Mercurial > emacs
annotate man/cl.texi @ 45390:c9c1367befe1
*** empty log message ***
author | Miles Bader <miles@gnu.org> |
---|---|
date | Mon, 20 May 2002 04:40:53 +0000 |
parents | 5b63c25571a8 |
children | 0a70200bde27 |
rev | line source |
---|---|
25829 | 1 \input texinfo @c -*-texinfo-*- |
2 @setfilename ../info/cl | |
3 @settitle Common Lisp Extensions | |
4 | |
30009 | 5 @dircategory Emacs |
25829 | 6 @direntry |
7 * CL: (cl). Partial Common Lisp support for Emacs Lisp. | |
8 @end direntry | |
9 | |
10 @iftex | |
11 @finalout | |
12 @end iftex | |
13 | |
14 @ifinfo | |
15 This file documents the GNU Emacs Common Lisp emulation package. | |
16 | |
17 Copyright (C) 1993 Free Software Foundation, Inc. | |
18 | |
32315 | 19 |
20 Permission is granted to copy, distribute and/or modify this document | |
21 under the terms of the GNU Free Documentation License, Version 1.1 or | |
22 any later version published by the Free Software Foundation; with no | |
23 Invariant Sections, with the Front-Cover texts being ``A GNU | |
24 Manual'', and with the Back-Cover Texts as in (a) below. A copy of the | |
25 license is included in the section entitled ``GNU Free Documentation | |
26 License'' in the Emacs manual. | |
27 | |
28 (a) The FSF's Back-Cover Text is: ``You have freedom to copy and modify | |
29 this GNU Manual, like GNU software. Copies published by the Free | |
30 Software Foundation raise funds for GNU development.'' | |
31 | |
32 This document is part of a collection distributed under the GNU Free | |
33 Documentation License. If you want to distribute this document | |
34 separately from the collection, you can do so by adding a copy of the | |
35 license to the document, as described in section 6 of the license. | |
25829 | 36 @end ifinfo |
37 | |
38 @titlepage | |
39 @sp 6 | |
40 @center @titlefont{Common Lisp Extensions} | |
41 @sp 4 | |
42 @center For GNU Emacs Lisp | |
43 @sp 1 | |
44 @center Version 2.02 | |
45 @sp 5 | |
46 @center Dave Gillespie | |
47 @center daveg@@synaptics.com | |
48 @page | |
49 | |
50 @vskip 0pt plus 1filll | |
51 Copyright @copyright{} 1993 Free Software Foundation, Inc. | |
52 | |
32315 | 53 Permission is granted to copy, distribute and/or modify this document |
54 under the terms of the GNU Free Documentation License, Version 1.1 or | |
55 any later version published by the Free Software Foundation; with no | |
56 Invariant Sections, with the Front-Cover texts being ``A GNU | |
57 Manual'', and with the Back-Cover Texts as in (a) below. A copy of the | |
58 license is included in the section entitled ``GNU Free Documentation | |
59 License'' in the Emacs manual. | |
60 | |
61 (a) The FSF's Back-Cover Text is: ``You have freedom to copy and modify | |
62 this GNU Manual, like GNU software. Copies published by the Free | |
63 Software Foundation raise funds for GNU development.'' | |
64 | |
65 This document is part of a collection distributed under the GNU Free | |
66 Documentation License. If you want to distribute this document | |
67 separately from the collection, you can do so by adding a copy of the | |
68 license to the document, as described in section 6 of the license. | |
25829 | 69 @end titlepage |
70 | |
31572
384603523f41
Remove obsolescences, especially hash tables and refs to Emacs 19.
