Mercurial > emacs
annotate doc/lispref/functions.texi @ 84743:21b080d14de6
(lisp, shortlisp): Remove termdev.elc.
author | Glenn Morris <rgm@gnu.org> |
---|---|
date | Fri, 21 Sep 2007 07:09:48 +0000 |
parents | 0ba80d073e27 |
children | b210bba3f477 |
rev | line source |
---|---|
84070 | 1 @c -*-texinfo-*- |
2 @c This is part of the GNU Emacs Lisp Reference Manual. | |
3 @c Copyright (C) 1990, 1991, 1992, 1993, 1994, 1995, 1998, 1999, 2001, | |
4 @c 2002, 2003, 2004, 2005, 2006, 2007 Free Software Foundation, Inc. | |
5 @c See the file elisp.texi for copying conditions. | |
84116
0ba80d073e27
(setfilename): Go up one more level to ../../info.
Glenn Morris <rgm@gnu.org>
parents:
84070
diff
changeset
|
6 @setfilename ../../info/functions |
84070 | 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 * Obsolete Functions:: Declaring functions obsolete. | |
25 * Inline Functions:: Defining functions that the compiler will open code. | |
26 * Function Safety:: Determining whether a function is safe to call. | |
27 * Related Topics:: Cross-references to specific Lisp primitives | |
28 that have a special bearing on how functions work. | |
29 @end menu | |
30 | |
31 @node What Is a Function | |
32 @section What Is a Function? | |
33 | |
34 In a general sense, a function is a rule for carrying on a computation | |
35 given several values called @dfn{arguments}. The result of the | |
36 computation is called the value of the function. The computation can | |
37 also have side effects: lasting changes in the values of variables or | |
38 the contents of data structures. | |
39 | |
40 Here are important terms for functions in Emacs Lisp and for other | |
41 function-like objects. | |
42 | |
43 @table @dfn | |
44 @item function | |
45 @cindex function | |
46 In Emacs Lisp, a @dfn{function} is anything that can be applied to | |
47 arguments in a Lisp program. In some cases, we use it more | |
48 specifically to mean a function written in Lisp. Special forms and | |
49 macros are not functions. | |
50 | |
51 @item primitive | |
52 @cindex primitive | |
53 @cindex subr | |
54 @cindex built-in function | |
55 A @dfn{primitive} is a function callable from Lisp that is written in C, | |
56 such as @code{car} or @code{append}. These functions are also called | |
57 @dfn{built-in functions}, or @dfn{subrs}. (Special forms are also | |
58 considered primitives.) | |
59 | |
60 Usually the reason we implement a function as a primitive is either | |
61 because it is fundamental, because it provides a low-level interface | |
62 to operating system services, or because it needs to run fast. | |
63 Primitives can be modified or added only by changing the C sources and | |
64 recompiling the editor. See @ref{Writing Emacs Primitives}. | |
65 | |
66 @item lambda expression | |
67 A @dfn{lambda expression} is a function written in Lisp. | |
68 These are described in the following section. | |
69 @ifnottex | |
70 @xref{Lambda Expressions}. | |
71 @end ifnottex | |
72 | |
73 @item special form | |
74 A @dfn{special form} is a primitive that is like a function but does not | |
75 evaluate all of its arguments in the usual way. It may evaluate only | |
76 some of the arguments, or may evaluate them in an unusual order, or | |
77 several times. Many special forms are described in @ref{Control | |
78 Structures}. | |
79 | |
80 @item macro | |
81 @cindex macro | |
82 A @dfn{macro} is a construct defined in Lisp by the programmer. It | |
83 differs from a function in that it translates a Lisp expression that you | |
84 write into an equivalent expression to be evaluated instead of the | |
85 original expression. Macros enable Lisp programmers to do the sorts of | |
86 things that special forms can do. @xref{Macros}, for how to define and | |
87 use macros. | |
88 | |
89 @item command | |
90 @cindex command | |
91 A @dfn{command} is an object that @code{command-execute} can invoke; it | |
92 is a possible definition for a key sequence. Some functions are | |
93 commands; a function written in Lisp is a command if it contains an | |
94 interactive declaration (@pxref{Defining Commands}). Such a function | |
95 can be called from Lisp expressions like other functions; in this case, | |
96 the fact that the function is a command makes no difference. | |
97 | |
98 Keyboard macros (strings and vectors) are commands also, even though | |
99 they are not functions. A symbol is a command if its function | |
100 definition is a command; such symbols can be invoked with @kbd{M-x}. | |
101 The symbol is a function as well if the definition is a function. | |
102 @xref{Interactive Call}. | |
103 | |
104 @item keystroke command | |
105 @cindex keystroke command | |
106 A @dfn{keystroke command} is a command that is bound to a key sequence | |
107 (typically one to three keystrokes). The distinction is made here | |
108 merely to avoid confusion with the meaning of ``command'' in non-Emacs | |
109 editors; for Lisp programs, the distinction is normally unimportant. | |
110 | |
111 @item byte-code function | |
112 A @dfn{byte-code function} is a function that has been compiled by the | |
113 byte compiler. @xref{Byte-Code Type}. | |
114 @end table | |
115 | |
116 @defun functionp object | |
117 This function returns @code{t} if @var{object} is any kind of | |
118 function, or a special form, or, recursively, a symbol whose function | |
119 definition is a function or special form. (This does not include | |
120 macros.) | |
121 @end defun | |
122 | |
123 Unlike @code{functionp}, the next three functions do @emph{not} | |
124 treat a symbol as its function definition. | |
125 | |
126 @defun subrp object | |
127 This function returns @code{t} if @var{object} is a built-in function | |
128 (i.e., a Lisp primitive). | |
129 | |
130 @example | |
131 @group | |
132 (subrp 'message) ; @r{@code{message} is a symbol,} | |
133 @result{} nil ; @r{not a subr object.} | |
134 @end group | |
135 @group | |
136 (subrp (symbol-function 'message)) | |
137 @result{} t | |
138 @end group | |
139 @end example | |
140 @end defun | |
141 | |
142 @defun byte-code-function-p object | |
143 This function returns @code{t} if @var{object} is a byte-code | |
144 function. For example: | |
145 | |
146 @example | |
147 @group | |
148 (byte-code-function-p (symbol-function 'next-line)) | |
149 @result{} t | |
150 @end group | |
151 @end example | |
152 @end defun | |
153 | |
154 @defun subr-arity subr | |
155 This function provides information about the argument list of a | |
156 primitive, @var{subr}. The returned value is a pair | |
157 @code{(@var{min} . @var{max})}. @var{min} is the minimum number of | |
158 args. @var{max} is the maximum number or the symbol @code{many}, for a | |
159 function with @code{&rest} arguments, or the symbol @code{unevalled} if | |
160 @var{subr} is a special form. | |
161 @end defun | |
162 | |
163 @node Lambda Expressions | |
164 @section Lambda Expressions | |
165 @cindex lambda expression | |
166 | |
167 A function written in Lisp is a list that looks like this: | |
168 | |
169 @example | |
170 (lambda (@var{arg-variables}@dots{}) | |
171 @r{[}@var{documentation-string}@r{]} | |
172 @r{[}@var{interactive-declaration}@r{]} | |
173 @var{body-forms}@dots{}) | |
174 @end example | |
175 | |
176 @noindent | |
177 Such a list is called a @dfn{lambda expression}. In Emacs Lisp, it | |
