2 @c This is part of the GNU Emacs Lisp Reference Manual.
3 @c Copyright (C) 1990, 1991, 1992, 1993, 1994, 1995, 1998, 2001, 2002,
4 @c 2003, 2004, 2005, 2006, 2007 Free Software Foundation, Inc.
5 @c See the file elisp.texi for copying conditions.
6 @setfilename ../../info/macros
7 @node Macros, Customization, Functions, Top
11 @dfn{Macros} enable you to define new control constructs and other
12 language features. A macro is defined much like a function, but instead
13 of telling how to compute a value, it tells how to compute another Lisp
14 expression which will in turn compute the value. We call this
15 expression the @dfn{expansion} of the macro.
17 Macros can do this because they operate on the unevaluated expressions
18 for the arguments, not on the argument values as functions do. They can
19 therefore construct an expansion containing these argument expressions
22 If you are using a macro to do something an ordinary function could
23 do, just for the sake of speed, consider using an inline function
24 instead. @xref{Inline Functions}.
27 * Simple Macro:: A basic example.
28 * Expansion:: How, when and why macros are expanded.
29 * Compiling Macros:: How macros are expanded by the compiler.
30 * Defining Macros:: How to write a macro definition.
31 * Backquote:: Easier construction of list structure.
32 * Problems with Macros:: Don't evaluate the macro arguments too many times.
33 Don't hide the user's variables.
34 * Indenting Macros:: Specifying how to indent macro calls.
38 @section A Simple Example of a Macro
40 Suppose we would like to define a Lisp construct to increment a
41 variable value, much like the @code{++} operator in C. We would like to
42 write @code{(inc x)} and have the effect of @code{(setq x (1+ x))}.
43 Here's a macro definition that does the job:
49 (list 'setq var (list '1+ var)))
53 When this is called with @code{(inc x)}, the argument @var{var} is the
54 symbol @code{x}---@emph{not} the @emph{value} of @code{x}, as it would
55 be in a function. The body of the macro uses this to construct the
56 expansion, which is @code{(setq x (1+ x))}. Once the macro definition
57 returns this expansion, Lisp proceeds to evaluate it, thus incrementing
61 @section Expansion of a Macro Call
62 @cindex expansion of macros
65 A macro call looks just like a function call in that it is a list which
66 starts with the name of the macro. The rest of the elements of the list
67 are the arguments of the macro.
69 Evaluation of the macro call begins like evaluation of a function call
70 except for one crucial difference: the macro arguments are the actual
71 expressions appearing in the macro call. They are not evaluated before
72 they are given to the macro definition. By contrast, the arguments of a
73 function are results of evaluating the elements of the function call
76 Having obtained the arguments, Lisp invokes the macro definition just
77 as a function is invoked. The argument variables of the macro are bound
78 to the argument values from the macro call, or to a list of them in the
79 case of a @code{&rest} argument. And the macro body executes and
80 returns its value just as a function body does.
82 The second crucial difference between macros and functions is that the
83 value returned by the macro body is not the value of the macro call.
84 Instead, it is an alternate expression for computing that value, also
85 known as the @dfn{expansion} of the macro. The Lisp interpreter
86 proceeds to evaluate the expansion as soon as it comes back from the
89 Since the expansion is evaluated in the normal manner, it may contain
90 calls to other macros. It may even be a call to the same macro, though
93 You can see the expansion of a given macro call by calling
96 @defun macroexpand form &optional environment
97 @cindex macro expansion
98 This function expands @var{form}, if it is a macro call. If the result
99 is another macro call, it is expanded in turn, until something which is
100 not a macro call results. That is the value returned by
101 @code{macroexpand}. If @var{form} is not a macro call to begin with, it
102 is returned as given.
104 Note that @code{macroexpand} does not look at the subexpressions of
105 @var{form} (although some macro definitions may do so). Even if they
106 are macro calls themselves, @code{macroexpand} does not expand them.
108 The function @code{macroexpand} does not expand calls to inline functions.
109 Normally there is no need for that, since a call to an inline function is
110 no harder to understand than a call to an ordinary function.
