Mercurial > emacs
annotate lispref/functions.texi @ 41828:b8e9dfdbefe3
*** empty log message ***
author | Eli Zaretskii <eliz@gnu.org> |
---|---|
date | Wed, 05 Dec 2001 18:41:34 +0000 |
parents | 0317f0f05c37 |
children | 3e0b4b8d4f7e |
rev | line source |
---|---|
6455 | 1 @c -*-texinfo-*- |
2 @c This is part of the GNU Emacs Lisp Reference Manual. | |
27189 | 3 @c Copyright (C) 1990, 1991, 1992, 1993, 1994, 1995, 1998, 1999 |
4 @c Free Software Foundation, Inc. | |
6455 | 5 @c See the file elisp.texi for copying conditions. |
6 @setfilename ../info/functions | |
7 @node Functions, Macros, Variables, Top | |
8 @chapter Functions | |
9 | |
10 A Lisp program is composed mainly of Lisp functions. This chapter | |
11 explains what functions are, how they accept arguments, and how to | |
12 define them. | |
13 | |
14 @menu | |
15 * What Is a Function:: Lisp functions vs. primitives; terminology. | |
16 * Lambda Expressions:: How functions are expressed as Lisp objects. | |
17 * Function Names:: A symbol can serve as the name of a function. | |
18 * Defining Functions:: Lisp expressions for defining functions. | |
19 * Calling Functions:: How to use an existing function. | |
20 * Mapping Functions:: Applying a function to each element of a list, etc. | |
21 * Anonymous Functions:: Lambda expressions are functions with no names. | |
22 * Function Cells:: Accessing or setting the function definition | |
23 of a symbol. | |
24 * Inline Functions:: Defining functions that the compiler will open code. | |
25 * Related Topics:: Cross-references to specific Lisp primitives | |
26 that have a special bearing on how functions work. | |
27 @end menu | |
28 | |
29 @node What Is a Function | |
30 @section What Is a Function? | |
31 | |
32 In a general sense, a function is a rule for carrying on a computation | |
33 given several values called @dfn{arguments}. The result of the | |
34 computation is called the value of the function. The computation can | |
35 also have side effects: lasting changes in the values of variables or | |
36 the contents of data structures. | |
37 | |
38 Here are important terms for functions in Emacs Lisp and for other | |
39 function-like objects. | |
40 | |
41 @table @dfn | |
42 @item function | |
43 @cindex function | |
44 In Emacs Lisp, a @dfn{function} is anything that can be applied to | |
45 arguments in a Lisp program. In some cases, we use it more | |
46 specifically to mean a function written in Lisp. Special forms and | |
47 macros are not functions. | |
48 | |
49 @item primitive | |
50 @cindex primitive | |
51 @cindex subr | |
52 @cindex built-in function | |
53 A @dfn{primitive} is a function callable from Lisp that is written in C, | |
54 such as @code{car} or @code{append}. These functions are also called | |
55 @dfn{built-in} functions or @dfn{subrs}. (Special forms are also | |
56 considered primitives.) | |
57 | |
22138
d4ac295a98b3
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21682
diff
changeset
|
58 Usually the reason we implement a function as a primitive is either |
d4ac295a98b3
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21682
diff
changeset
|
59 because it is fundamental, because it provides a low-level interface to |
d4ac295a98b3
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21682
diff
changeset
|
60 operating system services, or because it needs to run fast. Primitives |
d4ac295a98b3
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21682
diff
changeset
|
61 can be modified or added only by changing the C sources and recompiling |
d4ac295a98b3
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21682
diff
changeset
|
62 the editor. See @ref{Writing Emacs Primitives}. |
6455 | 63 |
64 @item lambda expression | |
65 A @dfn{lambda expression} is a function written in Lisp. | |
66 These are described in the following section. | |
27193 | 67 @ifnottex |
6455 | 68 @xref{Lambda Expressions}. |
27193 | 69 @end ifnottex |
6455 | 70 |
71 @item special form | |
72 A @dfn{special form} is a primitive that is like a function but does not | |
73 evaluate all of its arguments in the usual way. It may evaluate only | |
74 some of the arguments, or may evaluate them in an unusual order, or | |
75 several times. Many special forms are described in @ref{Control | |
76 Structures}. | |
77 | |
78 @item macro | |
79 @cindex macro | |
80 A @dfn{macro} is a construct defined in Lisp by the programmer. It | |
81 differs from a function in that it translates a Lisp expression that you | |
82 write into an equivalent expression to be evaluated instead of the | |
12098 | 83 original expression. Macros enable Lisp programmers to do the sorts of |
84 things that special forms can do. @xref{Macros}, for how to define and | |
85 use macros. | |
6455 | 86 |
87 @item command | |
88 @cindex command | |
89 A @dfn{command} is an object that @code{command-execute} can invoke; it | |
90 is a possible definition for a key sequence. Some functions are | |
91 commands; a function written in Lisp is a command if it contains an | |
92 interactive declaration (@pxref{Defining Commands}). Such a function | |
93 can be called from Lisp expressions like other functions; in this case, | |
94 the fact that the function is a command makes no difference. | |
95 | |
96 Keyboard macros (strings and vectors) are commands also, even though | |
97 they are not functions. A symbol is a command if its function | |
98 definition is a command; such symbols can be invoked with @kbd{M-x}. | |
99 The symbol is a function as well if the definition is a function. | |
100 @xref{Command Overview}. | |
101 | |
102 @item keystroke command | |
103 @cindex keystroke command | |
104 A @dfn{keystroke command} is a command that is bound to a key sequence | |
105 (typically one to three keystrokes). The distinction is made here | |
106 merely to avoid confusion with the meaning of ``command'' in non-Emacs | |
107 editors; for Lisp programs, the distinction is normally unimportant. | |
108 | |
109 @item byte-code function | |
110 A @dfn{byte-code function} is a function that has been compiled by the | |
111 byte compiler. @xref{Byte-Code Type}. | |
112 @end table | |
113 | |
22138
d4ac295a98b3
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21682
diff
changeset
|
114 @defun functionp object |
21682
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
115 This function returns @code{t} if @var{object} is any kind of function, |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
116 or a special form or macro. |
21007
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
117 @end defun |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
118 |
6455 | 119 @defun subrp object |
120 This function returns @code{t} if @var{object} is a built-in function | |
121 (i.e., a Lisp primitive). | |
122 | |
123 @example | |
124 @group | |
125 (subrp 'message) ; @r{@code{message} is a symbol,} | |
126 @result{} nil ; @r{not a subr object.} | |
127 @end group | |
128 @group | |
129 (subrp (symbol-function 'message)) | |
130 @result{} t | |
131 @end group | |
