Mercurial > emacs
annotate doc/lispref/functions.texi @ 94788:8d78530778f5
Switch to recommended form of GPLv3 permissions notice.
author | Glenn Morris <rgm@gnu.org> |
---|---|
date | Fri, 09 May 2008 04:38:00 +0000 |
parents | 310118b32104 |
children | 48cabcb73380 |
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, | |
87649 | 4 @c 2002, 2003, 2004, 2005, 2006, 2007, 2008 Free Software Foundation, Inc. |
84070 | 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. | |
86383
382d8b0539b2
(Declaring Functions): New section.
Glenn Morris <rgm@gnu.org>
parents:
85688
diff
changeset
|
26 * Declaring Functions:: Telling the compiler that a function is defined. |
84070 | 27 * Function Safety:: Determining whether a function is safe to call. |
28 * Related Topics:: Cross-references to specific Lisp primitives | |
29 that have a special bearing on how functions work. | |
30 @end menu | |
31 | |
32 @node What Is a Function | |
33 @section What Is a Function? | |
34 | |
35 In a general sense, a function is a rule for carrying on a computation | |
36 given several values called @dfn{arguments}. The result of the | |
37 computation is called the value of the function. The computation can | |
38 also have side effects: lasting changes in the values of variables or | |
39 the contents of data structures. | |
40 | |
41 Here are important terms for functions in Emacs Lisp and for other | |
42 function-like objects. | |
43 | |
44 @table @dfn | |
45 @item function | |
46 @cindex function | |
47 In Emacs Lisp, a @dfn{function} is anything that can be applied to | |
48 arguments in a Lisp program. In some cases, we use it more | |
49 specifically to mean a function written in Lisp. Special forms and | |
50 macros are not functions. | |
51 | |
52 @item primitive | |
53 @cindex primitive | |
54 @cindex subr | |
55 @cindex built-in function | |
56 A @dfn{primitive} is a function callable from Lisp that is written in C, | |
57 such as @code{car} or @code{append}. These functions are also called | |
58 @dfn{built-in functions}, or @dfn{subrs}. (Special forms are also | |
59 considered primitives.) | |
60 | |
61 Usually the reason we implement a function as a primitive is either | |
62 because it is fundamental, because it provides a low-level interface | |
63 to operating system services, or because it needs to run fast. | |
64 Primitives can be modified or added only by changing the C sources and | |
65 recompiling the editor. See @ref{Writing Emacs Primitives}. | |
66 | |
67 @item lambda expression | |
68 A @dfn{lambda expression} is a function written in Lisp. | |
69 These are described in the following section. | |
70 @ifnottex | |
71 @xref{Lambda Expressions}. | |
72 @end ifnottex | |
73 | |
74 @item special form | |
75 A @dfn{special form} is a primitive that is like a function but does not | |
76 evaluate all of its arguments in the usual way. It may evaluate only | |
77 some of the arguments, or may evaluate them in an unusual order, or | |
78 several times. Many special forms are described in @ref{Control | |
79 Structures}. | |
80 | |
81 @item macro | |
82 @cindex macro | |
83 A @dfn{macro} is a construct defined in Lisp by the programmer. It | |
84 differs from a function in that it translates a Lisp expression that you | |
85 write into an equivalent expression to be evaluated instead of the | |
86 original expression. Macros enable Lisp programmers to do the sorts of | |
87 things that special forms can do. @xref{Macros}, for how to define and | |
88 use macros. | |
89 | |
90 @item command | |
91 @cindex command | |
92 A @dfn{command} is an object that @code{command-execute} can invoke; it | |
93 is a possible definition for a key sequence. Some functions are | |
94 commands; a function written in Lisp is a command if it contains an | |
95 interactive declaration (@pxref{Defining Commands}). Such a function | |
96 can be called from Lisp expressions like other functions; in this case, | |
97 the fact that the function is a command makes no difference. | |
98 | |
99 Keyboard macros (strings and vectors) are commands also, even though | |
100 they are not functions. A symbol is a command if its function | |
101 definition is a command; such symbols can be invoked with @kbd{M-x}. | |
102 The symbol is a function as well if the definition is a function. | |
103 @xref{Interactive Call}. | |
104 | |
105 @item keystroke command | |
106 @cindex keystroke command | |
107 A @dfn{keystroke command} is a command that is bound to a key sequence | |
108 (typically one to three keystrokes). The distinction is made here | |
109 merely to avoid confusion with the meaning of ``command'' in non-Emacs | |
110 editors; for Lisp programs, the distinction is normally unimportant. | |
111 | |
112 @item byte-code function | |
113 A @dfn{byte-code function} is a function that has been compiled by the | |
114 byte compiler. @xref{Byte-Code Type}. | |
115 @end table | |
116 | |
117 @defun functionp object | |
118 This function returns @code{t} if @var{object} is any kind of | |
93735
310118b32104
(functionp): Return nil for special forms.
Stefan Monnier <monnier@iro.umontreal.ca>
parents:
87649
diff
changeset
|
119 function, i.e. can be passed to @code{funcall}. |
84070 | 120 @end defun |
121 | |
122 Unlike @code{functionp}, the next three functions do @emph{not} | |
123 treat a symbol as its function definition. | |
124 | |
125 @defun subrp object | |
126 This function returns @code{t} if @var{object} is a built-in function | |
127 (i.e., a Lisp primitive). | |
128 | |
129 @example | |
130 @group | |
131 (subrp 'message) ; @r{@code{message} is a symbol,} | |
132 @result{} nil ; @r{not a subr object.} | |
133 @end group | |
134 @group | |
135 (subrp (symbol-function 'message)) | |
136 @result{} t | |
137 @end group | |
138 @end example | |
139 @end defun | |
140 | |
141 @defun byte-code-function-p object | |
142 This function returns @code{t} if @var{object} is a byte-code | |
143 function. For example: | |
144 | |
145 @example | |
146 @group | |
147 (byte-code-function-p (symbol-function 'next-line)) | |
148 @result{} t | |
149 @end group | |
150 @end example | |
151 @end defun | |
152 | |
153 @defun subr-arity subr | |
154 This function provides information about the argument list of a | |
155 primitive, @var{subr}. The returned value is a pair | |
156 @code{(@var{min} . @var{max})}. @var{min} is the minimum number of | |
157 args. @var{max} is the maximum number or the symbol @code{many}, for a | |
158 function with @code{&rest} arguments, or the symbol @code{unevalled} if | |
159 @var{subr} is a special form. | |
160 @end defun | |
161 | |
162 @node Lambda Expressions | |
163 @section Lambda Expressions | |
164 @cindex lambda expression | |
165 | |
166 A function written in Lisp is a list that looks like this: | |
167 | |
168 @example | |
169 (lambda (@var{arg-variables}@dots{}) | |
170 @r{[}@var{documentation-string}@r{]} | |
171 @r{[}@var{interactive-declaration}@r{]} | |
172 @var{body-forms}@dots{}) | |
173 @end example | |
174 | |
175 @noindent | |
176 Such a list is called a @dfn{lambda expression}. In Emacs Lisp, it | |
177 actually is valid as an expression---it evaluates to itself. In some | |
178 other Lisp dialects, a lambda expression is not a valid expression at | |
179 all. In either case, its main use is not to be evaluated as an | |
180 expression, but to be called as a function. | |
181 | |
182 @menu | |
183 * Lambda Components:: The parts of a lambda expression. | |
184 * Simple Lambda:: A simple example. | |
185 * Argument List:: Details and special features of argument lists. | |
186 * Function Documentation:: How to put documentation in a function. | |
187 @end menu | |
188 | |
189 @node Lambda Components | |
190 @subsection Components of a Lambda Expression | |
191 | |
192 @ifnottex | |
193 | |
194 A function written in Lisp (a ``lambda expression'') is a list that | |
195 looks like this: | |
196 | |
197 @example | |
198 (lambda (@var{arg-variables}@dots{}) | |
199 [@var{documentation-string}] | |
200 [@var{interactive-declaration}] | |
201 @var{body-forms}@dots{}) | |
202 @end example | |
203 @end ifnottex | |
204 | |
205 @cindex lambda list | |
206 The first element of a lambda expression is always the symbol | |
207 @code{lambda}. This indicates that the list represents a function. The | |
208 reason functions are defined to start with @code{lambda} is so that | |
209 other lists, intended for other uses, will not accidentally be valid as | |
210 functions. | |
211 | |
212 The second element is a list of symbols---the argument variable names. | |
213 This is called the @dfn{lambda list}. When a Lisp function is called, | |
214 the argument values are matched up against the variables in the lambda | |
215 list, which are given local bindings with the values provided. | |
216 @xref{Local Variables}. | |
217 | |
218 The documentation string is a Lisp string object placed within the | |
219 function definition to describe the function for the Emacs help | |
220 facilities. @xref{Function Documentation}. | |
221 | |
222 The interactive declaration is a list of the form @code{(interactive | |
223 @var{code-string})}. This declares how to provide arguments if the | |
224 function is used interactively. Functions with this declaration are called | |
225 @dfn{commands}; they can be called using @kbd{M-x} or bound to a key. | |
226 Functions not intended to be called in this way should not have interactive | |
