Mercurial > emacs
annotate lispref/functions.texi @ 15005:4d6af48d7446
Under DJGPP v2, use <sys/config.h> instead of most of config.h.in.
author | Richard M. Stallman <rms@gnu.org> |
---|---|
date | Fri, 12 Apr 1996 15:41:24 +0000 |
parents | a6eb5f12b0f3 |
children | e3abbf649bee |
rev | line source |
---|---|
6455 | 1 @c -*-texinfo-*- |
2 @c This is part of the GNU Emacs Lisp Reference Manual. | |
3 @c Copyright (C) 1990, 1991, 1992, 1993, 1994 Free Software Foundation, Inc. | |
4 @c See the file elisp.texi for copying conditions. | |
5 @setfilename ../info/functions | |
6 @node Functions, Macros, Variables, Top | |
7 @chapter Functions | |
8 | |
9 A Lisp program is composed mainly of Lisp functions. This chapter | |
10 explains what functions are, how they accept arguments, and how to | |
11 define them. | |
12 | |
13 @menu | |
14 * What Is a Function:: Lisp functions vs. primitives; terminology. | |
15 * Lambda Expressions:: How functions are expressed as Lisp objects. | |
16 * Function Names:: A symbol can serve as the name of a function. | |
17 * Defining Functions:: Lisp expressions for defining functions. | |
18 * Calling Functions:: How to use an existing function. | |
19 * Mapping Functions:: Applying a function to each element of a list, etc. | |
20 * Anonymous Functions:: Lambda expressions are functions with no names. | |
21 * Function Cells:: Accessing or setting the function definition | |
22 of a symbol. | |
23 * Inline Functions:: Defining functions that the compiler will open code. | |
24 * Related Topics:: Cross-references to specific Lisp primitives | |
25 that have a special bearing on how functions work. | |
26 @end menu | |
27 | |
28 @node What Is a Function | |
29 @section What Is a Function? | |
30 | |
31 In a general sense, a function is a rule for carrying on a computation | |
32 given several values called @dfn{arguments}. The result of the | |
33 computation is called the value of the function. The computation can | |
34 also have side effects: lasting changes in the values of variables or | |
35 the contents of data structures. | |
36 | |
37 Here are important terms for functions in Emacs Lisp and for other | |
38 function-like objects. | |
39 | |
40 @table @dfn | |
41 @item function | |
42 @cindex function | |
43 In Emacs Lisp, a @dfn{function} is anything that can be applied to | |
44 arguments in a Lisp program. In some cases, we use it more | |
45 specifically to mean a function written in Lisp. Special forms and | |
46 macros are not functions. | |
47 | |
48 @item primitive | |
49 @cindex primitive | |
50 @cindex subr | |
51 @cindex built-in function | |
52 A @dfn{primitive} is a function callable from Lisp that is written in C, | |
53 such as @code{car} or @code{append}. These functions are also called | |
54 @dfn{built-in} functions or @dfn{subrs}. (Special forms are also | |
55 considered primitives.) | |
56 | |
57 Usually the reason that a function is a primitives is because it is | |
58 fundamental, because it provides a low-level interface to operating | |
59 system services, or because it needs to run fast. Primitives can be | |
60 modified or added only by changing the C sources and recompiling the | |
61 editor. See @ref{Writing Emacs Primitives}. | |
62 | |
63 @item lambda expression | |
64 A @dfn{lambda expression} is a function written in Lisp. | |
65 These are described in the following section. | |
66 @ifinfo | |
67 @xref{Lambda Expressions}. | |
68 @end ifinfo | |
69 | |
70 @item special form | |
71 A @dfn{special form} is a primitive that is like a function but does not | |
72 evaluate all of its arguments in the usual way. It may evaluate only | |
73 some of the arguments, or may evaluate them in an unusual order, or | |
74 several times. Many special forms are described in @ref{Control | |
75 Structures}. | |
76 | |
77 @item macro | |
78 @cindex macro | |
79 A @dfn{macro} is a construct defined in Lisp by the programmer. It | |
80 differs from a function in that it translates a Lisp expression that you | |
81 write into an equivalent expression to be evaluated instead of the | |
12098 | 82 original expression. Macros enable Lisp programmers to do the sorts of |
83 things that special forms can do. @xref{Macros}, for how to define and | |
84 use macros. | |
6455 | 85 |
86 @item command | |
87 @cindex command | |
88 A @dfn{command} is an object that @code{command-execute} can invoke; it | |
89 is a possible definition for a key sequence. Some functions are | |
90 commands; a function written in Lisp is a command if it contains an | |
91 interactive declaration (@pxref{Defining Commands}). Such a function | |
92 can be called from Lisp expressions like other functions; in this case, | |
93 the fact that the function is a command makes no difference. | |
94 | |
95 Keyboard macros (strings and vectors) are commands also, even though | |
96 they are not functions. A symbol is a command if its function | |
97 definition is a command; such symbols can be invoked with @kbd{M-x}. | |
98 The symbol is a function as well if the definition is a function. | |
99 @xref{Command Overview}. | |
100 | |
101 @item keystroke command | |
102 @cindex keystroke command | |
103 A @dfn{keystroke command} is a command that is bound to a key sequence | |
104 (typically one to three keystrokes). The distinction is made here | |
105 merely to avoid confusion with the meaning of ``command'' in non-Emacs | |
106 editors; for Lisp programs, the distinction is normally unimportant. | |
107 | |
108 @item byte-code function | |
109 A @dfn{byte-code function} is a function that has been compiled by the | |
110 byte compiler. @xref{Byte-Code Type}. | |
111 @end table | |
112 | |
113 @defun subrp object | |
114 This function returns @code{t} if @var{object} is a built-in function | |
115 (i.e., a Lisp primitive). | |
116 | |
117 @example | |
118 @group | |
119 (subrp 'message) ; @r{@code{message} is a symbol,} | |
