of when-let suggests there is no benefit of going down this path.
Trying cond
The version using cond appears to be more compact.
(let ((a 3))
(let* ((b1 (+ 0 a))
(b2 (+ 0 a)))
(cond
((eq b1 1)
(print "1")
b1)
((eq b2 2)
(print "2")
b2)
(T (print "else")
a))))
All boils down to the variables defined in the let* which in real life example would be used to avoid calculating the same value twice and improve readability. What should I do?
I'd prefer to think more in terms of blocks and returning values from them, instead working with goto and variables. If one really needs separate let-bound variables and their own scope:
(prog ((a 0))
(let ((b1 (+ 0 a)))
(when (eql b1 1)
(print "1")
(return b1)))
(let ((b2 (+ 0 a)))
(when (eql b2 2)
(print "2")
(return b2)))
(return
(progn
(print "else")
(return a))))
Somebody did now. I wanted it to be compatible with cond which raises a trouble: if you want the binding clauses to be like
(cond/binding
...
((var expr) <use var>)
...)
But you want to allow just general test clauses, then a function with one argument is ambiguous: should
(cond/binding
...
((car x) ...)
...)
call car or bind car? To make this work then you need to bind a useless variable in that case:
(cond/binding
...
((useless (car x)) <useless not used here>)
...)
And that means you either need to insert ignore or ignorable declarations all over the place, or live with compiler warnings.
So, well, I decided it would be better to go the other way: you have to say when you want to bind a variable. And you do that by a clause like:
(cond/binding
...
((bind var expr) <var is bound here>)
...)
And note that bind is magic in the syntax (so this means you can't call a function called bind, but that's OK as I already use bind as a keyword in other macros.
The macro also tries hard (well, hard given I basically just typed it in and it's had no testing) to actually behave like cond: returning multiple values, for instance.
So this:
(cond/binding
((f x y z) t)
((bind x 3) (print x) (values x t))
(t (values nil nil))
(1))
expands to
(block #:cond/binding
(when (f x y z)
(return-from #:cond/binding (progn t)))
(let ((x 3))
(when x
(return-from #:cond/binding
(progn (print x) (values x t)))))
(when t
(return-from #:cond/binding (progn (values nil nil))))
(let ((r 1))
(when r
(return-from #:cond/binding r))))
(where all the blocks are the same block).
So, here:
(defmacro cond/binding (&body clauses)
;; Like COND but it can bind variables. All clauses are (should be)
;; like COND, except that a clause of the form ((bind var <expr>)
;; ...) will bind a variable. Note that bind has to be literally
;; the symbol BIND: it's magic in the syntax.
(let ((bn (make-symbol "COND/BINDING")))
`(block ,bn
,#(mapcar
(lambda (clause)
(unless (consp clause)
(error "bad clause ~S" clause))
(case (length clause)
(1
`(let ((r ,(car clause)))
(when r (return-from ,bn r))))
(otherwise
(destructuring-bind (test/binding &body forms) clause
(typecase test/binding
(cons
(case (car test/binding)
((bind)
(unless (and (= (length test/binding) 3)
(symbolp (second test/binding)))
(error "bad binding clause ~S" test/binding))
(destructuring-bind (var expr) (rest test/binding)
`(let ((,var ,expr))
(when ,var
(return-from ,bn
(progn ,#forms))))))
(otherwise
`(when ,test/binding
(return-from ,bn
(progn ,#forms))))))
(t
`(when ,test/binding
(return-from ,bn
(progn ,#forms)))))))))
clauses))))
Caveat emptor.
If I understand you problem correctly, then you can use or and rely on the fact that when is evaluated to nil if the condition is not true, e.g.,
(defun example (a)
(or
(let ((b1 (+ 0 a)))
(when (eql b1 1)
(print "1")
b1))
(let ((b2 (+ 0 a)))
(when (eql b2 2)
(print "2")
b2))
(progn
(print "else")
a)))
Using macrolet is the best solution so far. That allows me to bypass the limitations of when-let and not all bindins in the let form have to evaluate to true.