227 declarations. @xref{Defining Commands}, for how to write an interactive | |
228 declaration. | |
229 | |
230 @cindex body of function | |
231 The rest of the elements are the @dfn{body} of the function: the Lisp | |
232 code to do the work of the function (or, as a Lisp programmer would say, | |
233 ``a list of Lisp forms to evaluate''). The value returned by the | |
234 function is the value returned by the last element of the body. | |
235 | |
236 @node Simple Lambda | |
237 @subsection A Simple Lambda-Expression Example | |
238 | |
239 Consider for example the following function: | |
240 | |
241 @example | |
242 (lambda (a b c) (+ a b c)) | |
243 @end example | |
244 | |
245 @noindent | |
246 We can call this function by writing it as the @sc{car} of an | |
247 expression, like this: | |
248 | |
249 @example | |
250 @group | |
251 ((lambda (a b c) (+ a b c)) | |
252 1 2 3) | |
253 @end group | |
254 @end example | |
255 | |
256 @noindent | |
257 This call evaluates the body of the lambda expression with the variable | |
258 @code{a} bound to 1, @code{b} bound to 2, and @code{c} bound to 3. | |
259 Evaluation of the body adds these three numbers, producing the result 6; | |
260 therefore, this call to the function returns the value 6. | |
261 | |
262 Note that the arguments can be the results of other function calls, as in | |
263 this example: | |
264 | |
265 @example | |
266 @group | |
267 ((lambda (a b c) (+ a b c)) | |
268 1 (* 2 3) (- 5 4)) | |
269 @end group | |
270 @end example | |
271 | |
272 @noindent | |
273 This evaluates the arguments @code{1}, @code{(* 2 3)}, and @code{(- 5 | |
274 4)} from left to right. Then it applies the lambda expression to the | |
275 argument values 1, 6 and 1 to produce the value 8. | |
276 | |
277 It is not often useful to write a lambda expression as the @sc{car} of | |
278 a form in this way. You can get the same result, of making local | |
279 variables and giving them values, using the special form @code{let} | |
280 (@pxref{Local Variables}). And @code{let} is clearer and easier to use. | |
281 In practice, lambda expressions are either stored as the function | |
282 definitions of symbols, to produce named functions, or passed as | |
283 arguments to other functions (@pxref{Anonymous Functions}). | |
284 | |
285 However, calls to explicit lambda expressions were very useful in the | |
286 old days of Lisp, before the special form @code{let} was invented. At | |
287 that time, they were the only way to bind and initialize local | |
288 variables. | |
289 | |
290 @node Argument List | |
291 @subsection Other Features of Argument Lists | |
292 @kindex wrong-number-of-arguments | |
293 @cindex argument binding | |
294 @cindex binding arguments | |
295 @cindex argument lists, features | |
296 | |
297 Our simple sample function, @code{(lambda (a b c) (+ a b c))}, | |
298 specifies three argument variables, so it must be called with three | |
299 arguments: if you try to call it with only two arguments or four | |
300 arguments, you get a @code{wrong-number-of-arguments} error. | |
301 | |
302 It is often convenient to write a function that allows certain | |
303 arguments to be omitted. For example, the function @code{substring} | |
304 accepts three arguments---a string, the start index and the end | |
305 index---but the third argument defaults to the @var{length} of the | |
306 string if you omit it. It is also convenient for certain functions to | |
307 accept an indefinite number of arguments, as the functions @code{list} | |
308 and @code{+} do. | |
309 | |
310 @cindex optional arguments | |
311 @cindex rest arguments | |
312 @kindex &optional | |
313 @kindex &rest | |
314 To specify optional arguments that may be omitted when a function | |
315 is called, simply include the keyword @code{&optional} before the optional | |
316 arguments. To specify a list of zero or more extra arguments, include the | |
317 keyword @code{&rest} before one final argument. | |
318 | |
319 Thus, the complete syntax for an argument list is as follows: | |
320 | |
321 @example | |
322 @group | |
323 (@var{required-vars}@dots{} | |
324 @r{[}&optional @var{optional-vars}@dots{}@r{]} | |
325 @r{[}&rest @var{rest-var}@r{]}) | |
326 @end group | |
327 @end example | |
328 | |
329 @noindent | |
330 The square brackets indicate that the @code{&optional} and @code{&rest} | |
331 clauses, and the variables that follow them, are optional. | |
332 | |
333 A call to the function requires one actual argument for each of the | |
334 @var{required-vars}. There may be actual arguments for zero or more of | |
335 the @var{optional-vars}, and there cannot be any actual arguments beyond | |
336 that unless the lambda list uses @code{&rest}. In that case, there may | |
337 be any number of extra actual arguments. | |
338 | |
339 If actual arguments for the optional and rest variables are omitted, | |
340 then they always default to @code{nil}. There is no way for the | |
341 function to distinguish between an explicit argument of @code{nil} and | |
342 an omitted argument. However, the body of the function is free to | |
343 consider @code{nil} an abbreviation for some other meaningful value. | |
344 This is what @code{substring} does; @code{nil} as the third argument to | |
345 @code{substring} means to use the length of the string supplied. | |
346 | |
347 @cindex CL note---default optional arg | |
348 @quotation | |
349 @b{Common Lisp note:} Common Lisp allows the function to specify what | |
350 default value to use when an optional argument is omitted; Emacs Lisp | |
351 always uses @code{nil}. Emacs Lisp does not support ``supplied-p'' | |
352 variables that tell you whether an argument was explicitly passed. | |
353 @end quotation | |
354 | |
355 For example, an argument list that looks like this: | |
356 | |
357 @example | |
358 (a b &optional c d &rest e) | |
359 @end example | |
360 | |
361 @noindent | |
362 binds @code{a} and @code{b} to the first two actual arguments, which are | |
363 required. If one or two more arguments are provided, @code{c} and | |
364 @code{d} are bound to them respectively; any arguments after the first | |
365 four are collected into a list and @code{e} is bound to that list. If | |
366 there are only two arguments, @code{c} is @code{nil}; if two or three | |
367 arguments, @code{d} is @code{nil}; if four arguments or fewer, @code{e} | |
368 is @code{nil}. | |
369 | |
370 There is no way to have required arguments following optional | |
371 ones---it would not make sense. To see why this must be so, suppose | |
372 that @code{c} in the example were optional and @code{d} were required. | |
373 Suppose three actual arguments are given; which variable would the | |
374 third argument be for? Would it be used for the @var{c}, or for | |
375 @var{d}? One can argue for both possibilities. Similarly, it makes | |
376 no sense to have any more arguments (either required or optional) | |
377 after a @code{&rest} argument. | |
378 | |
379 Here are some examples of argument lists and proper calls: | |
380 | |
381 @smallexample | |
382 ((lambda (n) (1+ n)) ; @r{One required:} | |
383 1) ; @r{requires exactly one argument.} | |
384 @result{} 2 | |
385 ((lambda (n &optional n1) ; @r{One required and one optional:} | |
386 (if n1 (+ n n1) (1+ n))) ; @r{1 or 2 arguments.} | |
387 1 2) | |
388 @result{} 3 | |
389 ((lambda (n &rest ns) ; @r{One required and one rest:} | |
390 (+ n (apply '+ ns))) ; @r{1 or more arguments.} | |
391 1 2 3 4 5) | |
392 @result{} 15 | |
393 @end smallexample | |
394 | |
395 @node Function Documentation | |
396 @subsection Documentation Strings of Functions | |
397 @cindex documentation of function | |
398 | |
399 A lambda expression may optionally have a @dfn{documentation string} just | |
400 after the lambda list. This string does not affect execution of the | |
401 function; it is a kind of comment, but a systematized comment which | |
402 actually appears inside the Lisp world and can be used by the Emacs help | |
403 facilities. @xref{Documentation}, for how the @var{documentation-string} is | |
404 accessed. | |
405 | |
406 It is a good idea to provide documentation strings for all the | |
407 functions in your program, even those that are called only from within | |
408 your program. Documentation strings are like comments, except that they | |
409 are easier to access. | |
410 | |
411 The first line of the documentation string should stand on its own, | |
412 because @code{apropos} displays just this first line. It should consist | |
413 of one or two complete sentences that summarize the function's purpose. | |
414 | |
415 The start of the documentation string is usually indented in the | |
416 source file, but since these spaces come before the starting | |
417 double-quote, they are not part of the string. Some people make a | |
418 practice of indenting any additional lines of the string so that the | |
419 text lines up in the program source. @emph{That is a mistake.} The | |
420 indentation of the following lines is inside the string; what looks | |
421 nice in the source code will look ugly when displayed by the help | |
422 commands. | |
423 | |
424 You may wonder how the documentation string could be optional, since | |
425 there are required components of the function that follow it (the body). | |
426 Since evaluation of a string returns that string, without any side effects, | |