120 @result{} nil ; @r{not a subr object.} | |
121 @end group | |
122 @group | |
123 (subrp (symbol-function 'message)) | |
124 @result{} t | |
125 @end group | |
126 @end example | |
127 @end defun | |
128 | |
129 @defun byte-code-function-p object | |
130 This function returns @code{t} if @var{object} is a byte-code | |
131 function. For example: | |
132 | |
133 @example | |
134 @group | |
135 (byte-code-function-p (symbol-function 'next-line)) | |
136 @result{} t | |
137 @end group | |
138 @end example | |
139 @end defun | |
140 | |
141 @node Lambda Expressions | |
142 @section Lambda Expressions | |
143 @cindex lambda expression | |
144 | |
145 A function written in Lisp is a list that looks like this: | |
146 | |
147 @example | |
148 (lambda (@var{arg-variables}@dots{}) | |
149 @r{[}@var{documentation-string}@r{]} | |
150 @r{[}@var{interactive-declaration}@r{]} | |
151 @var{body-forms}@dots{}) | |
152 @end example | |
153 | |
154 @noindent | |
12098 | 155 Such a list is called a @dfn{lambda expression}. In Emacs Lisp, it |
156 actually is valid as an expression---it evaluates to itself. In some | |
157 other Lisp dialects, a lambda expression is not a valid expression at | |
158 all. In either case, its main use is not to be evaluated as an | |
159 expression, but to be called as a function. | |
6455 | 160 |
161 @menu | |
162 * Lambda Components:: The parts of a lambda expression. | |
163 * Simple Lambda:: A simple example. | |
164 * Argument List:: Details and special features of argument lists. | |
165 * Function Documentation:: How to put documentation in a function. | |
166 @end menu | |
167 | |
168 @node Lambda Components | |
169 @subsection Components of a Lambda Expression | |
170 | |
171 @ifinfo | |
172 | |
173 A function written in Lisp (a ``lambda expression'') is a list that | |
174 looks like this: | |
175 | |
176 @example | |
177 (lambda (@var{arg-variables}@dots{}) | |
178 [@var{documentation-string}] | |
179 [@var{interactive-declaration}] | |
180 @var{body-forms}@dots{}) | |
181 @end example | |
182 @end ifinfo | |
183 | |
184 @cindex lambda list | |
185 The first element of a lambda expression is always the symbol | |
186 @code{lambda}. This indicates that the list represents a function. The | |
187 reason functions are defined to start with @code{lambda} is so that | |
188 other lists, intended for other uses, will not accidentally be valid as | |
189 functions. | |
190 | |
191 The second element is a list of symbols--the argument variable names. | |
192 This is called the @dfn{lambda list}. When a Lisp function is called, | |
193 the argument values are matched up against the variables in the lambda | |
194 list, which are given local bindings with the values provided. | |
195 @xref{Local Variables}. | |
196 | |
7193 | 197 The documentation string is a Lisp string object placed within the |
198 function definition to describe the function for the Emacs help | |
199 facilities. @xref{Function Documentation}. | |
6455 | 200 |
201 The interactive declaration is a list of the form @code{(interactive | |
202 @var{code-string})}. This declares how to provide arguments if the | |
203 function is used interactively. Functions with this declaration are called | |
204 @dfn{commands}; they can be called using @kbd{M-x} or bound to a key. | |
205 Functions not intended to be called in this way should not have interactive | |
206 declarations. @xref{Defining Commands}, for how to write an interactive | |
207 declaration. | |
208 | |
209 @cindex body of function | |
210 The rest of the elements are the @dfn{body} of the function: the Lisp | |
211 code to do the work of the function (or, as a Lisp programmer would say, | |
212 ``a list of Lisp forms to evaluate''). The value returned by the | |
213 function is the value returned by the last element of the body. | |
214 | |
215 @node Simple Lambda | |
216 @subsection A Simple Lambda-Expression Example | |
217 | |
218 Consider for example the following function: | |
219 | |
220 @example | |
221 (lambda (a b c) (+ a b c)) | |
222 @end example | |
223 | |
224 @noindent | |
225 We can call this function by writing it as the @sc{car} of an | |
226 expression, like this: | |
227 | |
228 @example | |
229 @group | |
230 ((lambda (a b c) (+ a b c)) | |
231 1 2 3) | |
232 @end group | |
233 @end example | |
234 | |
235 @noindent | |
236 This call evaluates the body of the lambda expression with the variable | |
237 @code{a} bound to 1, @code{b} bound to 2, and @code{c} bound to 3. | |
238 Evaluation of the body adds these three numbers, producing the result 6; | |
239 therefore, this call to the function returns the value 6. | |
240 | |
241 Note that the arguments can be the results of other function calls, as in | |
242 this example: | |
243 | |
244 @example | |
245 @group | |
246 ((lambda (a b c) (+ a b c)) | |
247 1 (* 2 3) (- 5 4)) | |
248 @end group | |
249 @end example | |
250 | |
251 @noindent | |
252 This evaluates the arguments @code{1}, @code{(* 2 3)}, and @code{(- 5 | |
7193 | 253 4)} from left to right. Then it applies the lambda expression to the |
254 argument values 1, 6 and 1 to produce the value 8. | |
6455 | 255 |
256 It is not often useful to write a lambda expression as the @sc{car} of | |
257 a form in this way. You can get the same result, of making local | |
258 variables and giving them values, using the special form @code{let} | |
259 (@pxref{Local Variables}). And @code{let} is clearer and easier to use. | |
260 In practice, lambda expressions are either stored as the function | |
261 definitions of symbols, to produce named functions, or passed as | |
262 arguments to other functions (@pxref{Anonymous Functions}). | |
263 | |
264 However, calls to explicit lambda expressions were very useful in the | |
265 old days of Lisp, before the special form @code{let} was invented. At | |
266 that time, they were the only way to bind and initialize local | |
267 variables. | |