132 @end example | |
133 @end defun | |
134 | |
135 @defun byte-code-function-p object | |
136 This function returns @code{t} if @var{object} is a byte-code | |
137 function. For example: | |
138 | |
139 @example | |
140 @group | |
141 (byte-code-function-p (symbol-function 'next-line)) | |
142 @result{} t | |
143 @end group | |
144 @end example | |
145 @end defun | |
146 | |
33999 | 147 @defun subr-arity subr |
148 @tindex subr-arity | |
149 This function provides information about the argument list of a | |
150 primitive, @var{subr}. The returned value is a pair | |
151 @code{(@var{min} . @var{max})}. @var{min} is the minimum number of | |
152 args. @var{max} is the maximum number or the symbol @code{many}, for a | |
153 function with @code{&rest} arguments, or the symbol @code{unevalled} if | |
154 @var{subr} is a special form. | |
155 @end defun | |
156 | |
6455 | 157 @node Lambda Expressions |
158 @section Lambda Expressions | |
159 @cindex lambda expression | |
160 | |
161 A function written in Lisp is a list that looks like this: | |
162 | |
163 @example | |
164 (lambda (@var{arg-variables}@dots{}) | |
165 @r{[}@var{documentation-string}@r{]} | |
166 @r{[}@var{interactive-declaration}@r{]} | |
167 @var{body-forms}@dots{}) | |
168 @end example | |
169 | |
170 @noindent | |
12098 | 171 Such a list is called a @dfn{lambda expression}. In Emacs Lisp, it |
172 actually is valid as an expression---it evaluates to itself. In some | |
173 other Lisp dialects, a lambda expression is not a valid expression at | |
174 all. In either case, its main use is not to be evaluated as an | |
175 expression, but to be called as a function. | |
6455 | 176 |
177 @menu | |
178 * Lambda Components:: The parts of a lambda expression. | |
179 * Simple Lambda:: A simple example. | |
180 * Argument List:: Details and special features of argument lists. | |
181 * Function Documentation:: How to put documentation in a function. | |
182 @end menu | |
183 | |
184 @node Lambda Components | |
185 @subsection Components of a Lambda Expression | |
186 | |
27193 | 187 @ifnottex |
6455 | 188 |
189 A function written in Lisp (a ``lambda expression'') is a list that | |
190 looks like this: | |
191 | |
192 @example | |
193 (lambda (@var{arg-variables}@dots{}) | |
194 [@var{documentation-string}] | |
195 [@var{interactive-declaration}] | |
196 @var{body-forms}@dots{}) | |
197 @end example | |
27193 | 198 @end ifnottex |
6455 | 199 |
200 @cindex lambda list | |
201 The first element of a lambda expression is always the symbol | |
202 @code{lambda}. This indicates that the list represents a function. The | |
203 reason functions are defined to start with @code{lambda} is so that | |
204 other lists, intended for other uses, will not accidentally be valid as | |
205 functions. | |
206 | |
21007
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
207 The second element is a list of symbols---the argument variable names. |
6455 | 208 This is called the @dfn{lambda list}. When a Lisp function is called, |
209 the argument values are matched up against the variables in the lambda | |
210 list, which are given local bindings with the values provided. | |
211 @xref{Local Variables}. | |
212 | |
7193 | 213 The documentation string is a Lisp string object placed within the |
214 function definition to describe the function for the Emacs help | |
215 facilities. @xref{Function Documentation}. | |
6455 | 216 |
217 The interactive declaration is a list of the form @code{(interactive | |
218 @var{code-string})}. This declares how to provide arguments if the | |
219 function is used interactively. Functions with this declaration are called | |
220 @dfn{commands}; they can be called using @kbd{M-x} or bound to a key. | |
221 Functions not intended to be called in this way should not have interactive | |
222 declarations. @xref{Defining Commands}, for how to write an interactive | |
223 declaration. | |
224 | |
225 @cindex body of function | |
226 The rest of the elements are the @dfn{body} of the function: the Lisp | |
227 code to do the work of the function (or, as a Lisp programmer would say, | |
228 ``a list of Lisp forms to evaluate''). The value returned by the | |
229 function is the value returned by the last element of the body. | |
230 | |
231 @node Simple Lambda | |
232 @subsection A Simple Lambda-Expression Example | |
233 | |
234 Consider for example the following function: | |
235 | |
236 @example | |
237 (lambda (a b c) (+ a b c)) | |
238 @end example | |
239 | |
240 @noindent | |
241 We can call this function by writing it as the @sc{car} of an | |
242 expression, like this: | |
243 | |
244 @example | |
245 @group | |
246 ((lambda (a b c) (+ a b c)) | |
247 1 2 3) | |
248 @end group | |
249 @end example | |
250 | |
251 @noindent | |
252 This call evaluates the body of the lambda expression with the variable | |
253 @code{a} bound to 1, @code{b} bound to 2, and @code{c} bound to 3. | |
254 Evaluation of the body adds these three numbers, producing the result 6; | |
255 therefore, this call to the function returns the value 6. | |
256 | |
257 Note that the arguments can be the results of other function calls, as in | |
258 this example: | |
259 | |
260 @example | |
261 @group | |
262 ((lambda (a b c) (+ a b c)) | |
263 1 (* 2 3) (- 5 4)) | |
264 @end group | |
265 @end example | |
266 | |
267 @noindent | |
268 This evaluates the arguments @code{1}, @code{(* 2 3)}, and @code{(- 5 | |
7193 | 269 4)} from left to right. Then it applies the lambda expression to the |
270 argument values 1, 6 and 1 to produce the value 8. | |
6455 | 271 |
272 It is not often useful to write a lambda expression as the @sc{car} of | |
273 a form in this way. You can get the same result, of making local | |
274 variables and giving them values, using the special form @code{let} | |
275 (@pxref{Local Variables}). And @code{let} is clearer and easier to use. | |
276 In practice, lambda expressions are either stored as the function | |
277 definitions of symbols, to produce named functions, or passed as | |
278 arguments to other functions (@pxref{Anonymous Functions}). | |
279 | |
280 However, calls to explicit lambda expressions were very useful in the | |
281 old days of Lisp, before the special form @code{let} was invented. At | |
282 that time, they were the only way to bind and initialize local | |
283 variables. | |
284 | |
285 @node Argument List | |
21007
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
286 @subsection Other Features of Argument Lists |
6455 | 287 @kindex wrong-number-of-arguments |
288 @cindex argument binding | |
289 @cindex binding arguments | |
290 | |
291 Our simple sample function, @code{(lambda (a b c) (+ a b c))}, | |
292 specifies three argument variables, so it must be called with three | |
293 arguments: if you try to call it with only two arguments or four | |
294 arguments, you get a @code{wrong-number-of-arguments} error. | |
295 | |
296 It is often convenient to write a function that allows certain | |
297 arguments to be omitted. For example, the function @code{substring} | |