112 If @var{environment} is provided, it specifies an alist of macro
113 definitions that shadow the currently defined macros. Byte compilation
119 (list 'setq var (list '1+ var)))
124 (macroexpand '(inc r))
125 @result{} (setq r (1+ r))
129 (defmacro inc2 (var1 var2)
130 (list 'progn (list 'inc var1) (list 'inc var2)))
135 (macroexpand '(inc2 r s))
136 @result{} (progn (inc r) (inc s)) ; @r{@code{inc} not expanded here.}
142 @defun macroexpand-all form &optional environment
143 @code{macroexpand-all} expands macros like @code{macroexpand}, but
144 will look for and expand all macros in @var{form}, not just at the
145 top-level. If no macros are expanded, the return value is @code{eq}
148 Repeating the example used for @code{macroexpand} above with
149 @code{macroexpand-all}, we see that @code{macroexpand-all} @emph{does}
150 expand the embedded calls to @code{inc}:
153 (macroexpand-all '(inc2 r s))
154 @result{} (progn (setq r (1+ r)) (setq s (1+ s)))
159 @node Compiling Macros
160 @section Macros and Byte Compilation
161 @cindex byte-compiling macros
163 You might ask why we take the trouble to compute an expansion for a
164 macro and then evaluate the expansion. Why not have the macro body
165 produce the desired results directly? The reason has to do with
168 When a macro call appears in a Lisp program being compiled, the Lisp
169 compiler calls the macro definition just as the interpreter would, and
170 receives an expansion. But instead of evaluating this expansion, it
171 compiles the expansion as if it had appeared directly in the program.
172 As a result, the compiled code produces the value and side effects
173 intended for the macro, but executes at full compiled speed. This would
174 not work if the macro body computed the value and side effects
175 itself---they would be computed at compile time, which is not useful.
177 In order for compilation of macro calls to work, the macros must
178 already be defined in Lisp when the calls to them are compiled. The
179 compiler has a special feature to help you do this: if a file being
180 compiled contains a @code{defmacro} form, the macro is defined
181 temporarily for the rest of the compilation of that file. To make this
182 feature work, you must put the @code{defmacro} in the same file where it
183 is used, and before its first use.
185 Byte-compiling a file executes any @code{require} calls at top-level
186 in the file. This is in case the file needs the required packages for
187 proper compilation. One way to ensure that necessary macro definitions
188 are available during compilation is to require the files that define
189 them (@pxref{Named Features}). To avoid loading the macro definition files
190 when someone @emph{runs} the compiled program, write
191 @code{eval-when-compile} around the @code{require} calls (@pxref{Eval
194 @node Defining Macros
195 @section Defining Macros
197 A Lisp macro is a list whose @sc{car} is @code{macro}. Its @sc{cdr} should
198 be a function; expansion of the macro works by applying the function
199 (with @code{apply}) to the list of unevaluated argument-expressions
202 It is possible to use an anonymous Lisp macro just like an anonymous
203 function, but this is never done, because it does not make sense to pass
204 an anonymous macro to functionals such as @code{mapcar}. In practice,
205 all Lisp macros have names, and they are usually defined with the
206 special form @code{defmacro}.
208 @defspec defmacro name argument-list body-forms@dots{}
209 @code{defmacro} defines the symbol @var{name} as a macro that looks
213 (macro lambda @var{argument-list} . @var{body-forms})
216 (Note that the @sc{cdr} of this list is a function---a lambda expression.)
217 This macro object is stored in the function cell of @var{name}. The
218 value returned by evaluating the @code{defmacro} form is @var{name}, but
219 usually we ignore this value.
221 The shape and meaning of @var{argument-list} is the same as in a
222 function, and the keywords @code{&rest} and @code{&optional} may be used
223 (@pxref{Argument List}). Macros may have a documentation string, but
224 any @code{interactive} declaration is ignored since macros cannot be
225 called interactively.
228 The body of the macro definition can include a @code{declare} form,
229 which can specify how @key{TAB} should indent macro calls, and how to
230 step through them for Edebug.
232 @defmac declare @var{specs}@dots{}
233 @anchor{Definition of declare}
234 A @code{declare} form is used in a macro definition to specify various
235 additional information about it. Two kinds of specification are
239 @item (debug @var{edebug-form-spec})
240 Specify how to step through macro calls for Edebug.
241 @xref{Instrumenting Macro Calls}.
243 @item (indent @var{indent-spec})
244 Specify how to indent calls to this macro. @xref{Indenting Macros},
248 A @code{declare} form only has its special effect in the body of a
249 @code{defmacro} form if it immediately follows the documentation
250 string, if present, or the argument list otherwise. (Strictly
251 speaking, @emph{several} @code{declare} forms can follow the
252 documentation string or argument list, but since a @code{declare} form
253 can have several @var{specs}, they can always be combined into a
254 single form.) When used at other places in a @code{defmacro} form, or
255 outside a @code{defmacro} form, @code{declare} just returns @code{nil}
256 without evaluating any @var{specs}.