268 | |
269 @node Argument List | |
270 @subsection Advanced Features of Argument Lists | |
271 @kindex wrong-number-of-arguments | |
272 @cindex argument binding | |
273 @cindex binding arguments | |
274 | |
275 Our simple sample function, @code{(lambda (a b c) (+ a b c))}, | |
276 specifies three argument variables, so it must be called with three | |
277 arguments: if you try to call it with only two arguments or four | |
278 arguments, you get a @code{wrong-number-of-arguments} error. | |
279 | |
280 It is often convenient to write a function that allows certain | |
281 arguments to be omitted. For example, the function @code{substring} | |
282 accepts three arguments---a string, the start index and the end | |
283 index---but the third argument defaults to the @var{length} of the | |
284 string if you omit it. It is also convenient for certain functions to | |
7193 | 285 accept an indefinite number of arguments, as the functions @code{list} |
6455 | 286 and @code{+} do. |
287 | |
288 @cindex optional arguments | |
289 @cindex rest arguments | |
290 @kindex &optional | |
291 @kindex &rest | |
292 To specify optional arguments that may be omitted when a function | |
293 is called, simply include the keyword @code{&optional} before the optional | |
294 arguments. To specify a list of zero or more extra arguments, include the | |
295 keyword @code{&rest} before one final argument. | |
296 | |
297 Thus, the complete syntax for an argument list is as follows: | |
298 | |
299 @example | |
300 @group | |
301 (@var{required-vars}@dots{} | |
302 @r{[}&optional @var{optional-vars}@dots{}@r{]} | |
303 @r{[}&rest @var{rest-var}@r{]}) | |
304 @end group | |
305 @end example | |
306 | |
307 @noindent | |
308 The square brackets indicate that the @code{&optional} and @code{&rest} | |
309 clauses, and the variables that follow them, are optional. | |
310 | |
311 A call to the function requires one actual argument for each of the | |
312 @var{required-vars}. There may be actual arguments for zero or more of | |
313 the @var{optional-vars}, and there cannot be any actual arguments beyond | |
314 that unless the lambda list uses @code{&rest}. In that case, there may | |
315 be any number of extra actual arguments. | |
316 | |
317 If actual arguments for the optional and rest variables are omitted, | |
7193 | 318 then they always default to @code{nil}. There is no way for the |
6455 | 319 function to distinguish between an explicit argument of @code{nil} and |
7193 | 320 an omitted argument. However, the body of the function is free to |
321 consider @code{nil} an abbreviation for some other meaningful value. | |
322 This is what @code{substring} does; @code{nil} as the third argument to | |
323 @code{substring} means to use the length of the string supplied. | |
6455 | 324 |
325 @cindex CL note---default optional arg | |
326 @quotation | |
327 @b{Common Lisp note:} Common Lisp allows the function to specify what | |
328 default value to use when an optional argument is omitted; Emacs Lisp | |
329 always uses @code{nil}. | |
330 @end quotation | |
331 | |
332 For example, an argument list that looks like this: | |
333 | |
334 @example | |
335 (a b &optional c d &rest e) | |
336 @end example | |
337 | |
338 @noindent | |
339 binds @code{a} and @code{b} to the first two actual arguments, which are | |
340 required. If one or two more arguments are provided, @code{c} and | |
341 @code{d} are bound to them respectively; any arguments after the first | |
342 four are collected into a list and @code{e} is bound to that list. If | |
343 there are only two arguments, @code{c} is @code{nil}; if two or three | |
344 arguments, @code{d} is @code{nil}; if four arguments or fewer, @code{e} | |
345 is @code{nil}. | |
346 | |
347 There is no way to have required arguments following optional | |
348 ones---it would not make sense. To see why this must be so, suppose | |
349 that @code{c} in the example were optional and @code{d} were required. | |
350 Suppose three actual arguments are given; which variable would the third | |
351 argument be for? Similarly, it makes no sense to have any more | |
352 arguments (either required or optional) after a @code{&rest} argument. | |
353 | |
354 Here are some examples of argument lists and proper calls: | |
355 | |
356 @smallexample | |
357 ((lambda (n) (1+ n)) ; @r{One required:} | |
358 1) ; @r{requires exactly one argument.} | |
359 @result{} 2 | |
360 ((lambda (n &optional n1) ; @r{One required and one optional:} | |
361 (if n1 (+ n n1) (1+ n))) ; @r{1 or 2 arguments.} | |
362 1 2) | |
363 @result{} 3 | |
364 ((lambda (n &rest ns) ; @r{One required and one rest:} | |
365 (+ n (apply '+ ns))) ; @r{1 or more arguments.} | |
366 1 2 3 4 5) | |
367 @result{} 15 | |
368 @end smallexample | |
369 | |
370 @node Function Documentation | |
371 @subsection Documentation Strings of Functions | |
372 @cindex documentation of function | |
373 | |
374 A lambda expression may optionally have a @dfn{documentation string} just | |
375 after the lambda list. This string does not affect execution of the | |
376 function; it is a kind of comment, but a systematized comment which | |
377 actually appears inside the Lisp world and can be used by the Emacs help | |
378 facilities. @xref{Documentation}, for how the @var{documentation-string} is | |
379 accessed. | |
380 | |
12098 | 381 It is a good idea to provide documentation strings for all the |
382 functions in your program, even those that are only called from within | |
383 your program. Documentation strings are like comments, except that they | |
384 are easier to access. | |
6455 | 385 |
386 The first line of the documentation string should stand on its own, | |
387 because @code{apropos} displays just this first line. It should consist | |
388 of one or two complete sentences that summarize the function's purpose. | |
389 | |
12098 | 390 The start of the documentation string is usually indented in the source file, |