298 accepts three arguments---a string, the start index and the end | |
299 index---but the third argument defaults to the @var{length} of the | |
300 string if you omit it. It is also convenient for certain functions to | |
7193 | 301 accept an indefinite number of arguments, as the functions @code{list} |
6455 | 302 and @code{+} do. |
303 | |
304 @cindex optional arguments | |
305 @cindex rest arguments | |
306 @kindex &optional | |
307 @kindex &rest | |
308 To specify optional arguments that may be omitted when a function | |
309 is called, simply include the keyword @code{&optional} before the optional | |
310 arguments. To specify a list of zero or more extra arguments, include the | |
311 keyword @code{&rest} before one final argument. | |
312 | |
313 Thus, the complete syntax for an argument list is as follows: | |
314 | |
315 @example | |
316 @group | |
317 (@var{required-vars}@dots{} | |
318 @r{[}&optional @var{optional-vars}@dots{}@r{]} | |
319 @r{[}&rest @var{rest-var}@r{]}) | |
320 @end group | |
321 @end example | |
322 | |
323 @noindent | |
324 The square brackets indicate that the @code{&optional} and @code{&rest} | |
325 clauses, and the variables that follow them, are optional. | |
326 | |
327 A call to the function requires one actual argument for each of the | |
328 @var{required-vars}. There may be actual arguments for zero or more of | |
329 the @var{optional-vars}, and there cannot be any actual arguments beyond | |
330 that unless the lambda list uses @code{&rest}. In that case, there may | |
331 be any number of extra actual arguments. | |
332 | |
333 If actual arguments for the optional and rest variables are omitted, | |
7193 | 334 then they always default to @code{nil}. There is no way for the |
6455 | 335 function to distinguish between an explicit argument of @code{nil} and |
7193 | 336 an omitted argument. However, the body of the function is free to |
337 consider @code{nil} an abbreviation for some other meaningful value. | |
338 This is what @code{substring} does; @code{nil} as the third argument to | |
339 @code{substring} means to use the length of the string supplied. | |
6455 | 340 |
341 @cindex CL note---default optional arg | |
342 @quotation | |
343 @b{Common Lisp note:} Common Lisp allows the function to specify what | |
344 default value to use when an optional argument is omitted; Emacs Lisp | |
21007
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
345 always uses @code{nil}. Emacs Lisp does not support ``supplied-p'' |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
346 variables that tell you whether an argument was explicitly passed. |
6455 | 347 @end quotation |
348 | |
349 For example, an argument list that looks like this: | |
350 | |
351 @example | |
352 (a b &optional c d &rest e) | |
353 @end example | |
354 | |
355 @noindent | |
356 binds @code{a} and @code{b} to the first two actual arguments, which are | |
357 required. If one or two more arguments are provided, @code{c} and | |
358 @code{d} are bound to them respectively; any arguments after the first | |
359 four are collected into a list and @code{e} is bound to that list. If | |
360 there are only two arguments, @code{c} is @code{nil}; if two or three | |
361 arguments, @code{d} is @code{nil}; if four arguments or fewer, @code{e} | |
362 is @code{nil}. | |
363 | |
364 There is no way to have required arguments following optional | |
365 ones---it would not make sense. To see why this must be so, suppose | |
366 that @code{c} in the example were optional and @code{d} were required. | |
39168 | 367 Suppose three actual arguments are given; which variable would the |
368 third argument be for? Would it be used for the @var{c}, or for | |
369 @var{d}? One can argue for both possibilities. Similarly, it makes | |
370 no sense to have any more arguments (either required or optional) | |
371 after a @code{&rest} argument. | |
6455 | 372 |
373 Here are some examples of argument lists and proper calls: | |
374 | |
375 @smallexample | |
376 ((lambda (n) (1+ n)) ; @r{One required:} | |
377 1) ; @r{requires exactly one argument.} | |
378 @result{} 2 | |
379 ((lambda (n &optional n1) ; @r{One required and one optional:} | |
380 (if n1 (+ n n1) (1+ n))) ; @r{1 or 2 arguments.} | |
381 1 2) | |
382 @result{} 3 | |
383 ((lambda (n &rest ns) ; @r{One required and one rest:} | |
384 (+ n (apply '+ ns))) ; @r{1 or more arguments.} | |
385 1 2 3 4 5) | |
386 @result{} 15 | |
387 @end smallexample | |
388 | |
389 @node Function Documentation | |
390 @subsection Documentation Strings of Functions | |
391 @cindex documentation of function | |
392 | |
393 A lambda expression may optionally have a @dfn{documentation string} just | |
394 after the lambda list. This string does not affect execution of the | |
395 function; it is a kind of comment, but a systematized comment which | |
396 actually appears inside the Lisp world and can be used by the Emacs help | |
397 facilities. @xref{Documentation}, for how the @var{documentation-string} is | |
398 accessed. | |
399 | |
12098 | 400 It is a good idea to provide documentation strings for all the |
21682
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
401 functions in your program, even those that are called only from within |
12098 | 402 your program. Documentation strings are like comments, except that they |
403 are easier to access. | |
6455 | 404 |
405 The first line of the documentation string should stand on its own, | |
406 because @code{apropos} displays just this first line. It should consist | |
407 of one or two complete sentences that summarize the function's purpose. | |
408 | |
12098 | 409 The start of the documentation string is usually indented in the source file, |
410 but since these spaces come before the starting double-quote, they are not part of | |
6455 | 411 the string. Some people make a practice of indenting any additional |
7193 | 412 lines of the string so that the text lines up in the program source. |
413 @emph{This is a mistake.} The indentation of the following lines is | |
414 inside the string; what looks nice in the source code will look ugly | |
415 when displayed by the help commands. | |
6455 | 416 |
417 You may wonder how the documentation string could be optional, since | |
418 there are required components of the function that follow it (the body). | |
419 Since evaluation of a string returns that string, without any side effects, | |
420 it has no effect if it is not the last form in the body. Thus, in | |
421 practice, there is no confusion between the first form of the body and the | |
422 documentation string; if the only body form is a string then it serves both | |
423 as the return value and as the documentation. | |
424 | |
425 @node Function Names | |
426 @section Naming a Function | |
427 @cindex function definition | |
428 @cindex named function | |
429 @cindex function name | |
430 | |
431 In most computer languages, every function has a name; the idea of a | |
432 function without a name is nonsensical. In Lisp, a function in the | |