178 actually is valid as an expression---it evaluates to itself. In some | |
179 other Lisp dialects, a lambda expression is not a valid expression at | |
180 all. In either case, its main use is not to be evaluated as an | |
181 expression, but to be called as a function. | |
182 | |
183 @menu | |
184 * Lambda Components:: The parts of a lambda expression. | |
185 * Simple Lambda:: A simple example. | |
186 * Argument List:: Details and special features of argument lists. | |
187 * Function Documentation:: How to put documentation in a function. | |
188 @end menu | |
189 | |
190 @node Lambda Components | |
191 @subsection Components of a Lambda Expression | |
192 | |
193 @ifnottex | |
194 | |
195 A function written in Lisp (a ``lambda expression'') is a list that | |
196 looks like this: | |
197 | |
198 @example | |
199 (lambda (@var{arg-variables}@dots{}) | |
200 [@var{documentation-string}] | |
201 [@var{interactive-declaration}] | |
202 @var{body-forms}@dots{}) | |
203 @end example | |
204 @end ifnottex | |
205 | |
206 @cindex lambda list | |
207 The first element of a lambda expression is always the symbol | |
208 @code{lambda}. This indicates that the list represents a function. The | |
209 reason functions are defined to start with @code{lambda} is so that | |
210 other lists, intended for other uses, will not accidentally be valid as | |
211 functions. | |
212 | |
213 The second element is a list of symbols---the argument variable names. | |
214 This is called the @dfn{lambda list}. When a Lisp function is called, | |
215 the argument values are matched up against the variables in the lambda | |
216 list, which are given local bindings with the values provided. | |
217 @xref{Local Variables}. | |
218 | |
219 The documentation string is a Lisp string object placed within the | |
220 function definition to describe the function for the Emacs help | |
221 facilities. @xref{Function Documentation}. | |
222 | |
223 The interactive declaration is a list of the form @code{(interactive | |
224 @var{code-string})}. This declares how to provide arguments if the | |
225 function is used interactively. Functions with this declaration are called | |
226 @dfn{commands}; they can be called using @kbd{M-x} or bound to a key. | |
227 Functions not intended to be called in this way should not have interactive | |
228 declarations. @xref{Defining Commands}, for how to write an interactive | |
229 declaration. | |
230 | |
231 @cindex body of function | |
232 The rest of the elements are the @dfn{body} of the function: the Lisp | |
233 code to do the work of the function (or, as a Lisp programmer would say, | |
234 ``a list of Lisp forms to evaluate''). The value returned by the | |
235 function is the value returned by the last element of the body. | |
236 | |
237 @node Simple Lambda | |
238 @subsection A Simple Lambda-Expression Example | |
239 | |
240 Consider for example the following function: | |
241 | |
242 @example | |
243 (lambda (a b c) (+ a b c)) | |
244 @end example | |
245 | |
246 @noindent | |
247 We can call this function by writing it as the @sc{car} of an | |
248 expression, like this: | |
249 | |
250 @example | |
251 @group | |
252 ((lambda (a b c) (+ a b c)) | |
253 1 2 3) | |
254 @end group | |
255 @end example | |
256 | |
257 @noindent | |
258 This call evaluates the body of the lambda expression with the variable | |
259 @code{a} bound to 1, @code{b} bound to 2, and @code{c} bound to 3. | |
260 Evaluation of the body adds these three numbers, producing the result 6; | |
261 therefore, this call to the function returns the value 6. | |
262 | |
263 Note that the arguments can be the results of other function calls, as in | |
264 this example: | |
265 | |
266 @example | |
267 @group | |
268 ((lambda (a b c) (+ a b c)) | |
269 1 (* 2 3) (- 5 4)) | |
270 @end group | |
271 @end example | |
272 | |
273 @noindent | |
274 This evaluates the arguments @code{1}, @code{(* 2 3)}, and @code{(- 5 | |
275 4)} from left to right. Then it applies the lambda expression to the | |
276 argument values 1, 6 and 1 to produce the value 8. | |
277 | |
278 It is not often useful to write a lambda expression as the @sc{car} of | |
279 a form in this way. You can get the same result, of making local | |
280 variables and giving them values, using the special form @code{let} | |
281 (@pxref{Local Variables}). And @code{let} is clearer and easier to use. | |
282 In practice, lambda expressions are either stored as the function | |
283 definitions of symbols, to produce named functions, or passed as | |
284 arguments to other functions (@pxref{Anonymous Functions}). | |
285 | |
286 However, calls to explicit lambda expressions were very useful in the | |
287 old days of Lisp, before the special form @code{let} was invented. At | |
288 that time, they were the only way to bind and initialize local | |
289 variables. | |
290 | |
291 @node Argument List | |
292 @subsection Other Features of Argument Lists | |
293 @kindex wrong-number-of-arguments | |
294 @cindex argument binding | |
295 @cindex binding arguments | |
296 @cindex argument lists, features | |
297 | |
298 Our simple sample function, @code{(lambda (a b c) (+ a b c))}, | |
299 specifies three argument variables, so it must be called with three | |
300 arguments: if you try to call it with only two arguments or four | |
301 arguments, you get a @code{wrong-number-of-arguments} error. | |
302 | |
303 It is often convenient to write a function that allows certain | |
304 arguments to be omitted. For example, the function @code{substring} | |
305 accepts three arguments---a string, the start index and the end | |
306 index---but the third argument defaults to the @var{length} of the | |
307 string if you omit it. It is also convenient for certain functions to | |
308 accept an indefinite number of arguments, as the functions @code{list} | |
309 and @code{+} do. | |
310 | |
311 @cindex optional arguments | |
312 @cindex rest arguments | |
313 @kindex &optional | |
314 @kindex &rest | |
315 To specify optional arguments that may be omitted when a function | |
316 is called, simply include the keyword @code{&optional} before the optional | |
317 arguments. To specify a list of zero or more extra arguments, include the | |
318 keyword @code{&rest} before one final argument. | |
319 | |
320 Thus, the complete syntax for an argument list is as follows: | |
321 | |
322 @example | |
323 @group | |
324 (@var{required-vars}@dots{} | |
325 @r{[}&optional @var{optional-vars}@dots{}@r{]} | |
326 @r{[}&rest @var{rest-var}@r{]}) | |
327 @end group | |
328 @end example | |
329 | |
330 @noindent | |
331 The square brackets indicate that the @code{&optional} and @code{&rest} | |
332 clauses, and the variables that follow them, are optional. | |
333 | |
334 A call to the function requires one actual argument for each of the | |
335 @var{required-vars}. There may be actual arguments for zero or more of | |
336 the @var{optional-vars}, and there cannot be any actual arguments beyond | |
337 that unless the lambda list uses @code{&rest}. In that case, there may | |
338 be any number of extra actual arguments. | |
339 | |
340 If actual arguments for the optional and rest variables are omitted, | |
341 then they always default to @code{nil}. There is no way for the | |