391 but since these spaces come before the starting double-quote, they are not part of | |
6455 | 392 the string. Some people make a practice of indenting any additional |
7193 | 393 lines of the string so that the text lines up in the program source. |
394 @emph{This is a mistake.} The indentation of the following lines is | |
395 inside the string; what looks nice in the source code will look ugly | |
396 when displayed by the help commands. | |
6455 | 397 |
398 You may wonder how the documentation string could be optional, since | |
399 there are required components of the function that follow it (the body). | |
400 Since evaluation of a string returns that string, without any side effects, | |
401 it has no effect if it is not the last form in the body. Thus, in | |
402 practice, there is no confusion between the first form of the body and the | |
403 documentation string; if the only body form is a string then it serves both | |
404 as the return value and as the documentation. | |
405 | |
406 @node Function Names | |
407 @section Naming a Function | |
408 @cindex function definition | |
409 @cindex named function | |
410 @cindex function name | |
411 | |
412 In most computer languages, every function has a name; the idea of a | |
413 function without a name is nonsensical. In Lisp, a function in the | |
414 strictest sense has no name. It is simply a list whose first element is | |
415 @code{lambda}, or a primitive subr-object. | |
416 | |
417 However, a symbol can serve as the name of a function. This happens | |
418 when you put the function in the symbol's @dfn{function cell} | |
419 (@pxref{Symbol Components}). Then the symbol itself becomes a valid, | |
420 callable function, equivalent to the list or subr-object that its | |
421 function cell refers to. The contents of the function cell are also | |
422 called the symbol's @dfn{function definition}. The procedure of using a | |
423 symbol's function definition in place of the symbol is called | |
424 @dfn{symbol function indirection}; see @ref{Function Indirection}. | |
425 | |
426 In practice, nearly all functions are given names in this way and | |
427 referred to through their names. For example, the symbol @code{car} works | |
428 as a function and does what it does because the primitive subr-object | |
429 @code{#<subr car>} is stored in its function cell. | |
430 | |
431 We give functions names because it is convenient to refer to them by | |
432 their names in Lisp expressions. For primitive subr-objects such as | |
433 @code{#<subr car>}, names are the only way you can refer to them: there | |
434 is no read syntax for such objects. For functions written in Lisp, the | |
435 name is more convenient to use in a call than an explicit lambda | |
436 expression. Also, a function with a name can refer to itself---it can | |
437 be recursive. Writing the function's name in its own definition is much | |
438 more convenient than making the function definition point to itself | |
439 (something that is not impossible but that has various disadvantages in | |
440 practice). | |
441 | |
442 We often identify functions with the symbols used to name them. For | |
443 example, we often speak of ``the function @code{car}'', not | |
444 distinguishing between the symbol @code{car} and the primitive | |
445 subr-object that is its function definition. For most purposes, there | |
446 is no need to distinguish. | |
447 | |
448 Even so, keep in mind that a function need not have a unique name. While | |
449 a given function object @emph{usually} appears in the function cell of only | |
450 one symbol, this is just a matter of convenience. It is easy to store | |
451 it in several symbols using @code{fset}; then each of the symbols is | |
452 equally well a name for the same function. | |
453 | |
454 A symbol used as a function name may also be used as a variable; | |
455 these two uses of a symbol are independent and do not conflict. | |
456 | |
457 @node Defining Functions | |
7115
9a9e88e65617
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
6455
diff
changeset
|
458 @section Defining Functions |
6455 | 459 @cindex defining a function |
460 | |
461 We usually give a name to a function when it is first created. This | |
462 is called @dfn{defining a function}, and it is done with the | |
463 @code{defun} special form. | |
464 | |
465 @defspec defun name argument-list body-forms | |
466 @code{defun} is the usual way to define new Lisp functions. It | |
467 defines the symbol @var{name} as a function that looks like this: | |
468 | |
469 @example | |
470 (lambda @var{argument-list} . @var{body-forms}) | |
471 @end example | |
472 | |
473 @code{defun} stores this lambda expression in the function cell of | |
474 @var{name}. It returns the value @var{name}, but usually we ignore this | |
475 value. | |
476 | |
477 As described previously (@pxref{Lambda Expressions}), | |
478 @var{argument-list} is a list of argument names and may include the | |
479 keywords @code{&optional} and @code{&rest}. Also, the first two forms | |
480 in @var{body-forms} may be a documentation string and an interactive | |
481 declaration. | |
482 | |
483 There is no conflict if the same symbol @var{name} is also used as a | |
484 variable, since the symbol's value cell is independent of the function | |
485 cell. @xref{Symbol Components}. | |
486 | |
487 Here are some examples: | |
488 | |
489 @example | |
490 @group | |
491 (defun foo () 5) | |
492 @result{} foo | |
493 @end group | |
494 @group | |
495 (foo) | |
496 @result{} 5 | |
497 @end group | |
498 | |
499 @group | |
500 (defun bar (a &optional b &rest c) | |
501 (list a b c)) | |
502 @result{} bar | |
503 @end group | |
504 @group | |
505 (bar 1 2 3 4 5) | |
506 @result{} (1 2 (3 4 5)) | |
507 @end group | |
508 @group | |
509 (bar 1) | |
510 @result{} (1 nil nil) | |
511 @end group | |
512 @group | |
513 (bar) | |