433 strictest sense has no name. It is simply a list whose first element is | |
21682
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
434 @code{lambda}, a byte-code function object, or a primitive subr-object. |
6455 | 435 |
436 However, a symbol can serve as the name of a function. This happens | |
437 when you put the function in the symbol's @dfn{function cell} | |
438 (@pxref{Symbol Components}). Then the symbol itself becomes a valid, | |
439 callable function, equivalent to the list or subr-object that its | |
440 function cell refers to. The contents of the function cell are also | |
441 called the symbol's @dfn{function definition}. The procedure of using a | |
442 symbol's function definition in place of the symbol is called | |
443 @dfn{symbol function indirection}; see @ref{Function Indirection}. | |
444 | |
445 In practice, nearly all functions are given names in this way and | |
446 referred to through their names. For example, the symbol @code{car} works | |
447 as a function and does what it does because the primitive subr-object | |
448 @code{#<subr car>} is stored in its function cell. | |
449 | |
450 We give functions names because it is convenient to refer to them by | |
451 their names in Lisp expressions. For primitive subr-objects such as | |
452 @code{#<subr car>}, names are the only way you can refer to them: there | |
453 is no read syntax for such objects. For functions written in Lisp, the | |
454 name is more convenient to use in a call than an explicit lambda | |
455 expression. Also, a function with a name can refer to itself---it can | |
456 be recursive. Writing the function's name in its own definition is much | |
457 more convenient than making the function definition point to itself | |
458 (something that is not impossible but that has various disadvantages in | |
459 practice). | |
460 | |
461 We often identify functions with the symbols used to name them. For | |
462 example, we often speak of ``the function @code{car}'', not | |
463 distinguishing between the symbol @code{car} and the primitive | |
464 subr-object that is its function definition. For most purposes, there | |
465 is no need to distinguish. | |
466 | |
467 Even so, keep in mind that a function need not have a unique name. While | |
468 a given function object @emph{usually} appears in the function cell of only | |
469 one symbol, this is just a matter of convenience. It is easy to store | |
470 it in several symbols using @code{fset}; then each of the symbols is | |
471 equally well a name for the same function. | |
472 | |
22138
d4ac295a98b3
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21682
diff
changeset
|
473 A symbol used as a function name may also be used as a variable; these |
d4ac295a98b3
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21682
diff
changeset
|
474 two uses of a symbol are independent and do not conflict. (Some Lisp |
d4ac295a98b3
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21682
diff
changeset
|
475 dialects, such as Scheme, do not distinguish between a symbol's value |
d4ac295a98b3
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21682
diff
changeset
|
476 and its function definition; a symbol's value as a variable is also its |
d4ac295a98b3
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21682
diff
changeset
|
477 function definition.) If you have not given a symbol a function |
d4ac295a98b3
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21682
diff
changeset
|
478 definition, you cannot use it as a function; whether the symbol has a |
d4ac295a98b3
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21682
diff
changeset
|
479 value as a variable makes no difference to this. |
6455 | 480 |
481 @node Defining Functions | |
7115
9a9e88e65617
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
6455
diff
changeset
|
482 @section Defining Functions |
6455 | 483 @cindex defining a function |
484 | |
485 We usually give a name to a function when it is first created. This | |
486 is called @dfn{defining a function}, and it is done with the | |
487 @code{defun} special form. | |
488 | |
489 @defspec defun name argument-list body-forms | |
490 @code{defun} is the usual way to define new Lisp functions. It | |
491 defines the symbol @var{name} as a function that looks like this: | |
492 | |
493 @example | |
494 (lambda @var{argument-list} . @var{body-forms}) | |
495 @end example | |
496 | |
497 @code{defun} stores this lambda expression in the function cell of | |
498 @var{name}. It returns the value @var{name}, but usually we ignore this | |
499 value. | |
500 | |
501 As described previously (@pxref{Lambda Expressions}), | |
502 @var{argument-list} is a list of argument names and may include the | |
21007
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
503 keywords @code{&optional} and @code{&rest}. Also, the first two of the |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
504 @var{body-forms} may be a documentation string and an interactive |
6455 | 505 declaration. |
506 | |
507 There is no conflict if the same symbol @var{name} is also used as a | |
508 variable, since the symbol's value cell is independent of the function | |
509 cell. @xref{Symbol Components}. | |
510 | |
511 Here are some examples: | |
512 | |
513 @example | |
514 @group | |
515 (defun foo () 5) | |
516 @result{} foo | |
517 @end group | |
518 @group | |
519 (foo) | |
520 @result{} 5 | |
521 @end group | |
522 | |
523 @group | |
524 (defun bar (a &optional b &rest c) | |
525 (list a b c)) | |
526 @result{} bar | |
527 @end group | |
528 @group | |
529 (bar 1 2 3 4 5) | |
530 @result{} (1 2 (3 4 5)) | |
531 @end group | |
532 @group | |
533 (bar 1) | |
534 @result{} (1 nil nil) | |
535 @end group | |
536 @group | |
537 (bar) | |
538 @error{} Wrong number of arguments. | |
539 @end group | |
540 | |
541 @group | |
542 (defun capitalize-backwards () | |
543 "Upcase the last letter of a word." | |
544 (interactive) | |
545 (backward-word 1) | |
546 (forward-word 1) | |
547 (backward-char 1) | |
548 (capitalize-word 1)) | |
549 @result{} capitalize-backwards | |
550 @end group | |
551 @end example | |
552 | |
553 Be careful not to redefine existing functions unintentionally. | |
554 @code{defun} redefines even primitive functions such as @code{car} | |
555 without any hesitation or notification. Redefining a function already | |
556 defined is often done deliberately, and there is no way to distinguish | |
557 deliberate redefinition from unintentional redefinition. | |
558 @end defspec | |
559 | |
560 @defun defalias name definition | |
561 This special form defines the symbol @var{name} as a function, with | |
7193 | 562 definition @var{definition} (which can be any valid Lisp function). |
12098 | 563 |
564 The proper place to use @code{defalias} is where a specific function | |
565 name is being defined---especially where that name appears explicitly in | |
566 the source file being loaded. This is because @code{defalias} records | |
567 which file defined the function, just like @code{defun} | |