342 function to distinguish between an explicit argument of @code{nil} and | |
343 an omitted argument. However, the body of the function is free to | |
344 consider @code{nil} an abbreviation for some other meaningful value. | |
345 This is what @code{substring} does; @code{nil} as the third argument to | |
346 @code{substring} means to use the length of the string supplied. | |
347 | |
348 @cindex CL note---default optional arg | |
349 @quotation | |
350 @b{Common Lisp note:} Common Lisp allows the function to specify what | |
351 default value to use when an optional argument is omitted; Emacs Lisp | |
352 always uses @code{nil}. Emacs Lisp does not support ``supplied-p'' | |
353 variables that tell you whether an argument was explicitly passed. | |
354 @end quotation | |
355 | |
356 For example, an argument list that looks like this: | |
357 | |
358 @example | |
359 (a b &optional c d &rest e) | |
360 @end example | |
361 | |
362 @noindent | |
363 binds @code{a} and @code{b} to the first two actual arguments, which are | |
364 required. If one or two more arguments are provided, @code{c} and | |
365 @code{d} are bound to them respectively; any arguments after the first | |
366 four are collected into a list and @code{e} is bound to that list. If | |
367 there are only two arguments, @code{c} is @code{nil}; if two or three | |
368 arguments, @code{d} is @code{nil}; if four arguments or fewer, @code{e} | |
369 is @code{nil}. | |
370 | |
371 There is no way to have required arguments following optional | |
372 ones---it would not make sense. To see why this must be so, suppose | |
373 that @code{c} in the example were optional and @code{d} were required. | |
374 Suppose three actual arguments are given; which variable would the | |
375 third argument be for? Would it be used for the @var{c}, or for | |
376 @var{d}? One can argue for both possibilities. Similarly, it makes | |
377 no sense to have any more arguments (either required or optional) | |
378 after a @code{&rest} argument. | |
379 | |
380 Here are some examples of argument lists and proper calls: | |
381 | |
382 @smallexample | |
383 ((lambda (n) (1+ n)) ; @r{One required:} | |
384 1) ; @r{requires exactly one argument.} | |
385 @result{} 2 | |
386 ((lambda (n &optional n1) ; @r{One required and one optional:} | |
387 (if n1 (+ n n1) (1+ n))) ; @r{1 or 2 arguments.} | |
388 1 2) | |
389 @result{} 3 | |
390 ((lambda (n &rest ns) ; @r{One required and one rest:} | |
391 (+ n (apply '+ ns))) ; @r{1 or more arguments.} | |
392 1 2 3 4 5) | |
393 @result{} 15 | |
394 @end smallexample | |
395 | |
396 @node Function Documentation | |
397 @subsection Documentation Strings of Functions | |
398 @cindex documentation of function | |
399 | |
400 A lambda expression may optionally have a @dfn{documentation string} just | |
401 after the lambda list. This string does not affect execution of the | |
402 function; it is a kind of comment, but a systematized comment which | |
403 actually appears inside the Lisp world and can be used by the Emacs help | |
404 facilities. @xref{Documentation}, for how the @var{documentation-string} is | |
405 accessed. | |
406 | |
407 It is a good idea to provide documentation strings for all the | |
408 functions in your program, even those that are called only from within | |
409 your program. Documentation strings are like comments, except that they | |
410 are easier to access. | |
411 | |
412 The first line of the documentation string should stand on its own, | |
413 because @code{apropos} displays just this first line. It should consist | |
414 of one or two complete sentences that summarize the function's purpose. | |
415 | |
416 The start of the documentation string is usually indented in the | |
417 source file, but since these spaces come before the starting | |
418 double-quote, they are not part of the string. Some people make a | |
419 practice of indenting any additional lines of the string so that the | |
420 text lines up in the program source. @emph{That is a mistake.} The | |
421 indentation of the following lines is inside the string; what looks | |
422 nice in the source code will look ugly when displayed by the help | |
423 commands. | |
424 | |
425 You may wonder how the documentation string could be optional, since | |
426 there are required components of the function that follow it (the body). | |
427 Since evaluation of a string returns that string, without any side effects, | |
428 it has no effect if it is not the last form in the body. Thus, in | |
429 practice, there is no confusion between the first form of the body and the | |
430 documentation string; if the only body form is a string then it serves both | |
431 as the return value and as the documentation. | |
432 | |
433 The last line of the documentation string can specify calling | |
434 conventions different from the actual function arguments. Write | |
435 text like this: | |
436 | |
437 @example | |
438 \(fn @var{arglist}) | |
439 @end example | |
440 | |
441 @noindent | |
442 following a blank line, at the beginning of the line, with no newline | |
443 following it inside the documentation string. (The @samp{\} is used | |
444 to avoid confusing the Emacs motion commands.) The calling convention | |
445 specified in this way appears in help messages in place of the one | |
446 derived from the actual arguments of the function. | |
447 | |
448 This feature is particularly useful for macro definitions, since the | |
449 arguments written in a macro definition often do not correspond to the | |
450 way users think of the parts of the macro call. | |
451 | |
452 @node Function Names | |
453 @section Naming a Function | |
454 @cindex function definition | |
455 @cindex named function | |
456 @cindex function name | |
457 | |
458 In most computer languages, every function has a name; the idea of a | |
459 function without a name is nonsensical. In Lisp, a function in the | |
460 strictest sense has no name. It is simply a list whose first element is | |
461 @code{lambda}, a byte-code function object, or a primitive subr-object. | |
462 | |
463 However, a symbol can serve as the name of a function. This happens | |
464 when you put the function in the symbol's @dfn{function cell} | |
465 (@pxref{Symbol Components}). Then the symbol itself becomes a valid, | |
466 callable function, equivalent to the list or subr-object that its | |
467 function cell refers to. The contents of the function cell are also | |
468 called the symbol's @dfn{function definition}. The procedure of using a | |
469 symbol's function definition in place of the symbol is called | |
470 @dfn{symbol function indirection}; see @ref{Function Indirection}. | |
471 | |
472 In practice, nearly all functions are given names in this way and | |
473 referred to through their names. For example, the symbol @code{car} works | |
474 as a function and does what it does because the primitive subr-object | |
475 @code{#<subr car>} is stored in its function cell. | |
476 | |
477 We give functions names because it is convenient to refer to them by | |
478 their names in Lisp expressions. For primitive subr-objects such as | |
479 @code{#<subr car>}, names are the only way you can refer to them: there | |
480 is no read syntax for such objects. For functions written in Lisp, the | |