(let ((a 3))
(let ((result nil))
(macrolet ((ret-go (res)
`(progn
(setf result ,res)
(go finish))))
(tagbody
(let ((b1 (+ 0 a)))
(when (eq b1 1)
(print "1")
(ret-go b1)))
(let ((b2 (+ 0 a)))
(when (eq b2 2)
(print "2")
(ret-go b2)))
(when T
(print "else")
(setf result a))
(format t "=== ~A~%" a)
finish)
result)))
In Common Lisp, I can get a function to pass around with the #' syntax, like this:
(let ((x #'+))
(funcall x 1 2))
But suppose I want to set a function so I don't have to use funcall for it. Does Common Lisp have a local function name table, or just the global one that is assigned to with defun?
Is there a way to assign to a function symbol other than defun? Or more generally: is there a way I can do something similar to this nonworking example:
(setf #'x #'+)
(x 1 2)
You can define a local function using
flet and labels:
(flet ((f (x) (1+ (* 2 x))))
(f 7))
==> 15
You can also set function definition of a symbol using fdefinition:
(setf (fdefinition 'f) #'+)
(f 1 2 3)
==> 6
Note that let binds the
value cell of the symbol while flet bind the function cell.
When the symbol appears in the "function" position, the "function"
cell is used, while when it appears in the "value" position, the "value"
cell is used:
(setf (symbol-function 'x) #'car)
(setf (symbol-value 'x) #'cdr)
(x '(1 . 2))
==> 1
(funcall x '(1 . 2))
==> 2
Similarly,
(flet ((x (o) (car o)))
(let ((x #'cdr))
(cons (x '(1 . 2))
(funcall x '(1 . 2)))))
==> (1 . 2)
This is the difference between Lisp-1 and Lisp-2.
Finally, note that CLISP is just one implementation of the language ANSI Common Lisp.
One option for getting this kind of behavior is to write a macro to do it.
(defmacro flet* (assignments &body body)
(let ((assignments (mapcar
(lambda (assn)
(list (first assn) '(&rest args)
(list 'apply (second assn) 'args)))
assignments)))
`(flet ,assignments ,#body)))
This macro translates flet* into flet + apply like this:
(flet* ((x #'+)
(y #'*))
(pprint (x 1 2))
(pprint (y 3 4))
(pprint (x (y 2 3) 4)))
Becomes:
(flet ((x (&rest args) (apply #'+ args))
(y (&rest args) (apply #'* args)))
(pprint (x 1 2))
(pprint (y 3 4))
(pprint (x (y 2 3) 4)))
I'm new to lisp, and have been trying to learn Common Lisp by diving in and writing some code. I've read plenty of documentation on the subject, but it's taking a while to really sink in.
I have written a couple of macros (? and ??) for performing unit tests, but I'm having some difficulty. The code is at the end of the post, to avoid cluttering the actual question.
Here is an example of usage:
(??
(? "Arithmetic tests"
(? "Addition"
(= (+ 1 2) 3)
(= (+ 1 2 3) 6)
(= (+ -1 -3) -4))))
And an example of output:
[Arithmetic tests]
[Addition]
(PASS) '(= (+ 1 2) 3)'
(PASS) '(= (+ 1 2 3) 6)'
(PASS) '(= (+ -1 -3) -4)'
Results: 3 tests passed, 0 tests failed
Now, the existing code works. Unfortunately, the (? ...) macro is ugly, verbose, resistant to change - and I'm pretty sure also badly structured. For example, do I really have to use a list to store pieces of output code and then emit the contents at the end?
I'd like to modify the macro to permit description strings (or symbols) to optionally follow each test, whereupon it would replace the test literal in the output, thus:
(??