259 No macro absolutely needs a @code{declare} form, because that form
260 has no effect on how the macro expands, on what the macro means in the
261 program. It only affects secondary features: indentation and Edebug.
265 @cindex backquote (list substitution)
266 @cindex ` (list substitution)
269 Macros often need to construct large list structures from a mixture of
270 constants and nonconstant parts. To make this easier, use the @samp{`}
271 syntax (usually called @dfn{backquote}).
273 Backquote allows you to quote a list, but selectively evaluate
274 elements of that list. In the simplest case, it is identical to the
275 special form @code{quote} (@pxref{Quoting}). For example, these
276 two forms yield identical results:
280 `(a list of (+ 2 3) elements)
281 @result{} (a list of (+ 2 3) elements)
284 '(a list of (+ 2 3) elements)
285 @result{} (a list of (+ 2 3) elements)
289 @findex , @r{(with backquote)}
290 The special marker @samp{,} inside of the argument to backquote
291 indicates a value that isn't constant. Backquote evaluates the
292 argument of @samp{,} and puts the value in the list structure:
296 (list 'a 'list 'of (+ 2 3) 'elements)
297 @result{} (a list of 5 elements)
300 `(a list of ,(+ 2 3) elements)
301 @result{} (a list of 5 elements)
305 Substitution with @samp{,} is allowed at deeper levels of the list
306 structure also. For example:
310 (defmacro t-becomes-nil (variable)
311 `(if (eq ,variable t)
312 (setq ,variable nil)))
317 @equiv{} (if (eq foo t) (setq foo nil))
321 @findex ,@@ @r{(with backquote)}
322 @cindex splicing (with backquote)
323 You can also @dfn{splice} an evaluated value into the resulting list,
324 using the special marker @samp{,@@}. The elements of the spliced list
325 become elements at the same level as the other elements of the resulting
326 list. The equivalent code without using @samp{`} is often unreadable.
327 Here are some examples:
331 (setq some-list '(2 3))
335 (cons 1 (append some-list '(4) some-list))
336 @result{} (1 2 3 4 2 3)
339 `(1 ,@@some-list 4 ,@@some-list)
340 @result{} (1 2 3 4 2 3)
344 (setq list '(hack foo bar))
345 @result{} (hack foo bar)
350 (cons 'words (append (cdr list) '(as elements)))))
351 @result{} (use the words foo bar as elements)
354 `(use the words ,@@(cdr list) as elements)
355 @result{} (use the words foo bar as elements)
359 In old Emacs versions, before version 19.29, @samp{`} used a different
360 syntax which required an extra level of parentheses around the entire
361 backquote construct. Likewise, each @samp{,} or @samp{,@@} substitution
362 required an extra level of parentheses surrounding both the @samp{,} or
363 @samp{,@@} and the following expression. The old syntax required
364 whitespace between the @samp{`}, @samp{,} or @samp{,@@} and the
365 following expression.
367 This syntax is still accepted, for compatibility with old Emacs
368 versions, but support for it will be removed in the future.
370 @node Problems with Macros
371 @section Common Problems Using Macros
373 The basic facts of macro expansion have counterintuitive consequences.
374 This section describes some important consequences that can lead to
375 trouble, and rules to follow to avoid trouble.
378 * Wrong Time:: Do the work in the expansion, not in the macro.
379 * Argument Evaluation:: The expansion should evaluate each macro arg once.
380 * Surprising Local Vars:: Local variable bindings in the expansion
381 require special care.
382 * Eval During Expansion:: Don't evaluate them; put them in the expansion.
383 * Repeated Expansion:: Avoid depending on how many times expansion is done.
387 @subsection Wrong Time
389 The most common problem in writing macros is doing some of the
390 real work prematurely---while expanding the macro, rather than in the
391 expansion itself. For instance, one real package had this macro
395 (defmacro my-set-buffer-multibyte (arg)
396 (if (fboundp 'set-buffer-multibyte)
397 (set-buffer-multibyte arg)))
400 With this erroneous macro definition, the program worked fine when
401 interpreted but failed when compiled. This macro definition called
402 @code{set-buffer-multibyte} during compilation, which was wrong, and
403 then did nothing when the compiled package was run. The definition
404 that the programmer really wanted was this:
407 (defmacro my-set-buffer-multibyte (arg)
408 (if (fboundp 'set-buffer-multibyte)
409 `(set-buffer-multibyte ,arg)))
413 This macro expands, if appropriate, into a call to
414 @code{set-buffer-multibyte} that will be executed when the compiled
415 program is actually run.