427 it has no effect if it is not the last form in the body. Thus, in | |
428 practice, there is no confusion between the first form of the body and the | |
429 documentation string; if the only body form is a string then it serves both | |
430 as the return value and as the documentation. | |
431 | |
432 The last line of the documentation string can specify calling | |
433 conventions different from the actual function arguments. Write | |
434 text like this: | |
435 | |
436 @example | |
437 \(fn @var{arglist}) | |
438 @end example | |
439 | |
440 @noindent | |
441 following a blank line, at the beginning of the line, with no newline | |
442 following it inside the documentation string. (The @samp{\} is used | |
443 to avoid confusing the Emacs motion commands.) The calling convention | |
444 specified in this way appears in help messages in place of the one | |
445 derived from the actual arguments of the function. | |
446 | |
447 This feature is particularly useful for macro definitions, since the | |
448 arguments written in a macro definition often do not correspond to the | |
449 way users think of the parts of the macro call. | |
450 | |
451 @node Function Names | |
452 @section Naming a Function | |
453 @cindex function definition | |
454 @cindex named function | |
455 @cindex function name | |
456 | |
457 In most computer languages, every function has a name; the idea of a | |
458 function without a name is nonsensical. In Lisp, a function in the | |
459 strictest sense has no name. It is simply a list whose first element is | |
460 @code{lambda}, a byte-code function object, or a primitive subr-object. | |
461 | |
462 However, a symbol can serve as the name of a function. This happens | |
463 when you put the function in the symbol's @dfn{function cell} | |
464 (@pxref{Symbol Components}). Then the symbol itself becomes a valid, | |
465 callable function, equivalent to the list or subr-object that its | |
466 function cell refers to. The contents of the function cell are also | |
467 called the symbol's @dfn{function definition}. The procedure of using a | |
468 symbol's function definition in place of the symbol is called | |
469 @dfn{symbol function indirection}; see @ref{Function Indirection}. | |
470 | |
471 In practice, nearly all functions are given names in this way and | |
472 referred to through their names. For example, the symbol @code{car} works | |
473 as a function and does what it does because the primitive subr-object | |
474 @code{#<subr car>} is stored in its function cell. | |
475 | |
476 We give functions names because it is convenient to refer to them by | |
477 their names in Lisp expressions. For primitive subr-objects such as | |
478 @code{#<subr car>}, names are the only way you can refer to them: there | |
479 is no read syntax for such objects. For functions written in Lisp, the | |
480 name is more convenient to use in a call than an explicit lambda | |
481 expression. Also, a function with a name can refer to itself---it can | |
482 be recursive. Writing the function's name in its own definition is much | |
483 more convenient than making the function definition point to itself | |
484 (something that is not impossible but that has various disadvantages in | |
485 practice). | |
486 | |
487 We often identify functions with the symbols used to name them. For | |
488 example, we often speak of ``the function @code{car},'' not | |
489 distinguishing between the symbol @code{car} and the primitive | |
490 subr-object that is its function definition. For most purposes, the | |
491 distinction is not important. | |
492 | |
493 Even so, keep in mind that a function need not have a unique name. While | |
494 a given function object @emph{usually} appears in the function cell of only | |
495 one symbol, this is just a matter of convenience. It is easy to store | |
496 it in several symbols using @code{fset}; then each of the symbols is | |
497 equally well a name for the same function. | |
498 | |
499 A symbol used as a function name may also be used as a variable; these | |
500 two uses of a symbol are independent and do not conflict. (Some Lisp | |
501 dialects, such as Scheme, do not distinguish between a symbol's value | |
502 and its function definition; a symbol's value as a variable is also its | |
503 function definition.) If you have not given a symbol a function | |
504 definition, you cannot use it as a function; whether the symbol has a | |
505 value as a variable makes no difference to this. | |
506 | |
507 @node Defining Functions | |
508 @section Defining Functions | |
509 @cindex defining a function | |
510 | |
511 We usually give a name to a function when it is first created. This | |
512 is called @dfn{defining a function}, and it is done with the | |
513 @code{defun} special form. | |
514 | |
515 @defspec defun name argument-list body-forms | |
516 @code{defun} is the usual way to define new Lisp functions. It | |
517 defines the symbol @var{name} as a function that looks like this: | |
518 | |
519 @example | |
520 (lambda @var{argument-list} . @var{body-forms}) | |
521 @end example | |
522 | |
523 @code{defun} stores this lambda expression in the function cell of | |
524 @var{name}. It returns the value @var{name}, but usually we ignore this | |
525 value. | |
526 | |
527 As described previously, @var{argument-list} is a list of argument | |
528 names and may include the keywords @code{&optional} and @code{&rest} | |
529 (@pxref{Lambda Expressions}). Also, the first two of the | |
530 @var{body-forms} may be a documentation string and an interactive | |
531 declaration. | |
532 | |
533 There is no conflict if the same symbol @var{name} is also used as a | |
534 variable, since the symbol's value cell is independent of the function | |
535 cell. @xref{Symbol Components}. | |
536 | |
537 Here are some examples: | |
538 | |
539 @example | |
540 @group | |
541 (defun foo () 5) | |
542 @result{} foo | |
543 @end group | |
544 @group | |
545 (foo) | |
546 @result{} 5 | |
547 @end group | |
548 | |
549 @group | |
550 (defun bar (a &optional b &rest c) | |
551 (list a b c)) | |
552 @result{} bar | |
553 @end group | |
554 @group | |
555 (bar 1 2 3 4 5) | |
556 @result{} (1 2 (3 4 5)) | |
557 @end group | |
558 @group | |
559 (bar 1) | |
560 @result{} (1 nil nil) | |
561 @end group | |
562 @group | |
563 (bar) | |
564 @error{} Wrong number of arguments. | |
565 @end group | |
566 | |
567 @group | |
568 (defun capitalize-backwards () | |
569 "Upcase the last letter of a word." | |
570 (interactive) | |
571 (backward-word 1) | |
572 (forward-word 1) | |
573 (backward-char 1) | |
574 (capitalize-word 1)) | |
575 @result{} capitalize-backwards | |
576 @end group | |
577 @end example | |
578 | |
579 Be careful not to redefine existing functions unintentionally. | |
580 @code{defun} redefines even primitive functions such as @code{car} | |
581 without any hesitation or notification. Redefining a function already | |
582 defined is often done deliberately, and there is no way to distinguish | |
583 deliberate redefinition from unintentional redefinition. | |
584 @end defspec | |
585 | |
586 @cindex function aliases | |
587 @defun defalias name definition &optional docstring | |
588 @anchor{Definition of defalias} | |
589 This special form defines the symbol @var{name} as a function, with | |
590 definition @var{definition} (which can be any valid Lisp function). | |
591 It returns @var{definition}. | |
592 | |
593 If @var{docstring} is non-@code{nil}, it becomes the function | |
594 documentation of @var{name}. Otherwise, any documentation provided by | |
595 @var{definition} is used. | |
596 | |
597 The proper place to use @code{defalias} is where a specific function | |
598 name is being defined---especially where that name appears explicitly in | |
599 the source file being loaded. This is because @code{defalias} records | |
600 which file defined the function, just like @code{defun} | |
601 (@pxref{Unloading}). | |
602 | |
603 By contrast, in programs that manipulate function definitions for other | |
604 purposes, it is better to use @code{fset}, which does not keep such | |
605 records. @xref{Function Cells}. | |
606 @end defun | |
607 | |
608 You cannot create a new primitive function with @code{defun} or | |
609 @code{defalias}, but you can use them to change the function definition of | |
610 any symbol, even one such as @code{car} or @code{x-popup-menu} whose | |
611 normal definition is a primitive. However, this is risky: for | |
612 instance, it is next to impossible to redefine @code{car} without | |
613 breaking Lisp completely. Redefining an obscure function such as | |
614 @code{x-popup-menu} is less dangerous, but it still may not work as | |
615 you expect. If there are calls to the primitive from C code, they | |
616 call the primitive's C definition directly, so changing the symbol's | |
617 definition will have no effect on them. | |
618 | |
619 See also @code{defsubst}, which defines a function like @code{defun} | |
620 and tells the Lisp compiler to open-code it. @xref{Inline Functions}. | |
621 | |
622 @node Calling Functions | |
623 @section Calling Functions | |
624 @cindex function invocation | |
625 @cindex calling a function | |
626 | |
627 Defining functions is only half the battle. Functions don't do | |
628 anything until you @dfn{call} them, i.e., tell them to run. Calling a | |
629 function is also known as @dfn{invocation}. | |
630 | |