481 name is more convenient to use in a call than an explicit lambda | |
482 expression. Also, a function with a name can refer to itself---it can | |
483 be recursive. Writing the function's name in its own definition is much | |
484 more convenient than making the function definition point to itself | |
485 (something that is not impossible but that has various disadvantages in | |
486 practice). | |
487 | |
488 We often identify functions with the symbols used to name them. For | |
489 example, we often speak of ``the function @code{car},'' not | |
490 distinguishing between the symbol @code{car} and the primitive | |
491 subr-object that is its function definition. For most purposes, the | |
492 distinction is not important. | |
493 | |
494 Even so, keep in mind that a function need not have a unique name. While | |
495 a given function object @emph{usually} appears in the function cell of only | |
496 one symbol, this is just a matter of convenience. It is easy to store | |
497 it in several symbols using @code{fset}; then each of the symbols is | |
498 equally well a name for the same function. | |
499 | |
500 A symbol used as a function name may also be used as a variable; these | |
501 two uses of a symbol are independent and do not conflict. (Some Lisp | |
502 dialects, such as Scheme, do not distinguish between a symbol's value | |
503 and its function definition; a symbol's value as a variable is also its | |
504 function definition.) If you have not given a symbol a function | |
505 definition, you cannot use it as a function; whether the symbol has a | |
506 value as a variable makes no difference to this. | |
507 | |
508 @node Defining Functions | |
509 @section Defining Functions | |
510 @cindex defining a function | |
511 | |
512 We usually give a name to a function when it is first created. This | |
513 is called @dfn{defining a function}, and it is done with the | |
514 @code{defun} special form. | |
515 | |
516 @defspec defun name argument-list body-forms | |
517 @code{defun} is the usual way to define new Lisp functions. It | |
518 defines the symbol @var{name} as a function that looks like this: | |
519 | |
520 @example | |
521 (lambda @var{argument-list} . @var{body-forms}) | |
522 @end example | |
523 | |
524 @code{defun} stores this lambda expression in the function cell of | |
525 @var{name}. It returns the value @var{name}, but usually we ignore this | |
526 value. | |
527 | |
528 As described previously, @var{argument-list} is a list of argument | |
529 names and may include the keywords @code{&optional} and @code{&rest} | |
530 (@pxref{Lambda Expressions}). Also, the first two of the | |
531 @var{body-forms} may be a documentation string and an interactive | |
532 declaration. | |
533 | |
534 There is no conflict if the same symbol @var{name} is also used as a | |
535 variable, since the symbol's value cell is independent of the function | |
536 cell. @xref{Symbol Components}. | |
537 | |
538 Here are some examples: | |
539 | |
540 @example | |
541 @group | |
542 (defun foo () 5) | |
543 @result{} foo | |
544 @end group | |
545 @group | |
546 (foo) | |
547 @result{} 5 | |
548 @end group | |
549 | |
550 @group | |
551 (defun bar (a &optional b &rest c) | |
552 (list a b c)) | |
553 @result{} bar | |
554 @end group | |
555 @group | |
556 (bar 1 2 3 4 5) | |
557 @result{} (1 2 (3 4 5)) | |
558 @end group | |
559 @group | |
560 (bar 1) | |
561 @result{} (1 nil nil) | |
562 @end group | |
563 @group | |
564 (bar) | |
565 @error{} Wrong number of arguments. | |
566 @end group | |
567 | |
568 @group | |
569 (defun capitalize-backwards () | |
570 "Upcase the last letter of a word." | |
571 (interactive) | |
572 (backward-word 1) | |
573 (forward-word 1) | |
574 (backward-char 1) | |
575 (capitalize-word 1)) | |
576 @result{} capitalize-backwards | |
577 @end group | |
578 @end example | |
579 | |
580 Be careful not to redefine existing functions unintentionally. | |
581 @code{defun} redefines even primitive functions such as @code{car} | |
582 without any hesitation or notification. Redefining a function already | |
583 defined is often done deliberately, and there is no way to distinguish | |
584 deliberate redefinition from unintentional redefinition. | |
585 @end defspec | |
586 | |
587 @cindex function aliases | |
588 @defun defalias name definition &optional docstring | |
589 @anchor{Definition of defalias} | |
590 This special form defines the symbol @var{name} as a function, with | |
591 definition @var{definition} (which can be any valid Lisp function). | |
592 It returns @var{definition}. | |
593 | |
594 If @var{docstring} is non-@code{nil}, it becomes the function | |
595 documentation of @var{name}. Otherwise, any documentation provided by | |
596 @var{definition} is used. | |
597 | |
598 The proper place to use @code{defalias} is where a specific function | |
599 name is being defined---especially where that name appears explicitly in | |
600 the source file being loaded. This is because @code{defalias} records | |
601 which file defined the function, just like @code{defun} | |
602 (@pxref{Unloading}). | |
603 | |
604 By contrast, in programs that manipulate function definitions for other | |
605 purposes, it is better to use @code{fset}, which does not keep such | |
606 records. @xref{Function Cells}. | |
607 @end defun | |
608 | |
609 You cannot create a new primitive function with @code{defun} or | |
610 @code{defalias}, but you can use them to change the function definition of | |
611 any symbol, even one such as @code{car} or @code{x-popup-menu} whose | |
612 normal definition is a primitive. However, this is risky: for | |
613 instance, it is next to impossible to redefine @code{car} without | |
614 breaking Lisp completely. Redefining an obscure function such as | |
615 @code{x-popup-menu} is less dangerous, but it still may not work as | |
616 you expect. If there are calls to the primitive from C code, they | |
617 call the primitive's C definition directly, so changing the symbol's | |
618 definition will have no effect on them. | |
619 | |
620 See also @code{defsubst}, which defines a function like @code{defun} | |
621 and tells the Lisp compiler to open-code it. @xref{Inline Functions}. | |
622 | |
623 @node Calling Functions | |
624 @section Calling Functions | |
625 @cindex function invocation | |
626 @cindex calling a function | |
627 | |
628 Defining functions is only half the battle. Functions don't do | |
629 anything until you @dfn{call} them, i.e., tell them to run. Calling a | |
630 function is also known as @dfn{invocation}. | |
631 | |
632 The most common way of invoking a function is by evaluating a list. | |
633 For example, evaluating the list @code{(concat "a" "b")} calls the | |
634 function @code{concat} with arguments @code{"a"} and @code{"b"}. | |
635 @xref{Evaluation}, for a description of evaluation. | |
636 | |
637 When you write a list as an expression in your program, you specify | |
638 which function to call, and how many arguments to give it, in the text | |
639 of the program. Usually that's just what you want. Occasionally you | |
640 need to compute at run time which function to call. To do that, use | |
641 the function @code{funcall}. When you also need to determine at run | |
642 time how many arguments to pass, use @code{apply}. | |
643 | |