568 (@pxref{Unloading}). | |
569 | |
570 By contrast, in programs that manipulate function definitions for other | |
571 purposes, it is better to use @code{fset}, which does not keep such | |
572 records. | |
6455 | 573 @end defun |
574 | |
12098 | 575 See also @code{defsubst}, which defines a function like @code{defun} |
576 and tells the Lisp compiler to open-code it. @xref{Inline Functions}. | |
577 | |
6455 | 578 @node Calling Functions |
579 @section Calling Functions | |
580 @cindex function invocation | |
581 @cindex calling a function | |
582 | |
583 Defining functions is only half the battle. Functions don't do | |
584 anything until you @dfn{call} them, i.e., tell them to run. Calling a | |
585 function is also known as @dfn{invocation}. | |
586 | |
7193 | 587 The most common way of invoking a function is by evaluating a list. |
588 For example, evaluating the list @code{(concat "a" "b")} calls the | |
589 function @code{concat} with arguments @code{"a"} and @code{"b"}. | |
590 @xref{Evaluation}, for a description of evaluation. | |
6455 | 591 |
592 When you write a list as an expression in your program, the function | |
21007
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
593 name it calls is written in your program. This means that you choose |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
594 which function to call, and how many arguments to give it, when you |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
595 write the program. Usually that's just what you want. Occasionally you |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
596 need to compute at run time which function to call. To do that, use the |
21682
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
597 function @code{funcall}. When you also need to determine at run time |
22138
d4ac295a98b3
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21682
diff
changeset
|
598 how many arguments to pass, use @code{apply}. |
6455 | 599 |
600 @defun funcall function &rest arguments | |
601 @code{funcall} calls @var{function} with @var{arguments}, and returns | |
602 whatever @var{function} returns. | |
603 | |
604 Since @code{funcall} is a function, all of its arguments, including | |
605 @var{function}, are evaluated before @code{funcall} is called. This | |
606 means that you can use any expression to obtain the function to be | |
607 called. It also means that @code{funcall} does not see the expressions | |
608 you write for the @var{arguments}, only their values. These values are | |
609 @emph{not} evaluated a second time in the act of calling @var{function}; | |
610 @code{funcall} enters the normal procedure for calling a function at the | |
611 place where the arguments have already been evaluated. | |
612 | |
613 The argument @var{function} must be either a Lisp function or a | |
614 primitive function. Special forms and macros are not allowed, because | |
615 they make sense only when given the ``unevaluated'' argument | |
616 expressions. @code{funcall} cannot provide these because, as we saw | |
617 above, it never knows them in the first place. | |
618 | |
619 @example | |
620 @group | |
621 (setq f 'list) | |
622 @result{} list | |
623 @end group | |
624 @group | |
625 (funcall f 'x 'y 'z) | |
626 @result{} (x y z) | |
627 @end group | |
628 @group | |
629 (funcall f 'x 'y '(z)) | |
630 @result{} (x y (z)) | |
631 @end group | |
632 @group | |
633 (funcall 'and t nil) | |
634 @error{} Invalid function: #<subr and> | |
635 @end group | |
636 @end example | |
637 | |
26192 | 638 Compare these examples with the examples of @code{apply}. |
6455 | 639 @end defun |
640 | |
641 @defun apply function &rest arguments | |
642 @code{apply} calls @var{function} with @var{arguments}, just like | |
643 @code{funcall} but with one difference: the last of @var{arguments} is a | |
21007
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
644 list of objects, which are passed to @var{function} as separate |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
645 arguments, rather than a single list. We say that @code{apply} |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
646 @dfn{spreads} this list so that each individual element becomes an |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
647 argument. |
6455 | 648 |
649 @code{apply} returns the result of calling @var{function}. As with | |
650 @code{funcall}, @var{function} must either be a Lisp function or a | |
651 primitive function; special forms and macros do not make sense in | |
652 @code{apply}. | |
653 | |
654 @example | |
655 @group | |
656 (setq f 'list) | |
657 @result{} list | |
658 @end group | |
659 @group | |
660 (apply f 'x 'y 'z) | |
661 @error{} Wrong type argument: listp, z | |
662 @end group | |
663 @group | |
664 (apply '+ 1 2 '(3 4)) | |
665 @result{} 10 | |
666 @end group | |
667 @group | |
668 (apply '+ '(1 2 3 4)) | |
669 @result{} 10 | |
670 @end group | |
671 | |
672 @group | |
673 (apply 'append '((a b c) nil (x y z) nil)) | |
674 @result{} (a b c x y z) | |
675 @end group | |
676 @end example | |
677 | |
678 For an interesting example of using @code{apply}, see the description of | |
679 @code{mapcar}, in @ref{Mapping Functions}. | |
680 @end defun | |
681 | |
682 @cindex functionals | |
683 It is common for Lisp functions to accept functions as arguments or | |
684 find them in data structures (especially in hook variables and property | |
685 lists) and call them using @code{funcall} or @code{apply}. Functions | |
686 that accept function arguments are often called @dfn{functionals}. | |
687 | |
12098 | 688 Sometimes, when you call a functional, it is useful to supply a no-op |
689 function as the argument. Here are two different kinds of no-op | |
6455 | 690 function: |
691 | |
692 @defun identity arg | |
693 This function returns @var{arg} and has no side effects. | |
694 @end defun | |
695 | |
696 @defun ignore &rest args | |
697 This function ignores any arguments and returns @code{nil}. | |
698 @end defun | |
699 | |
700 @node Mapping Functions | |
701 @section Mapping Functions | |
702 @cindex mapping functions | |
703 | |
704 A @dfn{mapping function} applies a given function to each element of a | |
21007
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
705 list or other collection. Emacs Lisp has several such functions; |
6455 | 706 @code{mapcar} and @code{mapconcat}, which scan a list, are described |
21007
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
707 here. @xref{Creating Symbols}, for the function @code{mapatoms} which |
26192 | 708 maps over the symbols in an obarray. @xref{Hash Access}, for the |
709 function @code{maphash} which maps over key/value associations in a | |
710 hash table. | |
21682
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
711 |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
712 These mapping functions do not allow char-tables because a char-table |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
713 is a sparse array whose nominal range of indices is very large. To map |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