631 The most common way of invoking a function is by evaluating a list. | |
632 For example, evaluating the list @code{(concat "a" "b")} calls the | |
633 function @code{concat} with arguments @code{"a"} and @code{"b"}. | |
634 @xref{Evaluation}, for a description of evaluation. | |
635 | |
636 When you write a list as an expression in your program, you specify | |
637 which function to call, and how many arguments to give it, in the text | |
638 of the program. Usually that's just what you want. Occasionally you | |
639 need to compute at run time which function to call. To do that, use | |
640 the function @code{funcall}. When you also need to determine at run | |
641 time how many arguments to pass, use @code{apply}. | |
642 | |
643 @defun funcall function &rest arguments | |
644 @code{funcall} calls @var{function} with @var{arguments}, and returns | |
645 whatever @var{function} returns. | |
646 | |
647 Since @code{funcall} is a function, all of its arguments, including | |
648 @var{function}, are evaluated before @code{funcall} is called. This | |
649 means that you can use any expression to obtain the function to be | |
650 called. It also means that @code{funcall} does not see the | |
651 expressions you write for the @var{arguments}, only their values. | |
652 These values are @emph{not} evaluated a second time in the act of | |
653 calling @var{function}; the operation of @code{funcall} is like the | |
654 normal procedure for calling a function, once its arguments have | |
655 already been evaluated. | |
656 | |
657 The argument @var{function} must be either a Lisp function or a | |
658 primitive function. Special forms and macros are not allowed, because | |
659 they make sense only when given the ``unevaluated'' argument | |
660 expressions. @code{funcall} cannot provide these because, as we saw | |
661 above, it never knows them in the first place. | |
662 | |
663 @example | |
664 @group | |
665 (setq f 'list) | |
666 @result{} list | |
667 @end group | |
668 @group | |
669 (funcall f 'x 'y 'z) | |
670 @result{} (x y z) | |
671 @end group | |
672 @group | |
673 (funcall f 'x 'y '(z)) | |
674 @result{} (x y (z)) | |
675 @end group | |
676 @group | |
677 (funcall 'and t nil) | |
678 @error{} Invalid function: #<subr and> | |
679 @end group | |
680 @end example | |
681 | |
682 Compare these examples with the examples of @code{apply}. | |
683 @end defun | |
684 | |
685 @defun apply function &rest arguments | |
686 @code{apply} calls @var{function} with @var{arguments}, just like | |
687 @code{funcall} but with one difference: the last of @var{arguments} is a | |
688 list of objects, which are passed to @var{function} as separate | |
689 arguments, rather than a single list. We say that @code{apply} | |
690 @dfn{spreads} this list so that each individual element becomes an | |
691 argument. | |
692 | |
693 @code{apply} returns the result of calling @var{function}. As with | |
694 @code{funcall}, @var{function} must either be a Lisp function or a | |
695 primitive function; special forms and macros do not make sense in | |
696 @code{apply}. | |
697 | |
698 @example | |
699 @group | |
700 (setq f 'list) | |
701 @result{} list | |
702 @end group | |
703 @group | |
704 (apply f 'x 'y 'z) | |
705 @error{} Wrong type argument: listp, z | |
706 @end group | |
707 @group | |
708 (apply '+ 1 2 '(3 4)) | |
709 @result{} 10 | |
710 @end group | |
711 @group | |
712 (apply '+ '(1 2 3 4)) | |
713 @result{} 10 | |
714 @end group | |
715 | |
716 @group | |
717 (apply 'append '((a b c) nil (x y z) nil)) | |
718 @result{} (a b c x y z) | |
719 @end group | |
720 @end example | |
721 | |
722 For an interesting example of using @code{apply}, see @ref{Definition | |
723 of mapcar}. | |
724 @end defun | |
725 | |
726 @cindex functionals | |
727 It is common for Lisp functions to accept functions as arguments or | |
728 find them in data structures (especially in hook variables and property | |
729 lists) and call them using @code{funcall} or @code{apply}. Functions | |
730 that accept function arguments are often called @dfn{functionals}. | |
731 | |
732 Sometimes, when you call a functional, it is useful to supply a no-op | |
733 function as the argument. Here are two different kinds of no-op | |
734 function: | |
735 | |
736 @defun identity arg | |
737 This function returns @var{arg} and has no side effects. | |
738 @end defun | |
739 | |
740 @defun ignore &rest args | |
741 This function ignores any arguments and returns @code{nil}. | |
742 @end defun | |
743 | |
744 @node Mapping Functions | |
745 @section Mapping Functions | |
746 @cindex mapping functions | |
747 | |
748 A @dfn{mapping function} applies a given function (@emph{not} a | |
749 special form or macro) to each element of a list or other collection. | |
750 Emacs Lisp has several such functions; @code{mapcar} and | |
751 @code{mapconcat}, which scan a list, are described here. | |
752 @xref{Definition of mapatoms}, for the function @code{mapatoms} which | |
753 maps over the symbols in an obarray. @xref{Definition of maphash}, | |
754 for the function @code{maphash} which maps over key/value associations | |
755 in a hash table. | |
756 | |
757 These mapping functions do not allow char-tables because a char-table | |
758 is a sparse array whose nominal range of indices is very large. To map | |
759 over a char-table in a way that deals properly with its sparse nature, | |
760 use the function @code{map-char-table} (@pxref{Char-Tables}). | |
761 | |
762 @defun mapcar function sequence | |
763 @anchor{Definition of mapcar} | |
764 @code{mapcar} applies @var{function} to each element of @var{sequence} | |
765 in turn, and returns a list of the results. | |
766 | |
767 The argument @var{sequence} can be any kind of sequence except a | |
768 char-table; that is, a list, a vector, a bool-vector, or a string. The | |
769 result is always a list. The length of the result is the same as the | |
770 length of @var{sequence}. For example: | |
771 | |
772 @smallexample | |
773 @group | |
774 (mapcar 'car '((a b) (c d) (e f))) | |
775 @result{} (a c e) | |
776 (mapcar '1+ [1 2 3]) | |
777 @result{} (2 3 4) | |
778 (mapcar 'char-to-string "abc") | |
779 @result{} ("a" "b" "c") | |
780 @end group | |
781 | |
782 @group | |
783 ;; @r{Call each function in @code{my-hooks}.} | |
784 (mapcar 'funcall my-hooks) | |
785 @end group | |
786 | |
787 @group | |
788 (defun mapcar* (function &rest args) | |
789 "Apply FUNCTION to successive cars of all ARGS. | |
790 Return the list of results." | |
791 ;; @r{If no list is exhausted,} | |
792 (if (not (memq nil args)) | |
793 ;; @r{apply function to @sc{car}s.} | |
794 (cons (apply function (mapcar 'car args)) | |
795 (apply 'mapcar* function | |
796 ;; @r{Recurse for rest of elements.} | |
797 (mapcar 'cdr args))))) | |
798 @end group | |
799 | |
800 @group | |
801 (mapcar* 'cons '(a b c) '(1 2 3 4)) | |
802 @result{} ((a . 1) (b . 2) (c . 3)) | |
803 @end group | |
804 @end smallexample | |
805 @end defun | |
806 | |
807 @defun mapc function sequence | |
808 @code{mapc} is like @code{mapcar} except that @var{function} is used for | |
809 side-effects only---the values it returns are ignored, not collected | |
810 into a list. @code{mapc} always returns @var{sequence}. | |
811 @end defun | |
812 | |
813 @defun mapconcat function sequence separator | |
814 @code{mapconcat} applies @var{function} to each element of | |
815 @var{sequence}: the results, which must be strings, are concatenated. | |
816 Between each pair of result strings, @code{mapconcat} inserts the string | |
817 @var{separator}. Usually @var{separator} contains a space or comma or | |
818 other suitable punctuation. | |
819 | |
820 The argument @var{function} must be a function that can take one | |
821 argument and return a string. The argument @var{sequence} can be any | |
822 kind of sequence except a char-table; that is, a list, a vector, a | |
823 bool-vector, or a string. | |
824 | |
825 @smallexample | |
826 @group | |
827 (mapconcat 'symbol-name | |
828 '(The cat in the hat) | |
829 " ") | |
830 @result{} "The cat in the hat" | |
831 @end group | |
832 | |
833 @group | |
834 (mapconcat (function (lambda (x) (format "%c" (1+ x)))) | |
835 "HAL-8000" | |
836 "") | |
837 @result{} "IBM.9111" | |
838 @end group | |
839 @end smallexample | |
840 @end defun | |
841 | |
842 @node Anonymous Functions | |
843 @section Anonymous Functions | |
844 @cindex anonymous function | |
845 | |
846 In Lisp, a function is a list that starts with @code{lambda}, a | |
847 byte-code function compiled from such a list, or alternatively a | |
848 primitive subr-object; names are ``extra.'' Although usually functions | |
849 are defined with @code{defun} and given names at the same time, it is | |
850 occasionally more concise to use an explicit lambda expression---an | |
851 anonymous function. Such a list is valid wherever a function name is. | |
852 | |
853 Any method of creating such a list makes a valid function. Even this: | |
854 | |
855 @smallexample | |
856 @group | |
857 (setq silly (append '(lambda (x)) (list (list '+ (* 3 4) 'x)))) | |
858 @result{} (lambda (x) (+ 12 x)) | |
859 @end group | |
860 @end smallexample | |
861 | |
862 @noindent | |
863 This computes a list that looks like @code{(lambda (x) (+ 12 x))} and | |
864 makes it the value (@emph{not} the function definition!) of | |
865 @code{silly}. | |
866 | |
867 Here is how we might call this function: | |