514 @error{} Wrong number of arguments. | |
515 @end group | |
516 | |
517 @group | |
518 (defun capitalize-backwards () | |
519 "Upcase the last letter of a word." | |
520 (interactive) | |
521 (backward-word 1) | |
522 (forward-word 1) | |
523 (backward-char 1) | |
524 (capitalize-word 1)) | |
525 @result{} capitalize-backwards | |
526 @end group | |
527 @end example | |
528 | |
529 Be careful not to redefine existing functions unintentionally. | |
530 @code{defun} redefines even primitive functions such as @code{car} | |
531 without any hesitation or notification. Redefining a function already | |
532 defined is often done deliberately, and there is no way to distinguish | |
533 deliberate redefinition from unintentional redefinition. | |
534 @end defspec | |
535 | |
536 @defun defalias name definition | |
537 This special form defines the symbol @var{name} as a function, with | |
7193 | 538 definition @var{definition} (which can be any valid Lisp function). |
12098 | 539 |
540 The proper place to use @code{defalias} is where a specific function | |
541 name is being defined---especially where that name appears explicitly in | |
542 the source file being loaded. This is because @code{defalias} records | |
543 which file defined the function, just like @code{defun} | |
544 (@pxref{Unloading}). | |
545 | |
546 By contrast, in programs that manipulate function definitions for other | |
547 purposes, it is better to use @code{fset}, which does not keep such | |
548 records. | |
6455 | 549 @end defun |
550 | |
12098 | 551 See also @code{defsubst}, which defines a function like @code{defun} |
552 and tells the Lisp compiler to open-code it. @xref{Inline Functions}. | |
553 | |
6455 | 554 @node Calling Functions |
555 @section Calling Functions | |
556 @cindex function invocation | |
557 @cindex calling a function | |
558 | |
559 Defining functions is only half the battle. Functions don't do | |
560 anything until you @dfn{call} them, i.e., tell them to run. Calling a | |
561 function is also known as @dfn{invocation}. | |
562 | |
7193 | 563 The most common way of invoking a function is by evaluating a list. |
564 For example, evaluating the list @code{(concat "a" "b")} calls the | |
565 function @code{concat} with arguments @code{"a"} and @code{"b"}. | |
566 @xref{Evaluation}, for a description of evaluation. | |
6455 | 567 |
568 When you write a list as an expression in your program, the function | |
569 name is part of the program. This means that you choose which function | |
570 to call, and how many arguments to give it, when you write the program. | |
571 Usually that's just what you want. Occasionally you need to decide at | |
572 run time which function to call. To do that, use the functions | |
573 @code{funcall} and @code{apply}. | |
574 | |
575 @defun funcall function &rest arguments | |
576 @code{funcall} calls @var{function} with @var{arguments}, and returns | |
577 whatever @var{function} returns. | |
578 | |
579 Since @code{funcall} is a function, all of its arguments, including | |
580 @var{function}, are evaluated before @code{funcall} is called. This | |
581 means that you can use any expression to obtain the function to be | |
582 called. It also means that @code{funcall} does not see the expressions | |
583 you write for the @var{arguments}, only their values. These values are | |
584 @emph{not} evaluated a second time in the act of calling @var{function}; | |
585 @code{funcall} enters the normal procedure for calling a function at the | |
586 place where the arguments have already been evaluated. | |
587 | |
588 The argument @var{function} must be either a Lisp function or a | |
589 primitive function. Special forms and macros are not allowed, because | |
590 they make sense only when given the ``unevaluated'' argument | |
591 expressions. @code{funcall} cannot provide these because, as we saw | |
592 above, it never knows them in the first place. | |
593 | |
594 @example | |
595 @group | |
596 (setq f 'list) | |
597 @result{} list | |
598 @end group | |
599 @group | |
600 (funcall f 'x 'y 'z) | |
601 @result{} (x y z) | |
602 @end group | |
603 @group | |
604 (funcall f 'x 'y '(z)) | |
605 @result{} (x y (z)) | |
606 @end group | |
607 @group | |
608 (funcall 'and t nil) | |
609 @error{} Invalid function: #<subr and> | |
610 @end group | |
611 @end example | |
612 | |
613 Compare these example with the examples of @code{apply}. | |
614 @end defun | |
615 | |
616 @defun apply function &rest arguments | |
617 @code{apply} calls @var{function} with @var{arguments}, just like | |
618 @code{funcall} but with one difference: the last of @var{arguments} is a | |
619 list of arguments to give to @var{function}, rather than a single | |
7193 | 620 argument. We also say that @code{apply} @dfn{spreads} this list so that |
621 each individual element becomes an argument. | |
6455 | 622 |
623 @code{apply} returns the result of calling @var{function}. As with | |
624 @code{funcall}, @var{function} must either be a Lisp function or a | |
625 primitive function; special forms and macros do not make sense in | |
626 @code{apply}. | |
627 | |
628 @example | |
629 @group | |
630 (setq f 'list) | |
631 @result{} list | |
632 @end group | |
633 @group | |
634 (apply f 'x 'y 'z) | |
635 @error{} Wrong type argument: listp, z | |
636 @end group | |
637 @group | |
638 (apply '+ 1 2 '(3 4)) | |
639 @result{} 10 | |
640 @end group | |
641 @group | |
642 (apply '+ '(1 2 3 4)) | |
643 @result{} 10 | |
644 @end group | |
645 | |
646 @group | |
647 (apply 'append '((a b c) nil (x y z) nil)) | |
648 @result{} (a b c x y z) | |
649 @end group | |
650 @end example | |
651 | |
652 For an interesting example of using @code{apply}, see the description of | |
653 @code{mapcar}, in @ref{Mapping Functions}. | |
654 @end defun | |
655 | |
656 @cindex functionals | |
657 It is common for Lisp functions to accept functions as arguments or | |