(? "Arithmetic tests"
(? "Addition"
(= (+ 1 2) 3) "Adding 1 and 2 results in 3"
(= (+ 1 2 3) 6)
(= (+ -1 -3) -4))))
Output:
[Arithmetic tests]
[Addition]
(PASS) Adding 1 and 2 results in 3
(PASS) '(= (+ 1 2 3) 6)'
(PASS) '(= (+ -1 -3) -4)'
But unfortunately I can't find a sensible place in the macro to insert this change. Depending on where I put it, I get errors like you're not inside a backquote expression, label is not defined or body-forms is not defined. I know what these errors mean, but I can't find a way to avoid them.
Also, I'll be wanting to handle exceptions in the test, and treat that as a failure. Currently, there is no exception handling code - the test result is merely tested against nil. Again, it is not clear how I should add this functionality.
I'm thinking that maybe this macro is over-complex, due to my inexperience in writing macros; and perhaps if I simplify it, modification will be easier. I don't really want to separate it out into several smaller macros without good reason; but maybe there's a terser way to write it?
Can anyone help me out here, please?
A complete code listing follows:
(defmacro with-gensyms ((&rest names) &body body)
`(let ,(loop for n in names collect `(,n (gensym)))
,#body))
(defmacro while (condition &body body)
`(loop while ,condition do (progn ,#body)))
(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))))))
(defun starts-with-p (str1 str2)
"Determine whether `str1` starts with `str2`"
(let ((p (search str2 str1)))
(and p (= 0 p))))
(defmacro pop-first-char (string)
`(with-gensyms (c)
(if (> (length ,string) 0)
(progn
(setf c (schar ,string 0))
(if (> (length ,string) 1)
(setf ,string (subseq ,string 1))
(setf ,string ""))))
c))
(defmacro pop-chars (string count)
`(with-gensyms (result)
(setf result ())
(dotimes (index ,count)
(push (pop-first-char ,string) result))
result))
(defun format-ansi-codes (text)
(let ((result ()))
(while (> (length text) 0)
(cond
((starts-with-p text "\\e")
(push (code-char #o33) result)
(pop-chars text 2)
)
((starts-with-p text "\\r")
(push (code-char 13) result)
(pop-chars text 2)
)
(t (push (pop-first-char text) result))
))
(setf result (nreverse result))
(coerce result 'string)))
(defun kv-lookup (values key)
"Like getf, but works with 'keys as well as :keys, in both the list and the supplied key"
(setf key (if (typep key 'cons) (nth 1 key) key))
(while values
(let ((k (pop values)) (v (pop values)))
(setf k (if (typep k 'cons) (nth 1 k) k))
(if (eql (symbol-name key) (symbol-name k))
(return v)))))
(defun make-ansi-escape (ansi-name)
(let ((ansi-codes '( :normal "\\e[00m" :white "\\e[1;37m" :light-grey "\\e[0;37m" :dark-grey "\\e[1;30m"
:red "\\e[0;31m" :light-red "\\e[1;31m" :green "\\e[0;32m" :blue "\\e[1;34m" :dark-blue "\\e[1;34m"
:cyan "\\e[1;36m" :magenta "\\e[1;35m" :yellow "\\e[0;33m"
:bg-dark-grey "\\e[100m"
:bold "\\e[1m" :underline "\\e[4m"
:start-of-line "\\r" :clear-line "\\e[2K" :move-up "\\e[1A")))
(format-ansi-codes (kv-lookup ansi-codes ansi-name))
))
(defun format-ansi-escaped-arg (out-stream arg)
(cond
((typep arg 'symbol) (format out-stream "~a" (make-ansi-escape arg)))
((typep arg 'string) (format out-stream arg))
(t (format out-stream "~a" arg))
))
(defun format-ansi-escaped (out-stream &rest args)
(while args
(let ((arg (pop args)))
(if (typep arg 'list)
(let ((first-arg (eval (first arg))))
(format out-stream first-arg (second arg))
)
(format-ansi-escaped-arg out-stream arg)
))
))
(defmacro while-pop ((var sequence &optional result-form) &rest forms)
(with-gensyms (seq)
`(let (,var)
(progn
(do () ((not ,sequence))
(setf ,var (pop ,sequence))
(progn ,#forms))
,result-form))))
(defun report-start (form)
(format t "( ) '~a'~%" form))
(defun report-result (result form)
(format-ansi-escaped t "(" (if result :green :red) `("~:[FAIL~;PASS~]" ,result) :normal `(") '~a'~%" ,form))