868 | |
869 @example | |
870 @group | |
871 (funcall silly 1) | |
872 @result{} 13 | |
873 @end group | |
874 @end example | |
875 | |
876 @noindent | |
877 (It does @emph{not} work to write @code{(silly 1)}, because this function | |
878 is not the @emph{function definition} of @code{silly}. We have not given | |
879 @code{silly} any function definition, just a value as a variable.) | |
880 | |
881 Most of the time, anonymous functions are constants that appear in | |
882 your program. For example, you might want to pass one as an argument to | |
883 the function @code{mapcar}, which applies any given function to each | |
884 element of a list. | |
885 | |
886 Here we define a function @code{change-property} which | |
887 uses a function as its third argument: | |
888 | |
889 @example | |
890 @group | |
891 (defun change-property (symbol prop function) | |
892 (let ((value (get symbol prop))) | |
893 (put symbol prop (funcall function value)))) | |
894 @end group | |
895 @end example | |
896 | |
897 @noindent | |
898 Here we define a function that uses @code{change-property}, | |
899 passing it a function to double a number: | |
900 | |
901 @example | |
902 @group | |
903 (defun double-property (symbol prop) | |
904 (change-property symbol prop '(lambda (x) (* 2 x)))) | |
905 @end group | |
906 @end example | |
907 | |
908 @noindent | |
909 In such cases, we usually use the special form @code{function} instead | |
910 of simple quotation to quote the anonymous function, like this: | |
911 | |
912 @example | |
913 @group | |
914 (defun double-property (symbol prop) | |
915 (change-property symbol prop | |
916 (function (lambda (x) (* 2 x))))) | |
917 @end group | |
918 @end example | |
919 | |
920 Using @code{function} instead of @code{quote} makes a difference if you | |
921 compile the function @code{double-property}. For example, if you | |
922 compile the second definition of @code{double-property}, the anonymous | |
923 function is compiled as well. By contrast, if you compile the first | |
924 definition which uses ordinary @code{quote}, the argument passed to | |
925 @code{change-property} is the precise list shown: | |
926 | |
927 @example | |
928 (lambda (x) (* x 2)) | |
929 @end example | |
930 | |
931 @noindent | |
932 The Lisp compiler cannot assume this list is a function, even though it | |
933 looks like one, since it does not know what @code{change-property} will | |
934 do with the list. Perhaps it will check whether the @sc{car} of the third | |
935 element is the symbol @code{*}! Using @code{function} tells the | |
936 compiler it is safe to go ahead and compile the constant function. | |
937 | |
938 Nowadays it is possible to omit @code{function} entirely, like this: | |
939 | |
940 @example | |
941 @group | |
942 (defun double-property (symbol prop) | |
943 (change-property symbol prop (lambda (x) (* 2 x)))) | |
944 @end group | |
945 @end example | |
946 | |
947 @noindent | |
948 This is because @code{lambda} itself implies @code{function}. | |
949 | |
950 We sometimes write @code{function} instead of @code{quote} when | |
951 quoting the name of a function, but this usage is just a sort of | |
952 comment: | |
953 | |
954 @example | |
955 (function @var{symbol}) @equiv{} (quote @var{symbol}) @equiv{} '@var{symbol} | |
956 @end example | |
957 | |
958 @cindex @samp{#'} syntax | |
959 The read syntax @code{#'} is a short-hand for using @code{function}. | |
960 For example, | |
961 | |
962 @example | |
963 #'(lambda (x) (* x x)) | |
964 @end example | |
965 | |
966 @noindent | |
967 is equivalent to | |
968 | |
969 @example | |
970 (function (lambda (x) (* x x))) | |
971 @end example | |
972 | |
973 @defspec function function-object | |
974 @cindex function quoting | |
975 This special form returns @var{function-object} without evaluating it. | |
976 In this, it is equivalent to @code{quote}. However, it serves as a | |
977 note to the Emacs Lisp compiler that @var{function-object} is intended | |
978 to be used only as a function, and therefore can safely be compiled. | |
979 Contrast this with @code{quote}, in @ref{Quoting}. | |
980 @end defspec | |
981 | |
982 @xref{describe-symbols example}, for a realistic example using | |
983 @code{function} and an anonymous function. | |
984 | |
985 @node Function Cells | |
986 @section Accessing Function Cell Contents | |
987 | |
988 The @dfn{function definition} of a symbol is the object stored in the | |
989 function cell of the symbol. The functions described here access, test, | |
990 and set the function cell of symbols. | |
991 | |
992 See also the function @code{indirect-function}. @xref{Definition of | |
993 indirect-function}. | |
994 | |
995 @defun symbol-function symbol | |
996 @kindex void-function | |
997 This returns the object in the function cell of @var{symbol}. If the | |
998 symbol's function cell is void, a @code{void-function} error is | |
999 signaled. | |
1000 | |
1001 This function does not check that the returned object is a legitimate | |
1002 function. | |
1003 | |
1004 @example | |
1005 @group | |
1006 (defun bar (n) (+ n 2)) | |
1007 @result{} bar | |
1008 @end group | |
1009 @group | |
1010 (symbol-function 'bar) | |
1011 @result{} (lambda (n) (+ n 2)) | |
1012 @end group | |
1013 @group | |
1014 (fset 'baz 'bar) | |
1015 @result{} bar | |
1016 @end group | |
1017 @group | |
1018 (symbol-function 'baz) | |
1019 @result{} bar | |
1020 @end group | |
1021 @end example | |
1022 @end defun | |
1023 | |
1024 @cindex void function cell | |
1025 If you have never given a symbol any function definition, we say that | |
1026 that symbol's function cell is @dfn{void}. In other words, the function | |
1027 cell does not have any Lisp object in it. If you try to call such a symbol | |
1028 as a function, it signals a @code{void-function} error. | |
1029 | |
1030 Note that void is not the same as @code{nil} or the symbol | |
1031 @code{void}. The symbols @code{nil} and @code{void} are Lisp objects, | |
1032 and can be stored into a function cell just as any other object can be | |
1033 (and they can be valid functions if you define them in turn with | |
1034 @code{defun}). A void function cell contains no object whatsoever. | |
1035 | |
1036 You can test the voidness of a symbol's function definition with | |
1037 @code{fboundp}. After you have given a symbol a function definition, you | |
1038 can make it void once more using @code{fmakunbound}. | |
1039 | |
1040 @defun fboundp symbol | |
1041 This function returns @code{t} if the symbol has an object in its | |
1042 function cell, @code{nil} otherwise. It does not check that the object | |
1043 is a legitimate function. | |
1044 @end defun | |
1045 | |
1046 @defun fmakunbound symbol | |
1047 This function makes @var{symbol}'s function cell void, so that a | |
1048 subsequent attempt to access this cell will cause a | |
1049 @code{void-function} error. It returns @var{symbol}. (See also | |
1050 @code{makunbound}, in @ref{Void Variables}.) | |
1051 | |
1052 @example | |
1053 @group | |
1054 (defun foo (x) x) | |
1055 @result{} foo | |
1056 @end group | |
1057 @group | |
1058 (foo 1) | |
1059 @result{}1 | |
1060 @end group | |
1061 @group | |
1062 (fmakunbound 'foo) | |
1063 @result{} foo | |
1064 @end group | |
1065 @group | |
1066 (foo 1) | |
1067 @error{} Symbol's function definition is void: foo | |
1068 @end group | |
1069 @end example | |
1070 @end defun | |
1071 | |
1072 @defun fset symbol definition | |
1073 This function stores @var{definition} in the function cell of | |
1074 @var{symbol}. The result is @var{definition}. Normally | |
1075 @var{definition} should be a function or the name of a function, but | |
1076 this is not checked. The argument @var{symbol} is an ordinary evaluated | |
1077 argument. | |
1078 | |
1079 There are three normal uses of this function: | |
1080 | |
1081 @itemize @bullet | |
1082 @item | |
1083 Copying one symbol's function definition to another---in other words, | |
1084 making an alternate name for a function. (If you think of this as the | |
1085 definition of the new name, you should use @code{defalias} instead of | |
1086 @code{fset}; see @ref{Definition of defalias}.) | |
1087 | |
1088 @item | |
1089 Giving a symbol a function definition that is not a list and therefore | |
1090 cannot be made with @code{defun}. For example, you can use @code{fset} | |
1091 to give a symbol @code{s1} a function definition which is another symbol | |
1092 @code{s2}; then @code{s1} serves as an alias for whatever definition | |
1093 @code{s2} presently has. (Once again use @code{defalias} instead of | |
1094 @code{fset} if you think of this as the definition of @code{s1}.) | |
1095 | |
1096 @item | |
1097 In constructs for defining or altering functions. If @code{defun} | |
1098 were not a primitive, it could be written in Lisp (as a macro) using | |
1099 @code{fset}. | |
1100 @end itemize | |
1101 | |
1102 Here are examples of these uses: | |
1103 | |
1104 @example | |
1105 @group | |
1106 ;; @r{Save @code{foo}'s definition in @code{old-foo}.} | |
1107 (fset 'old-foo (symbol-function 'foo)) | |
1108 @end group | |
1109 | |
1110 @group | |
1111 ;; @r{Make the symbol @code{car} the function definition of @code{xfirst}.} | |
1112 ;; @r{(Most likely, @code{defalias} would be better than @code{fset} here.)} | |
1113 (fset 'xfirst 'car) | |
1114 @result{} car | |
1115 @end group | |
1116 @group | |
1117 (xfirst '(1 2 3)) | |