Dave Love <fx@gnu.org>
parents:
30009
diff
changeset
|
71 @node Top, Overview, (dir), (dir) |
25829 | 72 @chapter Common Lisp Extensions |
73 | |
74 @noindent | |
75 This document describes a set of Emacs Lisp facilities borrowed from | |
76 Common Lisp. All the facilities are described here in detail. While | |
77 this document does not assume any prior knowledge of Common Lisp, it | |
78 does assume a basic familiarity with Emacs Lisp. | |
79 | |
80 @menu | |
81 * Overview:: Installation, usage, etc. | |
82 * Program Structure:: Arglists, `eval-when', `defalias' | |
83 * Predicates:: `typep', `eql', and `equalp' | |
84 * Control Structure:: `setf', `do', `loop', etc. | |
85 * Macros:: Destructuring, `define-compiler-macro' | |
86 * Declarations:: `proclaim', `declare', etc. | |
87 * Symbols:: Property lists, `gensym' | |
88 * Numbers:: Predicates, functions, random numbers | |
89 * Sequences:: Mapping, functions, searching, sorting | |
90 * Lists:: `cadr', `sublis', `member*', `assoc*', etc. | |
91 * Structures:: `defstruct' | |
92 * Assertions:: `check-type', `assert', `ignore-errors'. | |
93 | |
94 * Efficiency Concerns:: Hints and techniques | |
95 * Common Lisp Compatibility:: All known differences with Steele | |
96 * Old CL Compatibility:: All known differences with old cl.el | |
97 * Porting Common Lisp:: Hints for porting Common Lisp code | |
98 | |
99 * Function Index:: | |
100 * Variable Index:: | |
101 @end menu | |
102 | |
103 @node Overview, Program Structure, Top, Top | |
104 @ifinfo | |
105 @chapter Overview | |
106 @end ifinfo | |
107 @iftex | |
108 @section Overview | |
109 @end iftex | |
110 | |
111 @noindent | |
112 Common Lisp is a huge language, and Common Lisp systems tend to be | |
113 massive and extremely complex. Emacs Lisp, by contrast, is rather | |
114 minimalist in the choice of Lisp features it offers the programmer. | |
115 As Emacs Lisp programmers have grown in number, and the applications | |
116 they write have grown more ambitious, it has become clear that Emacs | |
117 Lisp could benefit from many of the conveniences of Common Lisp. | |
118 | |
119 The @dfn{CL} package adds a number of Common Lisp functions and | |
120 control structures to Emacs Lisp. While not a 100% complete | |
121 implementation of Common Lisp, @dfn{CL} adds enough functionality | |
122 to make Emacs Lisp programming significantly more convenient. | |
123 | |
40590
e9ca70d27e37
Explain when not to use CL.
Richard M. Stallman <rms@gnu.org>
parents:
36931
diff
changeset
|
124 @strong{Please note:} the @dfn{CL} functions are not standard parts of |
e9ca70d27e37
Explain when not to use CL.
Richard M. Stallman <rms@gnu.org>
parents:
36931
diff
changeset
|
125 the Emacs Lisp name space, so it is legitimate for users to define |
e9ca70d27e37
Explain when not to use CL.
Richard M. Stallman <rms@gnu.org>
parents:
36931
diff
changeset
|
126 them with other, conflicting meanings. To avoid conflicting with |
e9ca70d27e37
Explain when not to use CL.
Richard M. Stallman <rms@gnu.org>
parents:
36931
diff
changeset
|
127 those user activities, we have a policy that packages installed in |
e9ca70d27e37
Explain when not to use CL.
Richard M. Stallman <rms@gnu.org>
parents:
36931
diff
changeset
|
128 Emacs must not load @dfn{CL} at run time. (It is ok for them to load |
e9ca70d27e37
Explain when not to use CL.
Richard M. Stallman <rms@gnu.org>
parents:
36931
diff
changeset
|
129 @dfn{CL} at compile time only, with @code{eval-when-compile}, and use |
e9ca70d27e37
Explain when not to use CL.
Richard M. Stallman <rms@gnu.org>
parents:
36931
diff
changeset
|
130 the macros it provides.) If you are writing packages that you plan to |
e9ca70d27e37
Explain when not to use CL.