658 find them in data structures (especially in hook variables and property | |
659 lists) and call them using @code{funcall} or @code{apply}. Functions | |
660 that accept function arguments are often called @dfn{functionals}. | |
661 | |
12098 | 662 Sometimes, when you call a functional, it is useful to supply a no-op |
663 function as the argument. Here are two different kinds of no-op | |
6455 | 664 function: |
665 | |
666 @defun identity arg | |
667 This function returns @var{arg} and has no side effects. | |
668 @end defun | |
669 | |
670 @defun ignore &rest args | |
671 This function ignores any arguments and returns @code{nil}. | |
672 @end defun | |
673 | |
674 @node Mapping Functions | |
675 @section Mapping Functions | |
676 @cindex mapping functions | |
677 | |
678 A @dfn{mapping function} applies a given function to each element of a | |
679 list or other collection. Emacs Lisp has three such functions; | |
680 @code{mapcar} and @code{mapconcat}, which scan a list, are described | |
681 here. For the third mapping function, @code{mapatoms}, see | |
682 @ref{Creating Symbols}. | |
683 | |
684 @defun mapcar function sequence | |
7193 | 685 @code{mapcar} applies @var{function} to each element of @var{sequence} |
686 in turn, and returns a list of the results. | |
6455 | 687 |
7193 | 688 The argument @var{sequence} may be a list, a vector, or a string. The |
6455 | 689 result is always a list. The length of the result is the same as the |
690 length of @var{sequence}. | |
691 | |
692 @smallexample | |
693 @group | |
694 @exdent @r{For example:} | |
695 | |
696 (mapcar 'car '((a b) (c d) (e f))) | |
697 @result{} (a c e) | |
698 (mapcar '1+ [1 2 3]) | |
699 @result{} (2 3 4) | |
700 (mapcar 'char-to-string "abc") | |
701 @result{} ("a" "b" "c") | |
702 @end group | |
703 | |
704 @group | |
705 ;; @r{Call each function in @code{my-hooks}.} | |
706 (mapcar 'funcall my-hooks) | |
707 @end group | |
708 | |
709 @group | |
710 (defun mapcar* (f &rest args) | |
711 "Apply FUNCTION to successive cars of all ARGS. | |
712 Return the list of results." | |
713 ;; @r{If no list is exhausted,} | |
714 (if (not (memq 'nil args)) | |
715 ;; @r{apply function to @sc{CAR}s.} | |
716 (cons (apply f (mapcar 'car args)) | |
717 (apply 'mapcar* f | |
718 ;; @r{Recurse for rest of elements.} | |
719 (mapcar 'cdr args))))) | |
720 @end group | |
721 | |
722 @group | |
723 (mapcar* 'cons '(a b c) '(1 2 3 4)) | |
724 @result{} ((a . 1) (b . 2) (c . 3)) | |
725 @end group | |
726 @end smallexample | |
727 @end defun | |
728 | |
729 @defun mapconcat function sequence separator | |
730 @code{mapconcat} applies @var{function} to each element of | |
731 @var{sequence}: the results, which must be strings, are concatenated. | |
732 Between each pair of result strings, @code{mapconcat} inserts the string | |
733 @var{separator}. Usually @var{separator} contains a space or comma or | |
734 other suitable punctuation. | |
735 | |
736 The argument @var{function} must be a function that can take one | |
737 argument and return a string. | |
738 | |
739 @smallexample | |
740 @group | |
741 (mapconcat 'symbol-name | |
742 '(The cat in the hat) | |
743 " ") | |
744 @result{} "The cat in the hat" | |
745 @end group | |
746 | |
747 @group | |
748 (mapconcat (function (lambda (x) (format "%c" (1+ x)))) | |
749 "HAL-8000" | |
750 "") | |
751 @result{} "IBM.9111" | |
752 @end group | |
753 @end smallexample | |
754 @end defun | |
755 | |
756 @node Anonymous Functions | |
757 @section Anonymous Functions | |
758 @cindex anonymous function | |
759 | |
760 In Lisp, a function is a list that starts with @code{lambda}, a | |
761 byte-code function compiled from such a list, or alternatively a | |
762 primitive subr-object; names are ``extra''. Although usually functions | |
763 are defined with @code{defun} and given names at the same time, it is | |
764 occasionally more concise to use an explicit lambda expression---an | |
765 anonymous function. Such a list is valid wherever a function name is. | |
766 | |
767 Any method of creating such a list makes a valid function. Even this: | |
768 | |
769 @smallexample | |
770 @group | |
771 (setq silly (append '(lambda (x)) (list (list '+ (* 3 4) 'x)))) | |
772 @result{} (lambda (x) (+ 12 x)) | |
773 @end group | |
774 @end smallexample | |
775 | |
776 @noindent | |
777 This computes a list that looks like @code{(lambda (x) (+ 12 x))} and | |
778 makes it the value (@emph{not} the function definition!) of | |
779 @code{silly}. | |
780 | |
781 Here is how we might call this function: | |
782 | |
783 @example | |
784 @group | |
785 (funcall silly 1) | |
786 @result{} 13 | |
787 @end group | |
788 @end example | |
789 | |
790 @noindent | |
791 (It does @emph{not} work to write @code{(silly 1)}, because this function | |
792 is not the @emph{function definition} of @code{silly}. We have not given | |
793 @code{silly} any function definition, just a value as a variable.) | |
794 | |
795 Most of the time, anonymous functions are constants that appear in | |
796 your program. For example, you might want to pass one as an argument | |
797 to the function @code{mapcar}, which applies any given function to each | |
798 element of a list. Here we pass an anonymous function that multiplies | |
799 a number by two: | |
800 | |
801 @example | |
802 @group | |
803 (defun double-each (list) | |
804 (mapcar '(lambda (x) (* 2 x)) list)) | |
805 @result{} double-each | |
806 @end group | |
807 @group | |
808 (double-each '(2 11)) | |
809 @result{} (4 22) | |
810 @end group | |
811 @end example | |
812 | |
813 @noindent | |
814 In such cases, we usually use the special form @code{function} instead | |
815 of simple quotation to quote the anonymous function. | |
816 | |
817 @defspec function function-object | |
818 @cindex function quoting | |
819 This special form returns @var{function-object} without evaluating it. | |
820 In this, it is equivalent to @code{quote}. However, it serves as a | |