417 @node Argument Evaluation
418 @subsection Evaluating Macro Arguments Repeatedly
420 When defining a macro you must pay attention to the number of times
421 the arguments will be evaluated when the expansion is executed. The
422 following macro (used to facilitate iteration) illustrates the problem.
423 This macro allows us to write a simple ``for'' loop such as one might
429 (defmacro for (var from init to final do &rest body)
430 "Execute a simple \"for\" loop.
431 For example, (for i from 1 to 10 do (print i))."
432 (list 'let (list (list var init))
433 (cons 'while (cons (list '<= var final)
434 (append body (list (list 'inc var)))))))
439 (for i from 1 to 3 do
440 (setq square (* i i))
441 (princ (format "\n%d %d" i square)))
447 (setq square (* i i))
448 (princ (format "\n%d %d" i square))
461 The arguments @code{from}, @code{to}, and @code{do} in this macro are
462 ``syntactic sugar''; they are entirely ignored. The idea is that you
463 will write noise words (such as @code{from}, @code{to}, and @code{do})
464 in those positions in the macro call.
466 Here's an equivalent definition simplified through use of backquote:
470 (defmacro for (var from init to final do &rest body)
471 "Execute a simple \"for\" loop.
472 For example, (for i from 1 to 10 do (print i))."
474 (while (<= ,var ,final)
480 Both forms of this definition (with backquote and without) suffer from
481 the defect that @var{final} is evaluated on every iteration. If
482 @var{final} is a constant, this is not a problem. If it is a more
483 complex form, say @code{(long-complex-calculation x)}, this can slow
484 down the execution significantly. If @var{final} has side effects,
485 executing it more than once is probably incorrect.
487 @cindex macro argument evaluation
488 A well-designed macro definition takes steps to avoid this problem by
489 producing an expansion that evaluates the argument expressions exactly
490 once unless repeated evaluation is part of the intended purpose of the
491 macro. Here is a correct expansion for the @code{for} macro:
498 (setq square (* i i))
499 (princ (format "%d %d" i square))
504 Here is a macro definition that creates this expansion:
508 (defmacro for (var from init to final do &rest body)
509 "Execute a simple for loop: (for i from 1 to 10 do (print i))."
518 Unfortunately, this fix introduces another problem,
519 described in the following section.
521 @node Surprising Local Vars
522 @subsection Local Variables in Macro Expansions
525 In the previous section, the definition of @code{for} was fixed as
526 follows to make the expansion evaluate the macro arguments the proper
531 (defmacro for (var from init to final do &rest body)
532 "Execute a simple for loop: (for i from 1 to 10 do (print i))."
544 The new definition of @code{for} has a new problem: it introduces a
545 local variable named @code{max} which the user does not expect. This
546 causes trouble in examples such as the following:
551 (for x from 0 to 10 do
552 (let ((this (frob x)))
559 The references to @code{max} inside the body of the @code{for}, which
560 are supposed to refer to the user's binding of @code{max}, really access
561 the binding made by @code{for}.
563 The way to correct this is to use an uninterned symbol instead of
564 @code{max} (@pxref{Creating Symbols}). The uninterned symbol can be
565 bound and referred to just like any other symbol, but since it is
566 created by @code{for}, we know that it cannot already appear in the
567 user's program. Since it is not interned, there is no way the user can
568 put it into the program later. It will never appear anywhere except
569 where put by @code{for}. Here is a definition of @code{for} that works
574 (defmacro for (var from init to final do &rest body)
575 "Execute a simple for loop: (for i from 1 to 10 do (print i))."
576 (let ((tempvar (make-symbol "max")))
579 (while (<= ,var ,tempvar)
586 This creates an uninterned symbol named @code{max} and puts it in the
587 expansion instead of the usual interned symbol @code{max} that appears
588 in expressions ordinarily.
590 @node Eval During Expansion
591 @subsection Evaluating Macro Arguments in Expansion
593 Another problem can happen if the macro definition itself
594 evaluates any of the macro argument expressions, such as by calling
595 @code{eval} (@pxref{Eval}). If the argument is supposed to refer to the
596 user's variables, you may have trouble if the user happens to use a
597 variable with the same name as one of the macro arguments. Inside the
598 macro body, the macro argument binding is the most local binding of this
599 variable, so any references inside the form being evaluated do refer to
600 it. Here is an example:
605 (list 'setq (eval a) t))
610 (foo x) @expansion{} (setq b t)
611 @result{} t ; @r{and @code{b} has been set.}
614 (foo a) @expansion{} (setq a t)
615 @result{} t ; @r{but this set @code{a}, not @code{c}.}
620 It makes a difference whether the user's variable is named @code{a} or
621 @code{x}, because @code{a} conflicts with the macro argument variable
624 Another problem with calling @code{eval} in a macro definition is that
625 it probably won't do what you intend in a compiled program. The
626 byte-compiler runs macro definitions while compiling the program, when
627 the program's own computations (which you might have wished to access
628 with @code{eval}) don't occur and its local variable bindings don't
631 To avoid these problems, @strong{don't evaluate an argument expression
632 while computing the macro expansion}. Instead, substitute the
633 expression into the macro expansion, so that its value will be computed
634 as part of executing the expansion. This is how the other examples in
637 @node Repeated Expansion
638 @subsection How Many Times is the Macro Expanded?