result)
(defmacro ? (name &body body-forms)
"Run any number of test forms, optionally nested within further (?) calls, and print the results of each test"
(with-gensyms (result indent indent-string)
(if (not body-forms)
:empty
(progn
(setf result () indent 0 indent-string " ")
(cond
((typep (first body-forms) 'integer)
(setf indent (pop body-forms))))
`(progn
(format t "~v#{~A~:*~}" ,indent ,indent-string)
(format-ansi-escaped t "[" :white ,name :normal "]~%")
(with-gensyms (test-results)
(setf test-results ())
,(while-pop (body-form body-forms `(progn ,#(nreverse result)))
(cond
( (EQL (first body-form) '?)
(push `(progn
(setf test-results (append test-results (? ',(nth 1 body-form) ,(1+ indent) ,#(nthcdr 2 body-form))))
(format t "~%")
test-results
) result)
)
(t
(push `(progn
(format t "~v#{~A~:*~}" ,(1+ indent) ,indent-string)
(report-start ',body-form)
(with-gensyms (result label)
(setf result ,body-form)
(format-ansi-escaped t :move-up :start-of-line :clear-line)
(format t "~v#{~A~:*~}" ,(1+ indent) ,indent-string)
(push (report-result result ',body-form) test-results)
test-results
)) result))))))))))
(defun ?? (&rest results)
"Run any number of tests, and print a summary afterward"
(setf results (flatten results))
(format-ansi-escaped t "~&" :white "Results: " :green `("~a test~:p passed" ,(count t results)) :normal ", "
(if (find NIL results) :red :normal) `("~a test~:p failed" ,(count NIL results))
:yellow `("~[~:;, ~:*~a test~:p not run~]" ,(count :skip results))
:brown `("~[~:;, ~:*~a empty test group~:p skipped~]" ,(count :empty results))
:normal "~%"))
For my part, the ? macro is rather technical and it's hard to follow the logic behind the formatting functions. So instead of tracking errors I'd like to suggest my own attempt, perhaps it'll be of use.
I think that actually your ?? doesn't want to evaluate anything, but rather to treat its body as individual tests or sections. If the body includes a list starting with ?, this list represents a section; other elements are test forms optionally followed by descriptions. So in my implementation ?? will be a macro, and ? will be just a symbol.
I start with wishful thinking. I suppose I can create individual tests using a function make-test-item and test sections using a function make-test-section (their implementation is unimportant for now), that I can display them using an auxiliary function display-test and compute results using the function results, which returns two values: the total number of tests and the number of passed ones. Then I'd like the code
(??
(? "Arithmetic tests"
(? "Addition"
(= (+ 1 2) 3) "Adding 1 and 2 results in 3"
(= (+ 1 2 3) 6)
(= (+ -1 -3) 4))
(? "Subtraction"
(= (- 1 2) 1)))
(= (sin 0) 0) "Sine of 0 equals 0")
to expand into something like
(let ((tests (list (make-test-section :header "Arithmetic tests"
:items (list (make-test-section :header "Addition"
:items (list (make-test-item :form '(= (+ 1 2) 3)
:description "Adding 1 and 2 results in 3"
:passp (= (+ 1 2) 3))
(make-test-item :form '(= (+ 1 2 3) 6)
:passp (= (+ 1 2 3) 6))
(make-test-item :form '(= (+ -1 -3) 4)
:passp (= (+ -1 -3) 4))))
(make-test-section :header "Subtraction"
:items (list (make-test-item :form '(= (- 1 2) 1)
:passp (= (- 1 2) 1))))))
(make-test-item :form '(= (sin 0) 0)
:passp (= (sin 0) 0)
:description "Sine of 0 equals 0"))))
(loop for test in tests
with total = 0
with passed = 0
do (display-test test 0 t)
do (multiple-value-bind (ttl p) (results test)
(incf total ttl)
(incf passed p))
finally (display-result total passed t)))
Here a list of tests is created; then we traverse it printing each test (0 denotes the zero level of indentation and t is as in format) and keeping track of the results, finally displaying the total results. I don't think explicit eval is needed here.