714 over a char-table in a way that deals properly with its sparse nature, |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
715 use the function @code{map-char-table} (@pxref{Char-Tables}). |
6455 | 716 |
717 @defun mapcar function sequence | |
7193 | 718 @code{mapcar} applies @var{function} to each element of @var{sequence} |
719 in turn, and returns a list of the results. | |
6455 | 720 |
21682
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
721 The argument @var{sequence} can be any kind of sequence except a |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
722 char-table; that is, a list, a vector, a bool-vector, or a string. The |
6455 | 723 result is always a list. The length of the result is the same as the |
724 length of @var{sequence}. | |
725 | |
726 @smallexample | |
727 @group | |
728 @exdent @r{For example:} | |
729 | |
730 (mapcar 'car '((a b) (c d) (e f))) | |
731 @result{} (a c e) | |
732 (mapcar '1+ [1 2 3]) | |
733 @result{} (2 3 4) | |
734 (mapcar 'char-to-string "abc") | |
735 @result{} ("a" "b" "c") | |
736 @end group | |
737 | |
738 @group | |
739 ;; @r{Call each function in @code{my-hooks}.} | |
740 (mapcar 'funcall my-hooks) | |
741 @end group | |
742 | |
743 @group | |
21682
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
744 (defun mapcar* (function &rest args) |
6455 | 745 "Apply FUNCTION to successive cars of all ARGS. |
746 Return the list of results." | |
747 ;; @r{If no list is exhausted,} | |
748 (if (not (memq 'nil args)) | |
21682
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
749 ;; @r{apply function to @sc{car}s.} |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
750 (cons (apply function (mapcar 'car args)) |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
751 (apply 'mapcar* function |
6455 | 752 ;; @r{Recurse for rest of elements.} |
753 (mapcar 'cdr args))))) | |
754 @end group | |
755 | |
756 @group | |
757 (mapcar* 'cons '(a b c) '(1 2 3 4)) | |
758 @result{} ((a . 1) (b . 2) (c . 3)) | |
759 @end group | |
760 @end smallexample | |
761 @end defun | |
762 | |
28556 | 763 @defun mapc function sequence |
764 @tindex mapc | |
765 @code{mapc} is like @code{mapcar} except that @var{function} is used for | |
766 side-effects only---the values it returns are ignored, not collected | |
767 into a list. @code{mapc} always returns @var{sequence}. | |
768 @end defun | |
769 | |
6455 | 770 @defun mapconcat function sequence separator |
771 @code{mapconcat} applies @var{function} to each element of | |
772 @var{sequence}: the results, which must be strings, are concatenated. | |
773 Between each pair of result strings, @code{mapconcat} inserts the string | |
774 @var{separator}. Usually @var{separator} contains a space or comma or | |
775 other suitable punctuation. | |
776 | |
777 The argument @var{function} must be a function that can take one | |
21682
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
778 argument and return a string. The argument @var{sequence} can be any |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
779 kind of sequence except a char-table; that is, a list, a vector, a |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
780 bool-vector, or a string. |
6455 | 781 |
782 @smallexample | |
783 @group | |
784 (mapconcat 'symbol-name | |
785 '(The cat in the hat) | |
786 " ") | |
787 @result{} "The cat in the hat" | |
788 @end group | |
789 | |
790 @group | |
791 (mapconcat (function (lambda (x) (format "%c" (1+ x)))) | |
792 "HAL-8000" | |
793 "") | |
794 @result{} "IBM.9111" | |
795 @end group | |
796 @end smallexample | |
797 @end defun | |
798 | |
799 @node Anonymous Functions | |
800 @section Anonymous Functions | |
801 @cindex anonymous function | |
802 | |
803 In Lisp, a function is a list that starts with @code{lambda}, a | |
804 byte-code function compiled from such a list, or alternatively a | |
805 primitive subr-object; names are ``extra''. Although usually functions | |
806 are defined with @code{defun} and given names at the same time, it is | |
807 occasionally more concise to use an explicit lambda expression---an | |
808 anonymous function. Such a list is valid wherever a function name is. | |
809 | |
810 Any method of creating such a list makes a valid function. Even this: | |
811 | |
812 @smallexample | |
813 @group | |
34179
82306439b33d
Don't quote lambdas.
Stefan Monnier <monnier@iro.umontreal.ca>
parents:
33999
diff
changeset
|
814 (setq silly (append (lambda (x)) (list (list '+ (* 3 4) 'x)))) |
6455 | 815 @result{} (lambda (x) (+ 12 x)) |
816 @end group | |
817 @end smallexample | |
818 | |
819 @noindent | |
820 This computes a list that looks like @code{(lambda (x) (+ 12 x))} and | |
821 makes it the value (@emph{not} the function definition!) of | |
822 @code{silly}. | |
823 | |
824 Here is how we might call this function: | |
825 | |
826 @example | |
827 @group | |
828 (funcall silly 1) | |
829 @result{} 13 | |
830 @end group | |
831 @end example | |
832 | |
833 @noindent | |
834 (It does @emph{not} work to write @code{(silly 1)}, because this function | |
835 is not the @emph{function definition} of @code{silly}. We have not given | |
836 @code{silly} any function definition, just a value as a variable.) | |
837 | |
838 Most of the time, anonymous functions are constants that appear in | |
21007
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
839 your program. For example, you might want to pass one as an argument to |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
840 the function @code{mapcar}, which applies any given function to each |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
841 element of a list. |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
842 |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
843 Here we define a function @code{change-property} which |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
844 uses a function as its third argument: |
6455 | 845 |
846 @example | |
847 @group | |
21007
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
848 (defun change-property (symbol prop function) |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
849 (let ((value (get symbol prop))) |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
850 (put symbol prop (funcall function value)))) |
6455 | 851 @end group |
21007
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
852 @end example |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
853 |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
854 @noindent |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
855 Here we define a function that uses @code{change-property}, |
21682
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
856 passing it a function to double a number: |
21007
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
857 |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
858 @example |
6455 | 859 @group |
21007
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
860 (defun double-property (symbol prop) |
34179
82306439b33d
Don't quote lambdas.