644 @defun funcall function &rest arguments | |
645 @code{funcall} calls @var{function} with @var{arguments}, and returns | |
646 whatever @var{function} returns. | |
647 | |
648 Since @code{funcall} is a function, all of its arguments, including | |
649 @var{function}, are evaluated before @code{funcall} is called. This | |
650 means that you can use any expression to obtain the function to be | |
651 called. It also means that @code{funcall} does not see the | |
652 expressions you write for the @var{arguments}, only their values. | |
653 These values are @emph{not} evaluated a second time in the act of | |
654 calling @var{function}; the operation of @code{funcall} is like the | |
655 normal procedure for calling a function, once its arguments have | |
656 already been evaluated. | |
657 | |
658 The argument @var{function} must be either a Lisp function or a | |
659 primitive function. Special forms and macros are not allowed, because | |
660 they make sense only when given the ``unevaluated'' argument | |
661 expressions. @code{funcall} cannot provide these because, as we saw | |
662 above, it never knows them in the first place. | |
663 | |
664 @example | |
665 @group | |
666 (setq f 'list) | |
667 @result{} list | |
668 @end group | |
669 @group | |
670 (funcall f 'x 'y 'z) | |
671 @result{} (x y z) | |
672 @end group | |
673 @group | |
674 (funcall f 'x 'y '(z)) | |
675 @result{} (x y (z)) | |
676 @end group | |
677 @group | |
678 (funcall 'and t nil) | |
679 @error{} Invalid function: #<subr and> | |
680 @end group | |
681 @end example | |
682 | |
683 Compare these examples with the examples of @code{apply}. | |
684 @end defun | |
685 | |
686 @defun apply function &rest arguments | |
687 @code{apply} calls @var{function} with @var{arguments}, just like | |
688 @code{funcall} but with one difference: the last of @var{arguments} is a | |
689 list of objects, which are passed to @var{function} as separate | |
690 arguments, rather than a single list. We say that @code{apply} | |
691 @dfn{spreads} this list so that each individual element becomes an | |
692 argument. | |
693 | |
694 @code{apply} returns the result of calling @var{function}. As with | |
695 @code{funcall}, @var{function} must either be a Lisp function or a | |
696 primitive function; special forms and macros do not make sense in | |
697 @code{apply}. | |
698 | |
699 @example | |
700 @group | |
701 (setq f 'list) | |
702 @result{} list | |
703 @end group | |
704 @group | |
705 (apply f 'x 'y 'z) | |
706 @error{} Wrong type argument: listp, z | |
707 @end group | |
708 @group | |
709 (apply '+ 1 2 '(3 4)) | |
710 @result{} 10 | |
711 @end group | |
712 @group | |
713 (apply '+ '(1 2 3 4)) | |
714 @result{} 10 | |
715 @end group | |
716 | |
717 @group | |
718 (apply 'append '((a b c) nil (x y z) nil)) | |
719 @result{} (a b c x y z) | |
720 @end group | |
721 @end example | |
722 | |
723 For an interesting example of using @code{apply}, see @ref{Definition | |
724 of mapcar}. | |
725 @end defun | |
726 | |
727 @cindex functionals | |
728 It is common for Lisp functions to accept functions as arguments or | |
729 find them in data structures (especially in hook variables and property | |
730 lists) and call them using @code{funcall} or @code{apply}. Functions | |
731 that accept function arguments are often called @dfn{functionals}. | |
732 | |
733 Sometimes, when you call a functional, it is useful to supply a no-op | |
734 function as the argument. Here are two different kinds of no-op | |
735 function: | |
736 | |
737 @defun identity arg | |
738 This function returns @var{arg} and has no side effects. | |
739 @end defun | |
740 | |
741 @defun ignore &rest args | |
742 This function ignores any arguments and returns @code{nil}. | |
743 @end defun | |
744 | |
745 @node Mapping Functions | |
746 @section Mapping Functions | |
747 @cindex mapping functions | |
748 | |
749 A @dfn{mapping function} applies a given function (@emph{not} a | |
750 special form or macro) to each element of a list or other collection. | |
751 Emacs Lisp has several such functions; @code{mapcar} and | |
752 @code{mapconcat}, which scan a list, are described here. | |
753 @xref{Definition of mapatoms}, for the function @code{mapatoms} which | |
754 maps over the symbols in an obarray. @xref{Definition of maphash}, | |
755 for the function @code{maphash} which maps over key/value associations | |
756 in a hash table. | |
757 | |
758 These mapping functions do not allow char-tables because a char-table | |
759 is a sparse array whose nominal range of indices is very large. To map | |
760 over a char-table in a way that deals properly with its sparse nature, | |
761 use the function @code{map-char-table} (@pxref{Char-Tables}). | |
762 | |
763 @defun mapcar function sequence | |
764 @anchor{Definition of mapcar} | |
765 @code{mapcar} applies @var{function} to each element of @var{sequence} | |
766 in turn, and returns a list of the results. | |
767 | |
768 The argument @var{sequence} can be any kind of sequence except a | |
769 char-table; that is, a list, a vector, a bool-vector, or a string. The | |
770 result is always a list. The length of the result is the same as the | |
771 length of @var{sequence}. For example: | |
772 | |
773 @smallexample | |
774 @group | |
775 (mapcar 'car '((a b) (c d) (e f))) | |
776 @result{} (a c e) | |
777 (mapcar '1+ [1 2 3]) | |
778 @result{} (2 3 4) | |
779 (mapcar 'char-to-string "abc") | |
780 @result{} ("a" "b" "c") | |
781 @end group | |
782 | |
783 @group | |
784 ;; @r{Call each function in @code{my-hooks}.} | |
785 (mapcar 'funcall my-hooks) | |
786 @end group | |
787 | |
788 @group | |
789 (defun mapcar* (function &rest args) | |
790 "Apply FUNCTION to successive cars of all ARGS. | |
791 Return the list of results." | |
792 ;; @r{If no list is exhausted,} | |
793 (if (not (memq nil args)) | |
794 ;; @r{apply function to @sc{car}s.} | |
795 (cons (apply function (mapcar 'car args)) | |
796 (apply 'mapcar* function | |
797 ;; @r{Recurse for rest of elements.} | |
798 (mapcar 'cdr args))))) | |
799 @end group | |
800 | |
801 @group | |
802 (mapcar* 'cons '(a b c) '(1 2 3 4)) | |
803 @result{} ((a . 1) (b . 2) (c . 3)) | |
804 @end group | |
805 @end smallexample | |
806 @end defun | |
807 | |
808 @defun mapc function sequence | |
809 @code{mapc} is like @code{mapcar} except that @var{function} is used for | |
810 side-effects only---the values it returns are ignored, not collected | |
811 into a list. @code{mapc} always returns @var{sequence}. | |
812 @end defun | |
813 | |
814 @defun mapconcat function sequence separator | |
815 @code{mapconcat} applies @var{function} to each element of | |
816 @var{sequence}: the results, which must be strings, are concatenated. | |
817 Between each pair of result strings, @code{mapconcat} inserts the string | |
818 @var{separator}. Usually @var{separator} contains a space or comma or | |
819 other suitable punctuation. | |
820 | |
821 The argument @var{function} must be a function that can take one | |
822 argument and return a string. The argument @var{sequence} can be any | |
823 kind of sequence except a char-table; that is, a list, a vector, a | |
824 bool-vector, or a string. | |
825 | |
826 @smallexample | |
827 @group | |
828 (mapconcat 'symbol-name | |
829 '(The cat in the hat) | |