1118 @result{} 1 | |
1119 @end group | |
1120 @group | |
1121 (symbol-function 'xfirst) | |
1122 @result{} car | |
1123 @end group | |
1124 @group | |
1125 (symbol-function (symbol-function 'xfirst)) | |
1126 @result{} #<subr car> | |
1127 @end group | |
1128 | |
1129 @group | |
1130 ;; @r{Define a named keyboard macro.} | |
1131 (fset 'kill-two-lines "\^u2\^k") | |
1132 @result{} "\^u2\^k" | |
1133 @end group | |
1134 | |
1135 @group | |
1136 ;; @r{Here is a function that alters other functions.} | |
1137 (defun copy-function-definition (new old) | |
1138 "Define NEW with the same function definition as OLD." | |
1139 (fset new (symbol-function old))) | |
1140 @end group | |
1141 @end example | |
1142 @end defun | |
1143 | |
1144 @code{fset} is sometimes used to save the old definition of a | |
1145 function before redefining it. That permits the new definition to | |
1146 invoke the old definition. But it is unmodular and unclean for a Lisp | |
1147 file to redefine a function defined elsewhere. If you want to modify | |
1148 a function defined by another package, it is cleaner to use | |
1149 @code{defadvice} (@pxref{Advising Functions}). | |
1150 | |
1151 @node Obsolete Functions | |
1152 @section Declaring Functions Obsolete | |
1153 | |
1154 You can use @code{make-obsolete} to declare a function obsolete. This | |
1155 indicates that the function may be removed at some stage in the future. | |
1156 | |
1157 @defun make-obsolete obsolete-name current-name &optional when | |
1158 This function makes the byte compiler warn that the function | |
1159 @var{obsolete-name} is obsolete. If @var{current-name} is a symbol, the | |
1160 warning message says to use @var{current-name} instead of | |
1161 @var{obsolete-name}. @var{current-name} does not need to be an alias for | |
1162 @var{obsolete-name}; it can be a different function with similar | |
1163 functionality. If @var{current-name} is a string, it is the warning | |
1164 message. | |
1165 | |
1166 If provided, @var{when} should be a string indicating when the function | |
1167 was first made obsolete---for example, a date or a release number. | |
1168 @end defun | |
1169 | |
1170 You can define a function as an alias and declare it obsolete at the | |
1171 same time using the macro @code{define-obsolete-function-alias}. | |
1172 | |
1173 @defmac define-obsolete-function-alias obsolete-name current-name &optional when docstring | |
1174 This macro marks the function @var{obsolete-name} obsolete and also | |
1175 defines it as an alias for the function @var{current-name}. It is | |
1176 equivalent to the following: | |
1177 | |
1178 @example | |
1179 (defalias @var{obsolete-name} @var{current-name} @var{docstring}) | |
1180 (make-obsolete @var{obsolete-name} @var{current-name} @var{when}) | |
1181 @end example | |
1182 @end defmac | |
1183 | |
1184 @node Inline Functions | |
1185 @section Inline Functions | |
1186 @cindex inline functions | |
1187 | |
1188 @findex defsubst | |
1189 You can define an @dfn{inline function} by using @code{defsubst} instead | |
1190 of @code{defun}. An inline function works just like an ordinary | |
1191 function except for one thing: when you compile a call to the function, | |
1192 the function's definition is open-coded into the caller. | |
1193 | |
1194 Making a function inline makes explicit calls run faster. But it also | |
1195 has disadvantages. For one thing, it reduces flexibility; if you | |
1196 change the definition of the function, calls already inlined still use | |
1197 the old definition until you recompile them. | |
1198 | |
1199 Another disadvantage is that making a large function inline can increase | |
1200 the size of compiled code both in files and in memory. Since the speed | |
1201 advantage of inline functions is greatest for small functions, you | |
1202 generally should not make large functions inline. | |
1203 | |
1204 Also, inline functions do not behave well with respect to debugging, | |
1205 tracing, and advising (@pxref{Advising Functions}). Since ease of | |
1206 debugging and the flexibility of redefining functions are important | |
1207 features of Emacs, you should not make a function inline, even if it's | |
1208 small, unless its speed is really crucial, and you've timed the code | |
1209 to verify that using @code{defun} actually has performance problems. | |
1210 | |
1211 It's possible to define a macro to expand into the same code that an | |
1212 inline function would execute. (@xref{Macros}.) But the macro would be | |
1213 limited to direct use in expressions---a macro cannot be called with | |
1214 @code{apply}, @code{mapcar} and so on. Also, it takes some work to | |
1215 convert an ordinary function into a macro. To convert it into an inline | |
1216 function is very easy; simply replace @code{defun} with @code{defsubst}. | |
1217 Since each argument of an inline function is evaluated exactly once, you | |
1218 needn't worry about how many times the body uses the arguments, as you | |
1219 do for macros. (@xref{Argument Evaluation}.) | |
1220 | |
1221 Inline functions can be used and open-coded later on in the same file, | |
1222 following the definition, just like macros. | |
1223 | |
86383
382d8b0539b2
(Declaring Functions): New section.
Glenn Morris <rgm@gnu.org>
parents:
85688
diff
changeset
|
1224 @node Declaring Functions |
382d8b0539b2
(Declaring Functions): New section.
Glenn Morris <rgm@gnu.org>
parents:
85688
diff
changeset
|
1225 @section Telling the Compiler that a Function is Defined |
382d8b0539b2
(Declaring Functions): New section.
Glenn Morris <rgm@gnu.org>
parents:
85688
diff
changeset
|
1226 @cindex function declaration |
382d8b0539b2
(Declaring Functions): New section.
Glenn Morris <rgm@gnu.org>
parents:
85688
diff
changeset
|
1227 @cindex declaring functions |
86826
9b003519e0dd
(Declaring Functions): Add findex. Mention `external' files.
Glenn Morris <rgm@gnu.org>
parents:
86447
diff
changeset
|
1228 @findex declare-function |
86383
382d8b0539b2
(Declaring Functions): New section.
Glenn Morris <rgm@gnu.org>
parents:
85688
diff
changeset
|
1229 |
86408
c4f8d28a93e7
(Declaring Functions): Clarify previous change.
Richard M. Stallman <rms@gnu.org>
parents:
86383
diff
changeset
|
1230 Byte-compiling a file often produces warnings about functions that the |
c4f8d28a93e7
(Declaring Functions): Clarify previous change.
Richard M. Stallman <rms@gnu.org>
parents:
86383
diff
changeset
|
1231 compiler doesn't know about (@pxref{Compiler Errors}). Sometimes this |
c4f8d28a93e7
(Declaring Functions): Clarify previous change.
Richard M. Stallman <rms@gnu.org>
parents:
86383
diff
changeset
|
1232 indicates a real problem, but usually the functions in question are |
c4f8d28a93e7
(Declaring Functions): Clarify previous change.
Richard M. Stallman <rms@gnu.org>
parents:
86383
diff
changeset
|
1233 defined in other files which would be loaded if that code is run. For |
c4f8d28a93e7
(Declaring Functions): Clarify previous change.
Richard M. Stallman <rms@gnu.org>
parents:
86383
diff
changeset
|
1234 example, byte-compiling @file{fortran.el} used to warn: |
86383
382d8b0539b2
(Declaring Functions): New section.
Glenn Morris <rgm@gnu.org>
parents:
85688
diff
changeset
|
1235 |
86408
c4f8d28a93e7
(Declaring Functions): Clarify previous change.
Richard M. Stallman <rms@gnu.org>
parents:
86383
diff
changeset
|
1236 @smallexample |
86383
382d8b0539b2
(Declaring Functions): New section.
Glenn Morris <rgm@gnu.org>
parents:
85688
diff
changeset
|
1237 In end of data: |
86408
c4f8d28a93e7
(Declaring Functions): Clarify previous change.
Richard M. Stallman <rms@gnu.org>
parents:
86383
diff
changeset
|
1238 fortran.el:2152:1:Warning: the function `gud-find-c-expr' is not known to be defined. |
c4f8d28a93e7
(Declaring Functions): Clarify previous change.
Richard M. Stallman <rms@gnu.org>
parents:
86383
diff
changeset
|
1239 @end smallexample |
86383
382d8b0539b2
(Declaring Functions): New section.
Glenn Morris <rgm@gnu.org>
parents:
85688
diff
changeset
|
1240 |
86408
c4f8d28a93e7
(Declaring Functions): Clarify previous change.
Richard M. Stallman <rms@gnu.org>
parents:
86383
diff
changeset
|
1241 In fact, @code{gud-find-c-expr} is only used in the function that |
c4f8d28a93e7
(Declaring Functions): Clarify previous change.
Richard M. Stallman <rms@gnu.org>
parents:
86383
diff
changeset
|
1242 Fortran mode uses for the local value of |
c4f8d28a93e7
(Declaring Functions): Clarify previous change.
Richard M. Stallman <rms@gnu.org>
parents:
86383
diff
changeset
|
1243 @code{gud-find-expr-function}, which is a callback from GUD; if it is |
c4f8d28a93e7
(Declaring Functions): Clarify previous change.
Richard M. Stallman <rms@gnu.org>
parents:
86383
diff
changeset
|
1244 called, the GUD functions will be loaded. When you know that such a |
c4f8d28a93e7
(Declaring Functions): Clarify previous change.
Richard M. Stallman <rms@gnu.org>
parents:
86383
diff
changeset
|
1245 warning does not indicate a real problem, it is good to suppress the |
c4f8d28a93e7
(Declaring Functions): Clarify previous change.
Richard M. Stallman <rms@gnu.org>
parents:
86383
diff
changeset
|
1246 warning. That makes new warnings which might mean real problems more |
c4f8d28a93e7
(Declaring Functions): Clarify previous change.
Richard M. Stallman <rms@gnu.org>
parents:
86383
diff
changeset
|
1247 visible. You do that with @code{declare-function}. |
86383
382d8b0539b2
(Declaring Functions): New section.