Richard M. Stallman <rms@gnu.org>
parents:
36931
diff
changeset
|
131 distribute and invite widespread use for, you might want to observe |
e9ca70d27e37
Explain when not to use CL.
Richard M. Stallman <rms@gnu.org>
parents:
36931
diff
changeset
|
132 the same rule. |
e9ca70d27e37
Explain when not to use CL.
Richard M. Stallman <rms@gnu.org>
parents:
36931
diff
changeset
|
133 |
25829 | 134 Some Common Lisp features have been omitted from this package |
135 for various reasons: | |
136 | |
137 @itemize @bullet | |
138 @item | |
139 Some features are too complex or bulky relative to their benefit | |
140 to Emacs Lisp programmers. CLOS and Common Lisp streams are fine | |
141 examples of this group. | |
142 | |
143 @item | |
144 Other features cannot be implemented without modification to the | |
145 Emacs Lisp interpreter itself, such as multiple return values, | |
146 lexical scoping, case-insensitive symbols, and complex numbers. | |
147 The @dfn{CL} package generally makes no attempt to emulate these | |
148 features. | |
149 | |
150 @item | |
151 Some features conflict with existing things in Emacs Lisp. For | |
152 example, Emacs' @code{assoc} function is incompatible with the | |
153 Common Lisp @code{assoc}. In such cases, this package usually | |
154 adds the suffix @samp{*} to the function name of the Common | |
155 Lisp version of the function (e.g., @code{assoc*}). | |
156 @end itemize | |
157 | |
158 The package described here was written by Dave Gillespie, | |
159 @file{daveg@@synaptics.com}. It is a total rewrite of the original | |
160 1986 @file{cl.el} package by Cesar Quiroz. Most features of the | |
161 the Quiroz package have been retained; any incompatibilities are | |
162 noted in the descriptions below. Care has been taken in this | |
163 version to ensure that each function is defined efficiently, | |
164 concisely, and with minimal impact on the rest of the Emacs | |
165 environment. | |
166 | |
167 @menu | |
168 * Usage:: How to use the CL package | |
169 * Organization:: The package's five component files | |
170 * Installation:: Compiling and installing CL | |
171 * Naming Conventions:: Notes on CL function names | |
172 @end menu | |
173 | |
174 @node Usage, Organization, Overview, Overview | |
175 @section Usage | |
176 | |
177 @noindent | |
178 Lisp code that uses features from the @dfn{CL} package should | |
179 include at the beginning: | |
180 | |
181 @example | |
182 (require 'cl) | |
183 @end example | |
184 | |
185 @noindent | |
186 If you want to ensure that the new (Gillespie) version of @dfn{CL} | |
187 is the one that is present, add an additional @code{(require 'cl-19)} | |
188 call: | |
189 | |
190 @example | |
191 (require 'cl) | |
192 (require 'cl-19) | |
193 @end example | |
194 | |
195 @noindent | |
196 The second call will fail (with ``@file{cl-19.el} not found'') if | |
197 the old @file{cl.el} package was in use. | |
198 | |
199 It is safe to arrange to load @dfn{CL} at all times, e.g., | |
200 in your @file{.emacs} file. But it's a good idea, for portability, | |
201 to @code{(require 'cl)} in your code even if you do this. | |
202 | |
203 @node Organization, Installation, Usage, Overview | |
204 @section Organization | |
205 | |
206 @noindent | |
207 The Common Lisp package is organized into four files: | |
208 | |
209 @table @file | |
210 @item cl.el | |
211 This is the ``main'' file, which contains basic functions | |
212 and information about the package. This file is relatively | |
213 compact---about 700 lines. | |
214 | |
215 @item cl-extra.el | |
216 This file contains the larger, more complex or unusual functions. | |
217 It is kept separate so that packages which only want to use Common | |
218 Lisp fundamentals like the @code{cadr} function won't need to pay | |
219 the overhead of loading the more advanced functions. | |
220 | |
221 @item cl-seq.el | |