It may not be the most exquisite piece of code ever, but it seems manageable. I supply missing definitions below, they are rather trivial (and can be improved) and have nothing to do with macros.
Now we pass on to the macros. Consider both pieces of code as data, then we want a list processing function which would turn the first one into the second. A few auxiliary functions would come in handy.
The major task is to parse the body of ?? and generate the list of test to go inside the let.
(defun test-item-form (form description)
`(make-test-item :form ',form :description ,description :passp ,form))
(defun test-section-form (header items)
`(make-test-section :header ,header :items (list ,#items)))
(defun parse-test (forms)
(let (new-forms)
(loop
(when (null forms)
(return (nreverse new-forms)))
(let ((f (pop forms)))
(cond ((and (listp f) (eq (first f) '?))
(push (test-section-form (second f) (parse-test (nthcdr 2 f))) new-forms))
((stringp (first forms))
(push (test-item-form f (pop forms)) new-forms))
(t (push (test-item-form f nil) new-forms)))))))
Here parse-test essentially absorbs the syntax of ??. Each iteration consumes one or two forms and collects corresponding make-... forms. The functions can be easily tested in REPL (and, of course, I did test them while writing).
Now the macro becomes quite simple:
(defmacro ?? (&body body)
`(let ((tests (list ,#(parse-test body))))
(loop for test in tests
with total = 0
with passed = 0
do (display-test test 0 t)
do (multiple-value-bind (ttl p) (results test)
(incf total ttl)
(incf passed p))
finally (display-result total passed t))))
It captures a few symbols, both in the variable name space and in the function one (the expansion may contain make-test-item and make-test-section). A clean solution with gensyms would be cumbersome, so I'd suggest just moving all the definitions in a separate package and exporting only ?? and ?.
For completeness, here is an implementation of the test API. Actually, it's what I started coding with and proceeded until I made sure the big let-form works; then I passed on to the macro part. This implementation is fairly sloppy; in particular, it doesn't support terminal colours and display-test can't even output a section into a string.
(defstruct test-item form description passp)
(defstruct test-section header items)
(defun results (test)
(etypecase test
(test-item (if (test-item-passp test)
(values 1 1)
(values 1 0)))
(test-section (let ((items-count 0)
(passed-count 0))
(dolist (i (test-section-items test) (values items-count passed-count))
(multiple-value-bind (i p) (results i)
(incf items-count i)
(incf passed-count p)))))))
(defparameter *test-indent* 2)
(defun display-test-item (i level stream)
(format stream "~V,0T~:[(FAIL)~;(PASS)~] ~:['~S'~;~:*~A~]~%"
(* level *test-indent*)
(test-item-passp i)
(test-item-description i)
(test-item-form i)))
(defun display-test-section-header (s level stream)
(format stream "~V,0T[~A]~%"
(* level *test-indent*)
(test-section-header s)))
(defun display-test (test level stream)
(etypecase test
(test-item (display-test-item test level stream))
(test-section
(display-test-section-header test level stream)
(dolist (i (test-section-items test))
(display-test i (1+ level) stream)))))
(defun display-result (total passed stream)
(format stream "Results: ~D test~:P passed, ~D test~:P failed.~%" passed (- total passed)))
All the code is licenced under WTFPL.