Glenn Morris <rgm@gnu.org>
parents:
85688
diff
changeset
|
1248 |
382d8b0539b2
(Declaring Functions): New section.
Glenn Morris <rgm@gnu.org>
parents:
85688
diff
changeset
|
1249 All you need to do is add a @code{declare-function} statement before the |
382d8b0539b2
(Declaring Functions): New section.
Glenn Morris <rgm@gnu.org>
parents:
85688
diff
changeset
|
1250 first use of the function in question: |
382d8b0539b2
(Declaring Functions): New section.
Glenn Morris <rgm@gnu.org>
parents:
85688
diff
changeset
|
1251 |
86408
c4f8d28a93e7
(Declaring Functions): Clarify previous change.
Richard M. Stallman <rms@gnu.org>
parents:
86383
diff
changeset
|
1252 @smallexample |
86383
382d8b0539b2
(Declaring Functions): New section.
Glenn Morris <rgm@gnu.org>
parents:
85688
diff
changeset
|
1253 (declare-function gud-find-c-expr "gud.el" nil) |
86447
598f0ef502b4
(Declaring Functions): Fix typo in directive.
Juanma Barranquero <lekktu@gmail.com>
parents:
86408
diff
changeset
|
1254 @end smallexample |
86383
382d8b0539b2
(Declaring Functions): New section.
Glenn Morris <rgm@gnu.org>
parents:
85688
diff
changeset
|
1255 |
382d8b0539b2
(Declaring Functions): New section.
Glenn Morris <rgm@gnu.org>
parents:
85688
diff
changeset
|
1256 This says that @code{gud-find-c-expr} is defined in @file{gud.el} (the |
86408
c4f8d28a93e7
(Declaring Functions): Clarify previous change.
Richard M. Stallman <rms@gnu.org>
parents:
86383
diff
changeset
|
1257 @samp{.el} can be omitted). The compiler takes for granted that that file |
c4f8d28a93e7
(Declaring Functions): Clarify previous change.
Richard M. Stallman <rms@gnu.org>
parents:
86383
diff
changeset
|
1258 really defines the function, and does not check. |
86447
598f0ef502b4
(Declaring Functions): Fix typo in directive.
Juanma Barranquero <lekktu@gmail.com>
parents:
86408
diff
changeset
|
1259 |
86408
c4f8d28a93e7
(Declaring Functions): Clarify previous change.
Richard M. Stallman <rms@gnu.org>
parents:
86383
diff
changeset
|
1260 The optional third argument specifies the argument list of |
c4f8d28a93e7
(Declaring Functions): Clarify previous change.
Richard M. Stallman <rms@gnu.org>
parents:
86383
diff
changeset
|
1261 @code{gud-find-c-expr}. In this case, it takes no arguments |
c4f8d28a93e7
(Declaring Functions): Clarify previous change.
Richard M. Stallman <rms@gnu.org>
parents:
86383
diff
changeset
|
1262 (@code{nil} is different from not specifying a value). In other |
c4f8d28a93e7
(Declaring Functions): Clarify previous change.
Richard M. Stallman <rms@gnu.org>
parents:
86383
diff
changeset
|
1263 cases, this might be something like @code{(file &optional overwrite)}. |
c4f8d28a93e7
(Declaring Functions): Clarify previous change.
Richard M. Stallman <rms@gnu.org>
parents:
86383
diff
changeset
|
1264 You don't have to specify the argument list, but if you do the |
c4f8d28a93e7
(Declaring Functions): Clarify previous change.
Richard M. Stallman <rms@gnu.org>
parents:
86383
diff
changeset
|
1265 byte compiler can check that the calls match the declaration. |
c4f8d28a93e7
(Declaring Functions): Clarify previous change.
Richard M. Stallman <rms@gnu.org>
parents:
86383
diff
changeset
|
1266 |
86845
62955c80c79b
(Declaring Functions): Add optional fourth
Glenn Morris <rgm@gnu.org>
parents:
86826
diff
changeset
|
1267 @defmac declare-function function file &optional arglist fileonly |
86408
c4f8d28a93e7
(Declaring Functions): Clarify previous change.
Richard M. Stallman <rms@gnu.org>
parents:
86383
diff
changeset
|
1268 Tell the byte compiler to assume that @var{function} is defined, with |
c4f8d28a93e7
(Declaring Functions): Clarify previous change.
Richard M. Stallman <rms@gnu.org>
parents:
86383
diff
changeset
|
1269 arguments @var{arglist}, and that the definition should come from |
86845
62955c80c79b
(Declaring Functions): Add optional fourth
Glenn Morris <rgm@gnu.org>
parents:
86826
diff
changeset
|
1270 the file @var{file}. @var{fileonly} non-nil means only check that |
62955c80c79b
(Declaring Functions): Add optional fourth
Glenn Morris <rgm@gnu.org>
parents:
86826
diff
changeset
|
1271 @var{file} exists, not that it actually defines @var{function}. |
86408
c4f8d28a93e7
(Declaring Functions): Clarify previous change.
Richard M. Stallman <rms@gnu.org>
parents:
86383
diff
changeset
|
1272 @end defmac |
86383
382d8b0539b2
(Declaring Functions): New section.
Glenn Morris <rgm@gnu.org>
parents:
85688
diff
changeset
|
1273 |
86408
c4f8d28a93e7
(Declaring Functions): Clarify previous change.
Richard M. Stallman <rms@gnu.org>
parents:
86383
diff
changeset
|
1274 To verify that these functions really are declared where |
c4f8d28a93e7
(Declaring Functions): Clarify previous change.
Richard M. Stallman <rms@gnu.org>
parents:
86383
diff
changeset
|
1275 @code{declare-function} says they are, use @code{check-declare-file} |
c4f8d28a93e7
(Declaring Functions): Clarify previous change.
Richard M. Stallman <rms@gnu.org>
parents:
86383
diff
changeset
|
1276 to check all @code{declare-function} calls in one source file, or use |
c4f8d28a93e7
(Declaring Functions): Clarify previous change.
Richard M. Stallman <rms@gnu.org>
parents:
86383
diff
changeset
|
1277 @code{check-declare-directory} check all the files in and under a |
c4f8d28a93e7
(Declaring Functions): Clarify previous change.
Richard M. Stallman <rms@gnu.org>
parents:
86383
diff
changeset
|
1278 certain directory. |
86383
382d8b0539b2
(Declaring Functions): New section.
Glenn Morris <rgm@gnu.org>
parents:
85688
diff
changeset
|
1279 |
86408
c4f8d28a93e7
(Declaring Functions): Clarify previous change.
Richard M. Stallman <rms@gnu.org>
parents:
86383
diff
changeset
|
1280 These commands find the file that ought to contain a function's |
c4f8d28a93e7
(Declaring Functions): Clarify previous change.
Richard M. Stallman <rms@gnu.org>
parents:
86383
diff
changeset
|
1281 definition using @code{locate-library}; if that finds no file, they |
c4f8d28a93e7
(Declaring Functions): Clarify previous change.
Richard M. Stallman <rms@gnu.org>
parents:
86383
diff
changeset
|
1282 expand the definition file name relative to the directory of the file |
c4f8d28a93e7
(Declaring Functions): Clarify previous change.
Richard M. Stallman <rms@gnu.org>
parents:
86383
diff
changeset
|
1283 that contains the @code{declare-function} call. |
c4f8d28a93e7
(Declaring Functions): Clarify previous change.
Richard M. Stallman <rms@gnu.org>
parents:
86383
diff
changeset
|
1284 |
c4f8d28a93e7
(Declaring Functions): Clarify previous change.
Richard M. Stallman <rms@gnu.org>
parents:
86383
diff
changeset
|
1285 You can also say that a function is defined by C code by specifying |
c4f8d28a93e7
(Declaring Functions): Clarify previous change.
Richard M. Stallman <rms@gnu.org>
parents:
86383
diff
changeset
|
1286 a file name ending in @samp{.c}. @code{check-declare-file} looks for |
c4f8d28a93e7
(Declaring Functions): Clarify previous change.
Richard M. Stallman <rms@gnu.org>
parents:
86383
diff
changeset
|
1287 these files in the C source code directory. This is useful only when |
c4f8d28a93e7
(Declaring Functions): Clarify previous change.
Richard M. Stallman <rms@gnu.org>
parents:
86383
diff
changeset
|
1288 you call a function that is defined only on certain systems. Most |
c4f8d28a93e7
(Declaring Functions): Clarify previous change.
Richard M. Stallman <rms@gnu.org>
parents:
86383
diff
changeset
|
1289 of the primitive functions of Emacs are always defined so they will |
c4f8d28a93e7
(Declaring Functions): Clarify previous change.
Richard M. Stallman <rms@gnu.org>
parents:
86383
diff
changeset
|
1290 never give you a warning. |
86383
382d8b0539b2
(Declaring Functions): New section.
Glenn Morris <rgm@gnu.org>
parents:
85688
diff
changeset
|
1291 |
86826
9b003519e0dd
(Declaring Functions): Add findex. Mention `external' files.
Glenn Morris <rgm@gnu.org>
parents:
86447
diff
changeset
|
1292 Sometimes a file will optionally use functions from an external package. |
9b003519e0dd
(Declaring Functions): Add findex. Mention `external' files.