640 Occasionally problems result from the fact that a macro call is
641 expanded each time it is evaluated in an interpreted function, but is
642 expanded only once (during compilation) for a compiled function. If the
643 macro definition has side effects, they will work differently depending
644 on how many times the macro is expanded.
646 Therefore, you should avoid side effects in computation of the
647 macro expansion, unless you really know what you are doing.
649 One special kind of side effect can't be avoided: constructing Lisp
650 objects. Almost all macro expansions include constructed lists; that is
651 the whole point of most macros. This is usually safe; there is just one
652 case where you must be careful: when the object you construct is part of a
653 quoted constant in the macro expansion.
655 If the macro is expanded just once, in compilation, then the object is
656 constructed just once, during compilation. But in interpreted
657 execution, the macro is expanded each time the macro call runs, and this
658 means a new object is constructed each time.
660 In most clean Lisp code, this difference won't matter. It can matter
661 only if you perform side-effects on the objects constructed by the macro
662 definition. Thus, to avoid trouble, @strong{avoid side effects on
663 objects constructed by macro definitions}. Here is an example of how
664 such side effects can get you into trouble:
668 (defmacro empty-object ()
669 (list 'quote (cons nil nil)))
673 (defun initialize (condition)
674 (let ((object (empty-object)))
676 (setcar object condition))
682 If @code{initialize} is interpreted, a new list @code{(nil)} is
683 constructed each time @code{initialize} is called. Thus, no side effect
684 survives between calls. If @code{initialize} is compiled, then the
685 macro @code{empty-object} is expanded during compilation, producing a
686 single ``constant'' @code{(nil)} that is reused and altered each time
687 @code{initialize} is called.
689 One way to avoid pathological cases like this is to think of
690 @code{empty-object} as a funny kind of constant, not as a memory
691 allocation construct. You wouldn't use @code{setcar} on a constant such
692 as @code{'(nil)}, so naturally you won't use it on @code{(empty-object)}
695 @node Indenting Macros
696 @section Indenting Macros
698 You can use the @code{declare} form in the macro definition to
699 specify how to @key{TAB} should indent indent calls to the macro. You
703 (declare (indent @var{indent-spec}))
707 Here are the possibilities for @var{indent-spec}:
711 This is the same as no property---use the standard indentation pattern.
713 Handle this function like a @samp{def} construct: treat the second
714 line as the start of a @dfn{body}.
715 @item an integer, @var{number}
716 The first @var{number} arguments of the function are
717 @dfn{distinguished} arguments; the rest are considered the body
718 of the expression. A line in the expression is indented according to
719 whether the first argument on it is distinguished or not. If the
720 argument is part of the body, the line is indented @code{lisp-body-indent}
721 more columns than the open-parenthesis starting the containing
722 expression. If the argument is distinguished and is either the first
723 or second argument, it is indented @emph{twice} that many extra columns.
724 If the argument is distinguished and not the first or second argument,
725 the line uses the standard pattern.
726 @item a symbol, @var{symbol}
727 @var{symbol} should be a function name; that function is called to
728 calculate the indentation of a line within this expression. The
729 function receives two arguments:
732 The value returned by @code{parse-partial-sexp} (a Lisp primitive for
733 indentation and nesting computation) when it parses up to the
734 beginning of this line.
736 The position at which the line being indented begins.
739 It should return either a number, which is the number of columns of
740 indentation for that line, or a list whose car is such a number. The
741 difference between returning a number and returning a list is that a
742 number says that all following lines at the same nesting level should
743 be indented just like this one; a list says that following lines might
744 call for different indentations. This makes a difference when the
745 indentation is being computed by @kbd{C-M-q}; if the value is a
746 number, @kbd{C-M-q} need not recalculate indentation for the following
747 lines until the end of the list.
751 arch-tag: d4cce66d-1047-45c3-bfde-db6719d6e82b