(defun merge-matrix (matrix-1 matrix-2)
(if (not (or (eql (matrix-rows matrix-1) (matrix-rows matrix-2)) (null matrix-1) (null matrix-2))) (error "Invalid dimensions."))
(cond
((null matrix-1) (copy-tree matrix-2))
((null matrix-2) (copy-tree matrix-1))
(t (let ((result (copy-tree matrix-1)))
(dotimes (i (matrix-rows matrix-1))
(setf (nth i result) (nconc (nth i result) (nth i matrix-2))))
result))))
(merge-matrix '((3 1) (1 3)) '((4 2) (1 1)))
*** - EVAL: variable NULL has no value
I receive an error like that how I can fix the problem, thanks
The OP's code works for me. However I felt motivated to improve it and
I implemented the same idea (but a bit more powerful).
The semantics are the same as Matlab's vertcat.
The function appends all arguments into one big matrix.
Note that due to the declarations my code should be super efficient.
(deftype mat ()
"Non-square matrices. Last index is columns, i.e. row-major order."
`(simple-array single-float 2))
(defun are-all-elements-typep (type ls)
(reduce #'(lambda (b x) (and b (typep x type)))
ls))
(defun are-all-matrix-heights-equalp (ls)
(let ((first-height (array-dimension (first ls) 0)))
(reduce #'(lambda (b x) (and b
(= first-height
(array-dimension x 0))))
ls)))
(defun vertcat (&rest rest)
(declare (type cons rest))
(unless (are-all-elements-typep 'mat rest)
(break "At least one of the arguments isn't a matrix."))
(unless (are-all-matrix-heights-equalp rest)
(break "All Matrices must have the same number of rows."))
(let* ((height (array-dimension (first rest) 0))
(widths (mapcar #'(lambda (mat) (array-dimension mat 1)) rest))
(result (make-array (list height
(reduce #'+ widths))
:element-type 'single-float))
(current-width 0))
(dotimes (m (length rest))
(let ((e (elt rest m)))
(destructuring-bind (y x) (array-dimensions e)
(dotimes (j y)
(dotimes (i x)
(setf (aref result j (+ current-width i))
(aref e j i))))
(incf current-width (elt widths m)))))
(the mat result)))
#+nil
(let ((a (make-array '(2 3)
:initial-contents '((1s0 2s0 3s0)
(2s0 4s0 5s0))
:element-type 'single-float))
(b (make-array '(2 2)
:initial-contents '((6s0 7s0)
(9s0 8s0))
:element-type 'single-float)))
(vertcat a b a))
;=> #2A ((1.0 2.0 3.0 6.0 7.0 1.0 2.0 3.0) (2.0 4.0 5.0 9.0 8.0 2.0 4.0 5.0))
The error message you're getting suggests that lisp is trying to treat one of your calls to null as a variable. I was able to replicate this behavior by defining matrix-rows like Frank Shearar did and deleting the parentheses around the ((null matrix-1) (copy-tree matrix-2)) s-expression, for example. I'd suggest you check your parentheses, either manually or using something like SLIME, which gave me a warning when I tried to compile the function.
Learning Common Lisp (using GNU CLISP 2.43) .. so might be a noob mistake. Example is the 'print prime numbers between x and y'
(defun is-prime (n)
(if (< n 2) (return-from is-prime NIL))
(do ((i 2 (1+ i)))
((= i n) T)
(if (= (mod n i) 0)
(return NIL))))
(defun next-prime-after (n)
(do ((i (1+ n) (1+ i)))
((is-prime i) i)))
(defmacro do-primes-v2 ((var start end) &body body)
`(do ((,var (if (is-prime ,start)
,start
(next-prime-after ,start))
(next-prime-after ,var)))
((> ,var ,end))
,#body))
(defmacro do-primes-v3 ((var start end) &body body)
(let ((loop-start (gensym))
(loop-end (gensym)))
`(do ((,loop-start ,start)
(,loop-end ,end)
(,var (if (is-prime ,loop-start)
,loop-start
(next-prime-after ,loop-start))
(next-prime-after ,var)))
((> ,var ,loop-end))
,#body )))
do-primes-v2 works perfectly.