Glenn Morris <rgm@gnu.org>
parents:
86447
diff
changeset
|
1293 If you prefix the filename in the @code{declare-function} statement with |
9b003519e0dd
(Declaring Functions): Add findex. Mention `external' files.
Glenn Morris <rgm@gnu.org>
parents:
86447
diff
changeset
|
1294 @samp{ext:}, then it will be checked if it is found, otherwise skipped |
9b003519e0dd
(Declaring Functions): Add findex. Mention `external' files.
Glenn Morris <rgm@gnu.org>
parents:
86447
diff
changeset
|
1295 without error. |
9b003519e0dd
(Declaring Functions): Add findex. Mention `external' files.
Glenn Morris <rgm@gnu.org>
parents:
86447
diff
changeset
|
1296 |
86845
62955c80c79b
(Declaring Functions): Add optional fourth
Glenn Morris <rgm@gnu.org>
parents:
86826
diff
changeset
|
1297 There are some function definitions that @samp{check-declare} does not |
62955c80c79b
(Declaring Functions): Add optional fourth
Glenn Morris <rgm@gnu.org>
parents:
86826
diff
changeset
|
1298 understand (e.g. @code{defstruct} and some other macros). In such cases, |
86884
bcf9ca31cd6d
(Declaring Functions): Improve previous change (arguments by name, not number).
Glenn Morris <rgm@gnu.org>
parents:
86845
diff
changeset
|
1299 you can pass a non-@code{nil} @var{fileonly} argument to |
bcf9ca31cd6d
(Declaring Functions): Improve previous change (arguments by name, not number).
Glenn Morris <rgm@gnu.org>
parents:
86845
diff
changeset
|
1300 @code{declare-function}, meaning to only check that the file exists, not |
bcf9ca31cd6d
(Declaring Functions): Improve previous change (arguments by name, not number).
Glenn Morris <rgm@gnu.org>
parents:
86845
diff
changeset
|
1301 that it actually defines the function. Note that to do this without |
bcf9ca31cd6d
(Declaring Functions): Improve previous change (arguments by name, not number).
Glenn Morris <rgm@gnu.org>
parents:
86845
diff
changeset
|
1302 having to specify an argument list, you should set the @var{arglist} |
bcf9ca31cd6d
(Declaring Functions): Improve previous change (arguments by name, not number).
Glenn Morris <rgm@gnu.org>
parents:
86845
diff
changeset
|
1303 argument to @code{t} (because @code{nil} means an empty argument list, as |
bcf9ca31cd6d
(Declaring Functions): Improve previous change (arguments by name, not number).
Glenn Morris <rgm@gnu.org>
parents:
86845
diff
changeset
|
1304 opposed to an unspecified one). |
86845
62955c80c79b
(Declaring Functions): Add optional fourth
Glenn Morris <rgm@gnu.org>
parents:
86826
diff
changeset
|
1305 |
84070 | 1306 @node Function Safety |
1307 @section Determining whether a Function is Safe to Call | |
1308 @cindex function safety | |
1309 @cindex safety of functions | |
1310 | |
1311 Some major modes such as SES call functions that are stored in user | |
1312 files. (@inforef{Top, ,ses}, for more information on SES.) User | |
1313 files sometimes have poor pedigrees---you can get a spreadsheet from | |
1314 someone you've just met, or you can get one through email from someone | |
1315 you've never met. So it is risky to call a function whose source code | |
1316 is stored in a user file until you have determined that it is safe. | |
1317 | |
1318 @defun unsafep form &optional unsafep-vars | |
1319 Returns @code{nil} if @var{form} is a @dfn{safe} Lisp expression, or | |
1320 returns a list that describes why it might be unsafe. The argument | |
1321 @var{unsafep-vars} is a list of symbols known to have temporary | |
1322 bindings at this point; it is mainly used for internal recursive | |
1323 calls. The current buffer is an implicit argument, which provides a | |
1324 list of buffer-local bindings. | |
1325 @end defun | |
1326 | |
1327 Being quick and simple, @code{unsafep} does a very light analysis and | |
1328 rejects many Lisp expressions that are actually safe. There are no | |
1329 known cases where @code{unsafep} returns @code{nil} for an unsafe | |
1330 expression. However, a ``safe'' Lisp expression can return a string | |
1331 with a @code{display} property, containing an associated Lisp | |
1332 expression to be executed after the string is inserted into a buffer. | |
1333 This associated expression can be a virus. In order to be safe, you | |
1334 must delete properties from all strings calculated by user code before | |
1335 inserting them into buffers. | |
1336 | |
1337 @ignore | |
1338 What is a safe Lisp expression? Basically, it's an expression that | |
1339 calls only built-in functions with no side effects (or only innocuous | |
1340 ones). Innocuous side effects include displaying messages and | |
1341 altering non-risky buffer-local variables (but not global variables). | |
1342 | |
1343 @table @dfn | |
1344 @item Safe expression | |
1345 @itemize | |
1346 @item | |
1347 An atom or quoted thing. | |
1348 @item | |
1349 A call to a safe function (see below), if all its arguments are | |
1350 safe expressions. | |
1351 @item | |
1352 One of the special forms @code{and}, @code{catch}, @code{cond}, | |
1353 @code{if}, @code{or}, @code{prog1}, @code{prog2}, @code{progn}, | |
1354 @code{while}, and @code{unwind-protect}], if all its arguments are | |
1355 safe. | |
1356 @item | |
1357 A form that creates temporary bindings (@code{condition-case}, | |
1358 @code{dolist}, @code{dotimes}, @code{lambda}, @code{let}, or | |
1359 @code{let*}), if all args are safe and the symbols to be bound are not | |
1360 explicitly risky (see @pxref{File Local Variables}). | |
1361 @item | |
1362 An assignment using @code{add-to-list}, @code{setq}, @code{push}, or | |
1363 @code{pop}, if all args are safe and the symbols to be assigned are | |
1364 not explicitly risky and they already have temporary or buffer-local | |
1365 bindings. | |
1366 @item | |
1367 One of [apply, mapc, mapcar, mapconcat] if the first argument is a | |
1368 safe explicit lambda and the other args are safe expressions. | |
1369 @end itemize | |
1370 | |
1371 @item Safe function | |
1372 @itemize | |
1373 @item | |
1374 A lambda containing safe expressions. | |
1375 @item | |
1376 A symbol on the list @code{safe-functions}, so the user says it's safe. | |
1377 @item | |
1378 A symbol with a non-@code{nil} @code{side-effect-free} property. | |
1379 @item | |
1380 A symbol with a non-@code{nil} @code{safe-function} property. Value t | |
1381 indicates a function that is safe but has innocuous side effects. | |
1382 Other values will someday indicate functions with classes of side | |
1383 effects that are not always safe. | |
1384 @end itemize | |
1385 | |
1386 The @code{side-effect-free} and @code{safe-function} properties are | |
1387 provided for built-in functions and for low-level functions and macros | |
1388 defined in @file{subr.el}. You can assign these properties for the | |
1389 functions you write. | |
1390 @end table | |
1391 @end ignore | |
1392 | |
1393 @node Related Topics | |
1394 @section Other Topics Related to Functions | |
1395 | |
1396 Here is a table of several functions that do things related to | |
1397 function calling and function definitions. They are documented | |
1398 elsewhere, but we provide cross references here. | |
1399 | |
1400 @table @code | |
1401 @item apply | |
1402 See @ref{Calling Functions}. | |
1403 | |
1404 @item autoload | |
1405 See @ref{Autoload}. | |
1406 | |
1407 @item call-interactively | |
1408 See @ref{Interactive Call}. | |
1409 | |
85688 | 1410 @item called-interactively-p |
1411 See @ref{Distinguish Interactive}. | |
1412 | |
84070 | 1413 @item commandp |
1414 See @ref{Interactive Call}. | |
1415 | |
1416 @item documentation | |
1417 See @ref{Accessing Documentation}. | |
1418 | |
1419 @item eval | |
1420 See @ref{Eval}. | |
1421 | |
1422 @item funcall | |
1423 See @ref{Calling Functions}. | |
1424 | |
1425 @item function | |
1426 See @ref{Anonymous Functions}. | |
1427 | |
1428 @item ignore | |
1429 See @ref{Calling Functions}. | |
1430 | |
1431 @item indirect-function | |
1432 See @ref{Function Indirection}. | |
1433 | |
1434 @item interactive | |
1435 See @ref{Using Interactive}. | |
1436 | |
1437 @item interactive-p | |
85688 | 1438 See @ref{Distinguish Interactive}. |
84070 | 1439 |
1440 @item mapatoms | |
1441 See @ref{Creating Symbols}. | |
1442 | |
1443 @item mapcar | |
1444 See @ref{Mapping Functions}. | |
1445 | |
1446 @item map-char-table | |
1447 See @ref{Char-Tables}. | |
1448 | |
1449 @item mapconcat | |
1450 See @ref{Mapping Functions}. | |
1451 | |
1452 @item undefined | |
1453 See @ref{Functions for Key Lookup}. | |
1454 @end table | |
1455 | |
1456 @ignore | |
1457 arch-tag: 39100cdf-8a55-4898-acba-595db619e8e2 | |
1458 @end ignore |