830 " ") | |
831 @result{} "The cat in the hat" | |
832 @end group | |
833 | |
834 @group | |
835 (mapconcat (function (lambda (x) (format "%c" (1+ x)))) | |
836 "HAL-8000" | |
837 "") | |
838 @result{} "IBM.9111" | |
839 @end group | |
840 @end smallexample | |
841 @end defun | |
842 | |
843 @node Anonymous Functions | |
844 @section Anonymous Functions | |
845 @cindex anonymous function | |
846 | |
847 In Lisp, a function is a list that starts with @code{lambda}, a | |
848 byte-code function compiled from such a list, or alternatively a | |
849 primitive subr-object; names are ``extra.'' Although usually functions | |
850 are defined with @code{defun} and given names at the same time, it is | |
851 occasionally more concise to use an explicit lambda expression---an | |
852 anonymous function. Such a list is valid wherever a function name is. | |
853 | |
854 Any method of creating such a list makes a valid function. Even this: | |
855 | |
856 @smallexample | |
857 @group | |
858 (setq silly (append '(lambda (x)) (list (list '+ (* 3 4) 'x)))) | |
859 @result{} (lambda (x) (+ 12 x)) | |
860 @end group | |
861 @end smallexample | |
862 | |
863 @noindent | |
864 This computes a list that looks like @code{(lambda (x) (+ 12 x))} and | |
865 makes it the value (@emph{not} the function definition!) of | |
866 @code{silly}. | |
867 | |
868 Here is how we might call this function: | |
869 | |
870 @example | |
871 @group | |
872 (funcall silly 1) | |
873 @result{} 13 | |
874 @end group | |
875 @end example | |
876 | |
877 @noindent | |
878 (It does @emph{not} work to write @code{(silly 1)}, because this function | |
879 is not the @emph{function definition} of @code{silly}. We have not given | |
880 @code{silly} any function definition, just a value as a variable.) | |
881 | |
882 Most of the time, anonymous functions are constants that appear in | |
883 your program. For example, you might want to pass one as an argument to | |
884 the function @code{mapcar}, which applies any given function to each | |
885 element of a list. | |
886 | |
887 Here we define a function @code{change-property} which | |
888 uses a function as its third argument: | |
889 | |
890 @example | |
891 @group | |
892 (defun change-property (symbol prop function) | |
893 (let ((value (get symbol prop))) | |
894 (put symbol prop (funcall function value)))) | |
895 @end group | |
896 @end example | |
897 | |
898 @noindent | |
899 Here we define a function that uses @code{change-property}, | |
900 passing it a function to double a number: | |
901 | |
902 @example | |
903 @group | |
904 (defun double-property (symbol prop) | |
905 (change-property symbol prop '(lambda (x) (* 2 x)))) | |
906 @end group | |
907 @end example | |
908 | |
909 @noindent | |
910 In such cases, we usually use the special form @code{function} instead | |
911 of simple quotation to quote the anonymous function, like this: | |
912 | |
913 @example | |
914 @group | |
915 (defun double-property (symbol prop) | |
916 (change-property symbol prop | |
917 (function (lambda (x) (* 2 x))))) | |
918 @end group | |
919 @end example | |
920 | |
921 Using @code{function} instead of @code{quote} makes a difference if you | |
922 compile the function @code{double-property}. For example, if you | |
923 compile the second definition of @code{double-property}, the anonymous | |
924 function is compiled as well. By contrast, if you compile the first | |
925 definition which uses ordinary @code{quote}, the argument passed to | |
926 @code{change-property} is the precise list shown: | |
927 | |
928 @example | |
929 (lambda (x) (* x 2)) | |
930 @end example | |
931 | |
932 @noindent | |
933 The Lisp compiler cannot assume this list is a function, even though it | |
934 looks like one, since it does not know what @code{change-property} will | |
935 do with the list. Perhaps it will check whether the @sc{car} of the third | |
936 element is the symbol @code{*}! Using @code{function} tells the | |
937 compiler it is safe to go ahead and compile the constant function. | |
938 | |
939 Nowadays it is possible to omit @code{function} entirely, like this: | |
940 | |
941 @example | |
942 @group | |
943 (defun double-property (symbol prop) | |
944 (change-property symbol prop (lambda (x) (* 2 x)))) | |
945 @end group | |
946 @end example | |
947 | |
948 @noindent | |
949 This is because @code{lambda} itself implies @code{function}. | |
950 | |
951 We sometimes write @code{function} instead of @code{quote} when | |
952 quoting the name of a function, but this usage is just a sort of | |
953 comment: | |
954 | |
955 @example | |
956 (function @var{symbol}) @equiv{} (quote @var{symbol}) @equiv{} '@var{symbol} | |
957 @end example | |
958 | |
959 @cindex @samp{#'} syntax | |
960 The read syntax @code{#'} is a short-hand for using @code{function}. | |
961 For example, | |
962 | |
963 @example | |
964 #'(lambda (x) (* x x)) | |
965 @end example | |
966 | |
967 @noindent | |
968 is equivalent to | |
969 | |
970 @example | |
971 (function (lambda (x) (* x x))) | |
972 @end example | |
973 | |
974 @defspec function function-object | |
975 @cindex function quoting | |
976 This special form returns @var{function-object} without evaluating it. | |
977 In this, it is equivalent to @code{quote}. However, it serves as a | |
978 note to the Emacs Lisp compiler that @var{function-object} is intended | |
979 to be used only as a function, and therefore can safely be compiled. | |
980 Contrast this with @code{quote}, in @ref{Quoting}. | |
981 @end defspec | |
982 | |
983 @xref{describe-symbols example}, for a realistic example using | |
984 @code{function} and an anonymous function. | |
985 | |
986 @node Function Cells | |
987 @section Accessing Function Cell Contents | |
988 | |
989 The @dfn{function definition} of a symbol is the object stored in the | |
990 function cell of the symbol. The functions described here access, test, | |
991 and set the function cell of symbols. | |
992 | |
993 See also the function @code{indirect-function}. @xref{Definition of | |
994 indirect-function}. | |
995 | |
996 @defun symbol-function symbol | |
997 @kindex void-function | |
998 This returns the object in the function cell of @var{symbol}. If the | |
999 symbol's function cell is void, a @code{void-function} error is | |
1000 signaled. | |
1001 | |
1002 This function does not check that the returned object is a legitimate | |
1003 function. | |
1004 | |
1005 @example | |
1006 @group | |
1007 (defun bar (n) (+ n 2)) | |
1008 @result{} bar | |
1009 @end group | |
1010 @group | |
1011 (symbol-function 'bar) | |
1012 @result{} (lambda (n) (+ n 2)) | |
1013 @end group | |
1014 @group | |
1015 (fset 'baz 'bar) | |
1016 @result{} bar | |
1017 @end group | |
1018 @group | |
1019 (symbol-function 'baz) | |
1020 @result{} bar | |
1021 @end group | |
1022 @end example | |
1023 @end defun | |
1024 | |
1025 @cindex void function cell | |
1026 If you have never given a symbol any function definition, we say that | |
1027 that symbol's function cell is @dfn{void}. In other words, the function | |
1028 cell does not have any Lisp object in it. If you try to call such a symbol | |
1029 as a function, it signals a @code{void-function} error. | |
1030 | |
1031 Note that void is not the same as @code{nil} or the symbol | |