821 note to the Emacs Lisp compiler that @var{function-object} is intended | |
822 to be used only as a function, and therefore can safely be compiled. | |
823 Contrast this with @code{quote}, in @ref{Quoting}. | |
824 @end defspec | |
825 | |
826 Using @code{function} instead of @code{quote} makes a difference | |
827 inside a function or macro that you are going to compile. For example: | |
828 | |
829 @example | |
830 @group | |
831 (defun double-each (list) | |
832 (mapcar (function (lambda (x) (* 2 x))) list)) | |
833 @result{} double-each | |
834 @end group | |
835 @group | |
836 (double-each '(2 11)) | |
837 @result{} (4 22) | |
838 @end group | |
839 @end example | |
840 | |
841 @noindent | |
842 If this definition of @code{double-each} is compiled, the anonymous | |
843 function is compiled as well. By contrast, in the previous definition | |
844 where ordinary @code{quote} is used, the argument passed to | |
845 @code{mapcar} is the precise list shown: | |
846 | |
847 @example | |
848 (lambda (x) (* x 2)) | |
849 @end example | |
850 | |
851 @noindent | |
852 The Lisp compiler cannot assume this list is a function, even though it | |
853 looks like one, since it does not know what @code{mapcar} does with the | |
854 list. Perhaps @code{mapcar} will check that the @sc{car} of the third | |
855 element is the symbol @code{*}! The advantage of @code{function} is | |
856 that it tells the compiler to go ahead and compile the constant | |
857 function. | |
858 | |
859 We sometimes write @code{function} instead of @code{quote} when | |
860 quoting the name of a function, but this usage is just a sort of | |
861 comment. | |
862 | |
863 @example | |
864 (function @var{symbol}) @equiv{} (quote @var{symbol}) @equiv{} '@var{symbol} | |
865 @end example | |
866 | |
867 See @code{documentation} in @ref{Accessing Documentation}, for a | |
868 realistic example using @code{function} and an anonymous function. | |
869 | |
870 @node Function Cells | |
871 @section Accessing Function Cell Contents | |
872 | |
873 The @dfn{function definition} of a symbol is the object stored in the | |
874 function cell of the symbol. The functions described here access, test, | |
875 and set the function cell of symbols. | |
876 | |
7193 | 877 See also the function @code{indirect-function} in @ref{Function |
878 Indirection}. | |
879 | |
6455 | 880 @defun symbol-function symbol |
881 @kindex void-function | |
882 This returns the object in the function cell of @var{symbol}. If the | |
883 symbol's function cell is void, a @code{void-function} error is | |
884 signaled. | |
885 | |
886 This function does not check that the returned object is a legitimate | |
887 function. | |
888 | |
889 @example | |
890 @group | |
891 (defun bar (n) (+ n 2)) | |
892 @result{} bar | |
893 @end group | |
894 @group | |
895 (symbol-function 'bar) | |
896 @result{} (lambda (n) (+ n 2)) | |
897 @end group | |
898 @group | |
899 (fset 'baz 'bar) | |
900 @result{} bar | |
901 @end group | |
902 @group | |
903 (symbol-function 'baz) | |
904 @result{} bar | |
905 @end group | |
906 @end example | |
907 @end defun | |
908 | |
909 @cindex void function cell | |
910 If you have never given a symbol any function definition, we say that | |
911 that symbol's function cell is @dfn{void}. In other words, the function | |
912 cell does not have any Lisp object in it. If you try to call such a symbol | |
913 as a function, it signals a @code{void-function} error. | |
914 | |
915 Note that void is not the same as @code{nil} or the symbol | |
916 @code{void}. The symbols @code{nil} and @code{void} are Lisp objects, | |
917 and can be stored into a function cell just as any other object can be | |
918 (and they can be valid functions if you define them in turn with | |
7193 | 919 @code{defun}). A void function cell contains no object whatsoever. |
6455 | 920 |
921 You can test the voidness of a symbol's function definition with | |
922 @code{fboundp}. After you have given a symbol a function definition, you | |
923 can make it void once more using @code{fmakunbound}. | |
924 | |
925 @defun fboundp symbol | |
926 This function returns @code{t} if the symbol has an object in its | |
927 function cell, @code{nil} otherwise. It does not check that the object | |
928 is a legitimate function. | |
929 @end defun | |
930 | |
931 @defun fmakunbound symbol | |
932 This function makes @var{symbol}'s function cell void, so that a | |
933 subsequent attempt to access this cell will cause a @code{void-function} | |
934 error. (See also @code{makunbound}, in @ref{Local Variables}.) | |
935 | |
936 @example | |
937 @group | |
938 (defun foo (x) x) | |
939 @result{} x | |
940 @end group | |
941 @group | |
7193 | 942 (foo 1) |
943 @result{}1 | |
944 @end group | |
945 @group | |
6455 | 946 (fmakunbound 'foo) |
947 @result{} x | |
948 @end group | |
949 @group | |
950 (foo 1) | |
951 @error{} Symbol's function definition is void: foo | |
952 @end group | |
953 @end example | |
954 @end defun | |
955 | |
956 @defun fset symbol object | |
957 This function stores @var{object} in the function cell of @var{symbol}. | |
958 The result is @var{object}. Normally @var{object} should be a function | |
959 or the name of a function, but this is not checked. | |
960 | |
961 There are three normal uses of this function: | |
962 | |
963 @itemize @bullet | |
964 @item | |
965 Copying one symbol's function definition to another. (In other words, | |
966 making an alternate name for a function.) | |
967 | |
968 @item | |
969 Giving a symbol a function definition that is not a list and therefore | |
7193 | 970 cannot be made with @code{defun}. For example, you can use @code{fset} |
971 to give a symbol @code{s1} a function definition which is another symbol | |
972 @code{s2}; then @code{s1} serves as an alias for whatever definition | |
973 @code{s2} presently has. | |
6455 | 974 |
975 @item | |