[13]> (do-primes-v2 (p 10 25) (format t "~d " p))
11 13 17 19 23
Next I tried using gensym to avoid naming clashes in macro expansion - do-primes-v3. However I'm stuck with a
*** - EVAL: variable #:G3498 has no value
Tried using macro-expand to see if i could spot the mistake but I can't.
[16]> (macroexpand-1 `(do-primes-v3 (p 10 25) (format t "~d " p)))
(DO
((#:G3502 10) (#:G3503 25)
(P (IF (IS-PRIME #:G3502) #:G3502 (NEXT-PRIME-AFTER #:G3502))
(NEXT-PRIME-AFTER P)))
((> P #:G3503)) (FORMAT T "~d " P)) ;
Use DO* instead of DO.
DO Initializes the bindings in a scope where they are not yet visible. DO* initializes the bindings in a scope where they are visible.
In this particular case var needs to reference the other binding loop-start.
You don't actually need the gensym here for avoiding variable capture, because you do not introduce any variables that would be "local to the macro". When you macroexpand your do-primes-v2, you will see that no variable is introduced that didn't exist outside of the macro.
You do need it for a different thing, though: avoiding multiple evaluation.
If you call the macro like this:
(do-primes-v2 (p (* x 2) (* y 3))
(format "~a~%" p))
it expands to
(do ((p (if (is-prime (* x 2))
(* x 2)
(next-prime-after (* x 2))
(next-prime-after p)))
((> p (* y 3))
(format "~a~%" p))
At best, this is inefficient, because those multiplications are done multiple times. However, if you use a function with side effects as inputs, like setf or incf, this can be a big problem.
Either move the binding of your loop-start and loop-end to an enclosing LET block or use DO*. The reason is that all loop variables in DO are bound "in parallel", so for the first binding, the (expanded) loop-start variable does not yet have a binding.
I know this doesn't really answer your question, but I do think it is relevant. In my experience, the type of macro you are attempting to write is a very common one. One problem I have with the way you have approached the problem is that it doesn't handle another common use case: functional composition.
I don't have the time to highlight some of the difficulties you will probably encounter using your macro, I will however highlight that, had you built your prime iterator geared towards functional composition, your macro turns out to be extremely simple, avoiding your question altogether.
Note: I have slightly modified some of your functions.
(defun is-prime (n)
(cond
((< n 2)
nil)
((= n 2)
t)
((evenp n)
nil)
(t
(do ((i 2 (1+ i)))
((= i n) t)
(when (or (= (mod n i) 0))
(return nil))))))
(defun next-prime (n)
(do ((i n (1+ i)))
((is-prime i) i)))
(defun prime-iterator (start-at)
(let ((current start-at))
(lambda ()
(let ((next-prime (next-prime current)))
(setf current (1+ next-prime))
next-prime))))
(defun map-primes/iterator (fn iterator end)
(do ((i (funcall iterator) (funcall iterator)))
((>= i end) nil)
(funcall fn i)))
(defun map-primes (fn start end)
(let ((iterator (prime-iterator start)))
(map-primes/iterator fn iterator end)))
(defmacro do-primes ((var start end) &body body)
`(map-primes #'(lambda (,var)
,#body)
,start ,end))
I too recommend that you look at Series. The generator pattern is also a very common occurrence in lisp programs. You may also want to look at Alexandria, in particular the function ALEXANDRIA:COMPOSE to see what cool stuff you can do with functional composition.
I suggest avoiding DO/DO* and macros altogether and instead going for Series (an implementation of which can be found on series.sourceforge.net).
If that's too complex then consider just generating a list of primes with recursion or a generator (for on-demand generation).