Stefan Monnier <monnier@iro.umontreal.ca>
parents:
33999
diff
changeset
|
861 (change-property symbol prop (lambda (x) (* 2 x)))) |
6455 | 862 @end group |
863 @end example | |
864 | |
865 @noindent | |
866 In such cases, we usually use the special form @code{function} instead | |
21007
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
867 of simple quotation to quote the anonymous function, like this: |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
868 |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
869 @example |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
870 @group |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
871 (defun double-property (symbol prop) |
22138
d4ac295a98b3
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21682
diff
changeset
|
872 (change-property symbol prop |
d4ac295a98b3
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21682
diff
changeset
|
873 (function (lambda (x) (* 2 x))))) |
21007
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
874 @end group |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
875 @end example |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
876 |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
877 Using @code{function} instead of @code{quote} makes a difference if you |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
878 compile the function @code{double-property}. For example, if you |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
879 compile the second definition of @code{double-property}, the anonymous |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
880 function is compiled as well. By contrast, if you compile the first |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
881 definition which uses ordinary @code{quote}, the argument passed to |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
882 @code{change-property} is the precise list shown: |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
883 |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
884 @example |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
885 (lambda (x) (* x 2)) |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
886 @end example |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
887 |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
888 @noindent |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
889 The Lisp compiler cannot assume this list is a function, even though it |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
890 looks like one, since it does not know what @code{change-property} will |
22138
d4ac295a98b3
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21682
diff
changeset
|
891 do with the list. Perhaps it will check whether the @sc{car} of the third |
21007
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
892 element is the symbol @code{*}! Using @code{function} tells the |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
893 compiler it is safe to go ahead and compile the constant function. |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
894 |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
895 We sometimes write @code{function} instead of @code{quote} when |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
896 quoting the name of a function, but this usage is just a sort of |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
897 comment: |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
898 |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
899 @example |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
900 (function @var{symbol}) @equiv{} (quote @var{symbol}) @equiv{} '@var{symbol} |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
901 @end example |
6455 | 902 |
25751
467b88fab665
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
22138
diff
changeset
|
903 @cindex @samp{#'} syntax |
22138
d4ac295a98b3
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21682
diff
changeset
|
904 The read syntax @code{#'} is a short-hand for using @code{function}. |
d4ac295a98b3
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21682
diff
changeset
|
905 For example, |
d4ac295a98b3
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21682
diff
changeset
|
906 |
d4ac295a98b3
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21682
diff
changeset
|
907 @example |
d4ac295a98b3
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21682
diff
changeset
|
908 #'(lambda (x) (* x x)) |
d4ac295a98b3
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21682
diff
changeset
|
909 @end example |
d4ac295a98b3
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21682
diff
changeset
|
910 |
d4ac295a98b3
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21682
diff
changeset
|
911 @noindent |
d4ac295a98b3
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21682
diff
changeset
|
912 is equivalent to |
d4ac295a98b3
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21682
diff
changeset
|
913 |
d4ac295a98b3
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21682
diff
changeset
|
914 @example |
d4ac295a98b3
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21682
diff
changeset
|
915 (function (lambda (x) (* x x))) |
d4ac295a98b3
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21682
diff
changeset
|
916 @end example |
d4ac295a98b3
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21682
diff
changeset
|
917 |
6455 | 918 @defspec function function-object |
919 @cindex function quoting | |
920 This special form returns @var{function-object} without evaluating it. | |
921 In this, it is equivalent to @code{quote}. However, it serves as a | |
922 note to the Emacs Lisp compiler that @var{function-object} is intended | |
923 to be used only as a function, and therefore can safely be compiled. | |
924 Contrast this with @code{quote}, in @ref{Quoting}. | |
925 @end defspec | |
926 | |
927 See @code{documentation} in @ref{Accessing Documentation}, for a | |
928 realistic example using @code{function} and an anonymous function. | |
929 | |
930 @node Function Cells | |
931 @section Accessing Function Cell Contents | |
932 | |
933 The @dfn{function definition} of a symbol is the object stored in the | |
934 function cell of the symbol. The functions described here access, test, | |
935 and set the function cell of symbols. | |
936 | |
7193 | 937 See also the function @code{indirect-function} in @ref{Function |
938 Indirection}. | |
939 | |
6455 | 940 @defun symbol-function symbol |
941 @kindex void-function | |
942 This returns the object in the function cell of @var{symbol}. If the | |
943 symbol's function cell is void, a @code{void-function} error is | |
944 signaled. | |
945 | |
946 This function does not check that the returned object is a legitimate | |
947 function. | |
948 | |
949 @example | |
950 @group | |
951 (defun bar (n) (+ n 2)) | |
952 @result{} bar | |
953 @end group | |
954 @group | |
955 (symbol-function 'bar) | |
956 @result{} (lambda (n) (+ n 2)) | |
957 @end group | |
958 @group | |
959 (fset 'baz 'bar) | |
960 @result{} bar | |
961 @end group | |
962 @group | |
963 (symbol-function 'baz) | |
964 @result{} bar | |
965 @end group | |
966 @end example | |
967 @end defun | |
968 | |
969 @cindex void function cell | |
970 If you have never given a symbol any function definition, we say that | |
971 that symbol's function cell is @dfn{void}. In other words, the function | |
972 cell does not have any Lisp object in it. If you try to call such a symbol | |
973 as a function, it signals a @code{void-function} error. | |
974 | |
975 Note that void is not the same as @code{nil} or the symbol | |
976 @code{void}. The symbols @code{nil} and @code{void} are Lisp objects, | |
977 and can be stored into a function cell just as any other object can be | |
978 (and they can be valid functions if you define them in turn with | |
7193 | 979 @code{defun}). A void function cell contains no object whatsoever. |
6455 | 980 |
981 You can test the voidness of a symbol's function definition with | |
982 @code{fboundp}. After you have given a symbol a function definition, you | |
983 can make it void once more using @code{fmakunbound}. | |
984 | |
985 @defun fboundp symbol | |
986 This function returns @code{t} if the symbol has an object in its | |
987 function cell, @code{nil} otherwise. It does not check that the object | |
988 is a legitimate function. | |
989 @end defun | |
990 | |
991 @defun fmakunbound symbol | |
992 This function makes @var{symbol}'s function cell void, so that a | |
993 subsequent attempt to access this cell will cause a @code{void-function} | |
22138
d4ac295a98b3
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21682
diff
changeset
|
994 error. (See also @code{makunbound}, in @ref{Void Variables}.) |
6455 | 995 |
996 @example | |
997 @group | |
998 (defun foo (x) x) | |
21007
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
999 @result{} foo |
6455 | 1000 @end group |
1001 @group | |
7193 | 1002 (foo 1) |
1003 @result{}1 | |
1004 @end group | |
1005 @group | |
6455 | 1006 (fmakunbound 'foo) |
21007
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
1007 @result{} foo |
6455 | 1008 @end group |
1009 @group | |
1010 (foo 1) | |
1011 @error{} Symbol's function definition is void: foo | |
1012 @end group | |
1013 @end example | |
1014 @end defun | |
1015 | |
16755
e3abbf649bee
Change arg name for fset.