976 In constructs for defining or altering functions. If @code{defun} | |
977 were not a primitive, it could be written in Lisp (as a macro) using | |
978 @code{fset}. | |
979 @end itemize | |
980 | |
981 Here are examples of the first two uses: | |
982 | |
983 @example | |
984 @group | |
985 ;; @r{Give @code{first} the same definition @code{car} has.} | |
986 (fset 'first (symbol-function 'car)) | |
987 @result{} #<subr car> | |
988 @end group | |
989 @group | |
990 (first '(1 2 3)) | |
991 @result{} 1 | |
992 @end group | |
993 | |
994 @group | |
995 ;; @r{Make the symbol @code{car} the function definition of @code{xfirst}.} | |
996 (fset 'xfirst 'car) | |
997 @result{} car | |
998 @end group | |
999 @group | |
1000 (xfirst '(1 2 3)) | |
1001 @result{} 1 | |
1002 @end group | |
1003 @group | |
1004 (symbol-function 'xfirst) | |
1005 @result{} car | |
1006 @end group | |
1007 @group | |
1008 (symbol-function (symbol-function 'xfirst)) | |
1009 @result{} #<subr car> | |
1010 @end group | |
1011 | |
1012 @group | |
1013 ;; @r{Define a named keyboard macro.} | |
1014 (fset 'kill-two-lines "\^u2\^k") | |
1015 @result{} "\^u2\^k" | |
1016 @end group | |
1017 @end example | |
7193 | 1018 |
1019 See also the related function @code{defalias}, in @ref{Defining | |
1020 Functions}. | |
6455 | 1021 @end defun |
1022 | |
1023 When writing a function that extends a previously defined function, | |
12098 | 1024 the following idiom is sometimes used: |
6455 | 1025 |
1026 @example | |
1027 (fset 'old-foo (symbol-function 'foo)) | |
1028 (defun foo () | |
1029 "Just like old-foo, except more so." | |
1030 @group | |
1031 (old-foo) | |
1032 (more-so)) | |
1033 @end group | |
1034 @end example | |
1035 | |
1036 @noindent | |
1037 This does not work properly if @code{foo} has been defined to autoload. | |
1038 In such a case, when @code{foo} calls @code{old-foo}, Lisp attempts | |
1039 to define @code{old-foo} by loading a file. Since this presumably | |
1040 defines @code{foo} rather than @code{old-foo}, it does not produce the | |
1041 proper results. The only way to avoid this problem is to make sure the | |
1042 file is loaded before moving aside the old definition of @code{foo}. | |
1043 | |
12098 | 1044 But it is unmodular and unclean, in any case, for a Lisp file to |
1045 redefine a function defined elsewhere. | |
1046 | |
6455 | 1047 @node Inline Functions |
1048 @section Inline Functions | |
1049 @cindex inline functions | |
1050 | |
1051 @findex defsubst | |
1052 You can define an @dfn{inline function} by using @code{defsubst} instead | |
1053 of @code{defun}. An inline function works just like an ordinary | |
1054 function except for one thing: when you compile a call to the function, | |
1055 the function's definition is open-coded into the caller. | |
1056 | |
1057 Making a function inline makes explicit calls run faster. But it also | |
1058 has disadvantages. For one thing, it reduces flexibility; if you change | |
1059 the definition of the function, calls already inlined still use the old | |
1060 definition until you recompile them. Since the flexibility of | |
1061 redefining functions is an important feature of Emacs, you should not | |
1062 make a function inline unless its speed is really crucial. | |
1063 | |
1064 Another disadvantage is that making a large function inline can increase | |
1065 the size of compiled code both in files and in memory. Since the speed | |
1066 advantage of inline functions is greatest for small functions, you | |
1067 generally should not make large functions inline. | |
1068 | |
1069 It's possible to define a macro to expand into the same code that an | |
1070 inline function would execute. But the macro would have a limitation: | |
1071 you can use it only explicitly---a macro cannot be called with | |
1072 @code{apply}, @code{mapcar} and so on. Also, it takes some work to | |
1073 convert an ordinary function into a macro. (@xref{Macros}.) To convert | |
1074 it into an inline function is very easy; simply replace @code{defun} | |
1075 with @code{defsubst}. Since each argument of an inline function is | |
1076 evaluated exactly once, you needn't worry about how many times the | |
1077 body uses the arguments, as you do for macros. (@xref{Argument | |
1078 Evaluation}.) | |
1079 | |
7193 | 1080 Inline functions can be used and open-coded later on in the same file, |
6455 | 1081 following the definition, just like macros. |
1082 | |
12098 | 1083 @c Emacs versions prior to 19 did not have inline functions. |
6455 | 1084 |
1085 @node Related Topics | |
1086 @section Other Topics Related to Functions | |
1087 | |
1088 Here is a table of several functions that do things related to | |
1089 function calling and function definitions. They are documented | |
1090 elsewhere, but we provide cross references here. | |
1091 | |
1092 @table @code | |
1093 @item apply | |
1094 See @ref{Calling Functions}. | |
1095 | |
1096 @item autoload | |
1097 See @ref{Autoload}. | |
1098 | |
1099 @item call-interactively | |
1100 See @ref{Interactive Call}. | |
1101 | |
1102 @item commandp | |
1103 See @ref{Interactive Call}. | |
1104 | |
1105 @item documentation | |
1106 See @ref{Accessing Documentation}. | |
1107 | |
1108 @item eval | |
1109 See @ref{Eval}. | |
1110 | |
1111 @item funcall | |
1112 See @ref{Calling Functions}. | |
1113 | |
1114 @item ignore | |
1115 See @ref{Calling Functions}. | |
1116 | |
1117 @item indirect-function | |
1118 See @ref{Function Indirection}. | |
1119 | |
1120 @item interactive | |
1121 See @ref{Using Interactive}. | |
1122 | |
1123 @item interactive-p | |
1124 See @ref{Interactive Call}. | |
1125 | |
1126 @item mapatoms | |
1127 See @ref{Creating Symbols}. | |
1128 | |
1129 @item mapcar | |
1130 See @ref{Mapping Functions}. | |
1131 | |
1132 @item mapconcat | |
1133 See @ref{Mapping Functions}. | |
1134 | |
1135 @item undefined | |
1136 See @ref{Key Lookup}. | |
1137 @end table | |
1138 |