1032 @code{void}. The symbols @code{nil} and @code{void} are Lisp objects, | |
1033 and can be stored into a function cell just as any other object can be | |
1034 (and they can be valid functions if you define them in turn with | |
1035 @code{defun}). A void function cell contains no object whatsoever. | |
1036 | |
1037 You can test the voidness of a symbol's function definition with | |
1038 @code{fboundp}. After you have given a symbol a function definition, you | |
1039 can make it void once more using @code{fmakunbound}. | |
1040 | |
1041 @defun fboundp symbol | |
1042 This function returns @code{t} if the symbol has an object in its | |
1043 function cell, @code{nil} otherwise. It does not check that the object | |
1044 is a legitimate function. | |
1045 @end defun | |
1046 | |
1047 @defun fmakunbound symbol | |
1048 This function makes @var{symbol}'s function cell void, so that a | |
1049 subsequent attempt to access this cell will cause a | |
1050 @code{void-function} error. It returns @var{symbol}. (See also | |
1051 @code{makunbound}, in @ref{Void Variables}.) | |
1052 | |
1053 @example | |
1054 @group | |
1055 (defun foo (x) x) | |
1056 @result{} foo | |
1057 @end group | |
1058 @group | |
1059 (foo 1) | |
1060 @result{}1 | |
1061 @end group | |
1062 @group | |
1063 (fmakunbound 'foo) | |
1064 @result{} foo | |
1065 @end group | |
1066 @group | |
1067 (foo 1) | |
1068 @error{} Symbol's function definition is void: foo | |
1069 @end group | |
1070 @end example | |
1071 @end defun | |
1072 | |
1073 @defun fset symbol definition | |
1074 This function stores @var{definition} in the function cell of | |
1075 @var{symbol}. The result is @var{definition}. Normally | |
1076 @var{definition} should be a function or the name of a function, but | |
1077 this is not checked. The argument @var{symbol} is an ordinary evaluated | |
1078 argument. | |
1079 | |
1080 There are three normal uses of this function: | |
1081 | |
1082 @itemize @bullet | |
1083 @item | |
1084 Copying one symbol's function definition to another---in other words, | |
1085 making an alternate name for a function. (If you think of this as the | |
1086 definition of the new name, you should use @code{defalias} instead of | |
1087 @code{fset}; see @ref{Definition of defalias}.) | |
1088 | |
1089 @item | |
1090 Giving a symbol a function definition that is not a list and therefore | |
1091 cannot be made with @code{defun}. For example, you can use @code{fset} | |
1092 to give a symbol @code{s1} a function definition which is another symbol | |
1093 @code{s2}; then @code{s1} serves as an alias for whatever definition | |
1094 @code{s2} presently has. (Once again use @code{defalias} instead of | |
1095 @code{fset} if you think of this as the definition of @code{s1}.) | |
1096 | |
1097 @item | |
1098 In constructs for defining or altering functions. If @code{defun} | |
1099 were not a primitive, it could be written in Lisp (as a macro) using | |
1100 @code{fset}. | |
1101 @end itemize | |
1102 | |
1103 Here are examples of these uses: | |
1104 | |
1105 @example | |
1106 @group | |
1107 ;; @r{Save @code{foo}'s definition in @code{old-foo}.} | |
1108 (fset 'old-foo (symbol-function 'foo)) | |
1109 @end group | |
1110 | |
1111 @group | |
1112 ;; @r{Make the symbol @code{car} the function definition of @code{xfirst}.} | |
1113 ;; @r{(Most likely, @code{defalias} would be better than @code{fset} here.)} | |
1114 (fset 'xfirst 'car) | |
1115 @result{} car | |
1116 @end group | |
1117 @group | |
1118 (xfirst '(1 2 3)) | |
1119 @result{} 1 | |
1120 @end group | |
1121 @group | |
1122 (symbol-function 'xfirst) | |
1123 @result{} car | |
1124 @end group | |
1125 @group | |
1126 (symbol-function (symbol-function 'xfirst)) | |
1127 @result{} #<subr car> | |
1128 @end group | |
1129 | |
1130 @group | |
1131 ;; @r{Define a named keyboard macro.} | |
1132 (fset 'kill-two-lines "\^u2\^k") | |
1133 @result{} "\^u2\^k" | |
1134 @end group | |
1135 | |
1136 @group | |
1137 ;; @r{Here is a function that alters other functions.} | |
1138 (defun copy-function-definition (new old) | |
1139 "Define NEW with the same function definition as OLD." | |
1140 (fset new (symbol-function old))) | |
1141 @end group | |
1142 @end example | |
1143 @end defun | |
1144 | |
1145 @code{fset} is sometimes used to save the old definition of a | |
1146 function before redefining it. That permits the new definition to | |
1147 invoke the old definition. But it is unmodular and unclean for a Lisp | |
1148 file to redefine a function defined elsewhere. If you want to modify | |
1149 a function defined by another package, it is cleaner to use | |
1150 @code{defadvice} (@pxref{Advising Functions}). | |
1151 | |
1152 @node Obsolete Functions | |
1153 @section Declaring Functions Obsolete | |
1154 | |
1155 You can use @code{make-obsolete} to declare a function obsolete. This | |
1156 indicates that the function may be removed at some stage in the future. | |
1157 | |
1158 @defun make-obsolete obsolete-name current-name &optional when | |
1159 This function makes the byte compiler warn that the function | |
1160 @var{obsolete-name} is obsolete. If @var{current-name} is a symbol, the | |
1161 warning message says to use @var{current-name} instead of | |
1162 @var{obsolete-name}. @var{current-name} does not need to be an alias for | |
1163 @var{obsolete-name}; it can be a different function with similar | |
1164 functionality. If @var{current-name} is a string, it is the warning | |
1165 message. | |
1166 | |
1167 If provided, @var{when} should be a string indicating when the function | |
1168 was first made obsolete---for example, a date or a release number. | |
1169 @end defun | |
1170 | |
1171 You can define a function as an alias and declare it obsolete at the | |
1172 same time using the macro @code{define-obsolete-function-alias}. | |
1173 | |
1174 @defmac define-obsolete-function-alias obsolete-name current-name &optional when docstring | |
1175 This macro marks the function @var{obsolete-name} obsolete and also | |
1176 defines it as an alias for the function @var{current-name}. It is | |
1177 equivalent to the following: | |
1178 | |
1179 @example | |
1180 (defalias @var{obsolete-name} @var{current-name} @var{docstring}) | |
1181 (make-obsolete @var{obsolete-name} @var{current-name} @var{when}) | |
1182 @end example | |
1183 @end defmac | |
1184 | |
1185 @node Inline Functions | |
1186 @section Inline Functions | |
1187 @cindex inline functions | |
1188 | |
1189 @findex defsubst | |
1190 You can define an @dfn{inline function} by using @code{defsubst} instead | |
1191 of @code{defun}. An inline function works just like an ordinary | |
1192 function except for one thing: when you compile a call to the function, | |
1193 the function's definition is open-coded into the caller. | |
1194 | |
1195 Making a function inline makes explicit calls run faster. But it also | |
1196 has disadvantages. For one thing, it reduces flexibility; if you | |
1197 change the definition of the function, calls already inlined still use | |
1198 the old definition until you recompile them. | |
1199 | |
1200 Another disadvantage is that making a large function inline can increase | |
1201 the size of compiled code both in files and in memory. Since the speed | |
1202 advantage of inline functions is greatest for small functions, you | |