Richard M. Stallman <rms@gnu.org>
parents:
12098
diff
changeset
|
1016 @defun fset symbol definition |
21007
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
1017 This function stores @var{definition} in the function cell of |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
1018 @var{symbol}. The result is @var{definition}. Normally |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
1019 @var{definition} should be a function or the name of a function, but |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
1020 this is not checked. The argument @var{symbol} is an ordinary evaluated |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16755
diff
changeset
|
1021 argument. |
6455 | 1022 |
1023 There are three normal uses of this function: | |
1024 | |
1025 @itemize @bullet | |
1026 @item | |
21682
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
1027 Copying one symbol's function definition to another---in other words, |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
1028 making an alternate name for a function. (If you think of this as the |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
1029 definition of the new name, you should use @code{defalias} instead of |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
1030 @code{fset}; see @ref{Defining Functions}.) |
6455 | 1031 |
1032 @item | |
1033 Giving a symbol a function definition that is not a list and therefore | |
7193 | 1034 cannot be made with @code{defun}. For example, you can use @code{fset} |
1035 to give a symbol @code{s1} a function definition which is another symbol | |
1036 @code{s2}; then @code{s1} serves as an alias for whatever definition | |
21682
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
1037 @code{s2} presently has. (Once again use @code{defalias} instead of |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
1038 @code{fset} if you think of this as the definition of @code{s1}.) |
6455 | 1039 |
1040 @item | |
1041 In constructs for defining or altering functions. If @code{defun} | |
1042 were not a primitive, it could be written in Lisp (as a macro) using | |
1043 @code{fset}. | |
1044 @end itemize | |
1045 | |
21682
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
1046 Here are examples of these uses: |
6455 | 1047 |
1048 @example | |
1049 @group | |
21682
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
1050 ;; @r{Save @code{foo}'s definition in @code{old-foo}.} |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
1051 (fset 'old-foo (symbol-function 'foo)) |
6455 | 1052 @end group |
1053 | |
1054 @group | |
1055 ;; @r{Make the symbol @code{car} the function definition of @code{xfirst}.} | |
21682
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
1056 ;; @r{(Most likely, @code{defalias} would be better than @code{fset} here.)} |
6455 | 1057 (fset 'xfirst 'car) |
1058 @result{} car | |
1059 @end group | |
1060 @group | |
1061 (xfirst '(1 2 3)) | |
1062 @result{} 1 | |
1063 @end group | |
1064 @group | |
1065 (symbol-function 'xfirst) | |
1066 @result{} car | |
1067 @end group | |
1068 @group | |
1069 (symbol-function (symbol-function 'xfirst)) | |
1070 @result{} #<subr car> | |
1071 @end group | |
1072 | |
1073 @group | |
1074 ;; @r{Define a named keyboard macro.} | |
1075 (fset 'kill-two-lines "\^u2\^k") | |
1076 @result{} "\^u2\^k" | |
1077 @end group | |
21682
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
1078 |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
1079 @group |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
1080 ;; @r{Here is a function that alters other functions.} |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
1081 (defun copy-function-definition (new old) |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
1082 "Define NEW with the same function definition as OLD." |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
1083 (fset new (symbol-function old))) |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
1084 @end group |
6455 | 1085 @end example |
1086 @end defun | |
1087 | |
1088 When writing a function that extends a previously defined function, | |
12098 | 1089 the following idiom is sometimes used: |
6455 | 1090 |
1091 @example | |
1092 (fset 'old-foo (symbol-function 'foo)) | |
1093 (defun foo () | |
1094 "Just like old-foo, except more so." | |
1095 @group | |
1096 (old-foo) | |
1097 (more-so)) | |
1098 @end group | |
1099 @end example | |
1100 | |
1101 @noindent | |
1102 This does not work properly if @code{foo} has been defined to autoload. | |
1103 In such a case, when @code{foo} calls @code{old-foo}, Lisp attempts | |
1104 to define @code{old-foo} by loading a file. Since this presumably | |
1105 defines @code{foo} rather than @code{old-foo}, it does not produce the | |
1106 proper results. The only way to avoid this problem is to make sure the | |
1107 file is loaded before moving aside the old definition of @code{foo}. | |
1108 | |
12098 | 1109 But it is unmodular and unclean, in any case, for a Lisp file to |
21682
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
1110 redefine a function defined elsewhere. It is cleaner to use the advice |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
1111 facility (@pxref{Advising Functions}). |
12098 | 1112 |
6455 | 1113 @node Inline Functions |
1114 @section Inline Functions | |
1115 @cindex inline functions | |
1116 | |
1117 @findex defsubst | |
1118 You can define an @dfn{inline function} by using @code{defsubst} instead | |
1119 of @code{defun}. An inline function works just like an ordinary | |
1120 function except for one thing: when you compile a call to the function, | |
1121 the function's definition is open-coded into the caller. | |
1122 | |
1123 Making a function inline makes explicit calls run faster. But it also | |
1124 has disadvantages. For one thing, it reduces flexibility; if you change | |
1125 the definition of the function, calls already inlined still use the old | |
1126 definition until you recompile them. Since the flexibility of | |
1127 redefining functions is an important feature of Emacs, you should not | |
1128 make a function inline unless its speed is really crucial. | |
1129 | |
1130 Another disadvantage is that making a large function inline can increase | |
1131 the size of compiled code both in files and in memory. Since the speed | |
1132 advantage of inline functions is greatest for small functions, you | |
1133 generally should not make large functions inline. | |
1134 | |
1135 It's possible to define a macro to expand into the same code that an | |
21682
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
1136 inline function would execute. (@xref{Macros}.) But the macro would be |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
1137 limited to direct use in expressions---a macro cannot be called with |
6455 | 1138 @code{apply}, @code{mapcar} and so on. Also, it takes some work to |
21682
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
1139 convert an ordinary function into a macro. To convert it into an inline |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
1140 function is very easy; simply replace @code{defun} with @code{defsubst}. |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
1141 Since each argument of an inline function is evaluated exactly once, you |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
1142 needn't worry about how many times the body uses the arguments, as you |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
1143 do for macros. (@xref{Argument Evaluation}.) |
6455 | 1144 |
7193 | 1145 Inline functions can be used and open-coded later on in the same file, |
6455 | 1146 following the definition, just like macros. |
1147 | |
12098 | 1148 @c Emacs versions prior to 19 did not have inline functions. |
6455 | 1149 |
1150 @node Related Topics | |
1151 @section Other Topics Related to Functions | |
1152 | |
1153 Here is a table of several functions that do things related to | |
1154 function calling and function definitions. They are documented | |
1155 elsewhere, but we provide cross references here. | |
1156 | |
1157 @table @code | |
1158 @item apply | |
1159 See @ref{Calling Functions}. | |
1160 | |
1161 @item autoload | |
1162 See @ref{Autoload}. | |
1163 | |
1164 @item call-interactively | |
1165 See @ref{Interactive Call}. | |
1166 | |
1167 @item commandp | |
1168 See @ref{Interactive Call}. | |
1169 | |
1170 @item documentation | |
1171 See @ref{Accessing Documentation}. | |
1172 | |
1173 @item eval | |
1174 See @ref{Eval}. | |
1175 | |
1176 @item funcall | |
1177 See @ref{Calling Functions}. | |
1178 | |
21682
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
1179 @item function |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
1180 See @ref{Anonymous Functions}. |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
1181 |
6455 | 1182 @item ignore |
1183 See @ref{Calling Functions}. | |
1184 | |
1185 @item indirect-function | |
1186 See @ref{Function Indirection}. | |
1187 | |
1188 @item interactive | |
1189 See @ref{Using Interactive}. | |
1190 | |
1191 @item interactive-p | |
1192 See @ref{Interactive Call}. | |
1193 | |
1194 @item mapatoms | |
1195 See @ref{Creating Symbols}. | |
1196 | |
1197 @item mapcar | |
1198 See @ref{Mapping Functions}. | |
1199 | |
21682
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
1200 @item map-char-table |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
1201 See @ref{Char-Tables}. |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
1202 |
6455 | 1203 @item mapconcat |
1204 See @ref{Mapping Functions}. | |
1205 | |
1206 @item undefined | |
1207 See @ref{Key Lookup}. | |
1208 @end table | |
1209 |