1203 generally should not make large functions inline. | |
1204 | |
1205 Also, inline functions do not behave well with respect to debugging, | |
1206 tracing, and advising (@pxref{Advising Functions}). Since ease of | |
1207 debugging and the flexibility of redefining functions are important | |
1208 features of Emacs, you should not make a function inline, even if it's | |
1209 small, unless its speed is really crucial, and you've timed the code | |
1210 to verify that using @code{defun} actually has performance problems. | |
1211 | |
1212 It's possible to define a macro to expand into the same code that an | |
1213 inline function would execute. (@xref{Macros}.) But the macro would be | |
1214 limited to direct use in expressions---a macro cannot be called with | |
1215 @code{apply}, @code{mapcar} and so on. Also, it takes some work to | |
1216 convert an ordinary function into a macro. To convert it into an inline | |
1217 function is very easy; simply replace @code{defun} with @code{defsubst}. | |
1218 Since each argument of an inline function is evaluated exactly once, you | |
1219 needn't worry about how many times the body uses the arguments, as you | |
1220 do for macros. (@xref{Argument Evaluation}.) | |
1221 | |
1222 Inline functions can be used and open-coded later on in the same file, | |
1223 following the definition, just like macros. | |
1224 | |
1225 @node Function Safety | |
1226 @section Determining whether a Function is Safe to Call | |
1227 @cindex function safety | |
1228 @cindex safety of functions | |
1229 | |
1230 Some major modes such as SES call functions that are stored in user | |
1231 files. (@inforef{Top, ,ses}, for more information on SES.) User | |
1232 files sometimes have poor pedigrees---you can get a spreadsheet from | |
1233 someone you've just met, or you can get one through email from someone | |
1234 you've never met. So it is risky to call a function whose source code | |
1235 is stored in a user file until you have determined that it is safe. | |
1236 | |
1237 @defun unsafep form &optional unsafep-vars | |
1238 Returns @code{nil} if @var{form} is a @dfn{safe} Lisp expression, or | |
1239 returns a list that describes why it might be unsafe. The argument | |
1240 @var{unsafep-vars} is a list of symbols known to have temporary | |
1241 bindings at this point; it is mainly used for internal recursive | |
1242 calls. The current buffer is an implicit argument, which provides a | |
1243 list of buffer-local bindings. | |
1244 @end defun | |
1245 | |
1246 Being quick and simple, @code{unsafep} does a very light analysis and | |
1247 rejects many Lisp expressions that are actually safe. There are no | |
1248 known cases where @code{unsafep} returns @code{nil} for an unsafe | |
1249 expression. However, a ``safe'' Lisp expression can return a string | |
1250 with a @code{display} property, containing an associated Lisp | |
1251 expression to be executed after the string is inserted into a buffer. | |
1252 This associated expression can be a virus. In order to be safe, you | |
1253 must delete properties from all strings calculated by user code before | |
1254 inserting them into buffers. | |
1255 | |
1256 @ignore | |
1257 What is a safe Lisp expression? Basically, it's an expression that | |
1258 calls only built-in functions with no side effects (or only innocuous | |
1259 ones). Innocuous side effects include displaying messages and | |
1260 altering non-risky buffer-local variables (but not global variables). | |
1261 | |
1262 @table @dfn | |
1263 @item Safe expression | |
1264 @itemize | |
1265 @item | |
1266 An atom or quoted thing. | |
1267 @item | |
1268 A call to a safe function (see below), if all its arguments are | |
1269 safe expressions. | |
1270 @item | |
1271 One of the special forms @code{and}, @code{catch}, @code{cond}, | |
1272 @code{if}, @code{or}, @code{prog1}, @code{prog2}, @code{progn}, | |
1273 @code{while}, and @code{unwind-protect}], if all its arguments are | |
1274 safe. | |
1275 @item | |
1276 A form that creates temporary bindings (@code{condition-case}, | |
1277 @code{dolist}, @code{dotimes}, @code{lambda}, @code{let}, or | |
1278 @code{let*}), if all args are safe and the symbols to be bound are not | |
1279 explicitly risky (see @pxref{File Local Variables}). | |
1280 @item | |
1281 An assignment using @code{add-to-list}, @code{setq}, @code{push}, or | |
1282 @code{pop}, if all args are safe and the symbols to be assigned are | |
1283 not explicitly risky and they already have temporary or buffer-local | |
1284 bindings. | |
1285 @item | |
1286 One of [apply, mapc, mapcar, mapconcat] if the first argument is a | |
1287 safe explicit lambda and the other args are safe expressions. | |
1288 @end itemize | |
1289 | |
1290 @item Safe function | |
1291 @itemize | |
1292 @item | |
1293 A lambda containing safe expressions. | |
1294 @item | |
1295 A symbol on the list @code{safe-functions}, so the user says it's safe. | |
1296 @item | |
1297 A symbol with a non-@code{nil} @code{side-effect-free} property. | |
1298 @item | |
1299 A symbol with a non-@code{nil} @code{safe-function} property. Value t | |
1300 indicates a function that is safe but has innocuous side effects. | |
1301 Other values will someday indicate functions with classes of side | |
1302 effects that are not always safe. | |
1303 @end itemize | |
1304 | |
1305 The @code{side-effect-free} and @code{safe-function} properties are | |
1306 provided for built-in functions and for low-level functions and macros | |
1307 defined in @file{subr.el}. You can assign these properties for the | |
1308 functions you write. | |
1309 @end table | |
1310 @end ignore | |
1311 | |
1312 @node Related Topics | |
1313 @section Other Topics Related to Functions | |
1314 | |
1315 Here is a table of several functions that do things related to | |
1316 function calling and function definitions. They are documented | |
1317 elsewhere, but we provide cross references here. | |
1318 | |
1319 @table @code | |
1320 @item apply | |
1321 See @ref{Calling Functions}. | |
1322 | |
1323 @item autoload | |
1324 See @ref{Autoload}. | |
1325 | |
1326 @item call-interactively | |
1327 See @ref{Interactive Call}. | |
1328 | |
1329 @item commandp | |
1330 See @ref{Interactive Call}. | |
1331 | |
1332 @item documentation | |
1333 See @ref{Accessing Documentation}. | |
1334 | |
1335 @item eval | |
1336 See @ref{Eval}. | |
1337 | |
1338 @item funcall | |
1339 See @ref{Calling Functions}. | |
1340 | |
1341 @item function | |
1342 See @ref{Anonymous Functions}. | |
1343 | |
1344 @item ignore | |
1345 See @ref{Calling Functions}. | |
1346 | |
1347 @item indirect-function | |
1348 See @ref{Function Indirection}. | |
1349 | |
1350 @item interactive | |
1351 See @ref{Using Interactive}. | |
1352 | |
1353 @item interactive-p | |
1354 See @ref{Interactive Call}. | |
1355 | |
1356 @item mapatoms | |
1357 See @ref{Creating Symbols}. | |
1358 | |
1359 @item mapcar | |
1360 See @ref{Mapping Functions}. | |
1361 | |
1362 @item map-char-table | |
1363 See @ref{Char-Tables}. | |
1364 | |
1365 @item mapconcat | |
1366 See @ref{Mapping Functions}. | |
1367 | |
1368 @item undefined | |
1369 See @ref{Functions for Key Lookup}. | |
1370 @end table | |
1371 | |
1372 @ignore | |
1373 arch-tag: 39100cdf-8a55-4898-acba-595db619e8e2 | |
1374 @end ignore |