222 This file contains most of the advanced functions for operating | |
223 on sequences or lists, such as @code{delete-if} and @code{assoc*}. | |
224 | |
225 @item cl-macs.el | |
226 This file contains the features of the packages which are macros | |
227 instead of functions. Macros expand when the caller is compiled, | |
228 not when it is run, so the macros generally only need to be | |
229 present when the byte-compiler is running (or when the macros are | |
230 used in uncompiled code such as a @file{.emacs} file). Most of | |
231 the macros of this package are isolated in @file{cl-macs.el} so | |
232 that they won't take up memory unless you are compiling. | |
233 @end table | |
234 | |
235 The file @file{cl.el} includes all necessary @code{autoload} | |
236 commands for the functions and macros in the other three files. | |
237 All you have to do is @code{(require 'cl)}, and @file{cl.el} | |
238 will take care of pulling in the other files when they are | |
239 needed. | |
240 | |
241 There is another file, @file{cl-compat.el}, which defines some | |
242 routines from the older @file{cl.el} package that are no longer | |
243 present in the new package. This includes internal routines | |
244 like @code{setelt} and @code{zip-lists}, deprecated features | |
245 like @code{defkeyword}, and an emulation of the old-style | |
246 multiple-values feature. @xref{Old CL Compatibility}. | |
247 | |
248 @node Installation, Naming Conventions, Organization, Overview | |
249 @section Installation | |
250 | |
251 @noindent | |
252 Installation of the @dfn{CL} package is simple: Just put the | |
253 byte-compiled files @file{cl.elc}, @file{cl-extra.elc}, | |
254 @file{cl-seq.elc}, @file{cl-macs.elc}, and @file{cl-compat.elc} | |
255 into a directory on your @code{load-path}. | |
256 | |
257 There are no special requirements to compile this package: | |
258 The files do not have to be loaded before they are compiled, | |
259 nor do they need to be compiled in any particular order. | |
260 | |
261 You may choose to put the files into your main @file{lisp/} | |
262 directory, replacing the original @file{cl.el} file there. Or, | |
263 you could put them into a directory that comes before @file{lisp/} | |
264 on your @code{load-path} so that the old @file{cl.el} is | |
265 effectively hidden. | |
266 | |
267 Also, format the @file{cl.texinfo} file and put the resulting | |
268 Info files in the @file{info/} directory or another suitable place. | |
269 | |
270 You may instead wish to leave this package's components all in | |
271 their own directory, and then add this directory to your | |
31572
384603523f41
Remove obsolescences, especially hash tables and refs to Emacs 19.
Dave Love <fx@gnu.org>
parents:
30009
diff
changeset
|
272 @code{load-path} and @code{Info-directory-list}. |
25829 | 273 Add the directory to the front of the list so the old @dfn{CL} |
274 package and its documentation are hidden. | |
275 | |
31572
384603523f41
Remove obsolescences, especially hash tables and refs to Emacs 19.
Dave Love <fx@gnu.org>
parents:
30009
diff
changeset
|
276 @node Naming Conventions, , Installation, Overview |
25829 | 277 @section Naming Conventions |
278 | |
279 @noindent | |
280 Except where noted, all functions defined by this package have the | |
281 same names and calling conventions as their Common Lisp counterparts. | |
282 | |
283 Following is a complete list of functions whose names were changed | |
284 from Common Lisp, usually to avoid conflicts with Emacs. In each | |
285 case, a @samp{*} has been appended to the Common Lisp name to obtain | |
286 the Emacs name: | |
287 | |
288 @example | |
289 defun* defsubst* defmacro* function* | |
290 member* assoc* rassoc* get* | |
291 remove* delete* mapcar* sort* | |
292 floor* ceiling* truncate* round* | |
36931 | 293 mod* rem* random* |
25829 | 294 @end example |
295 | |
296 Internal function and variable names in the package are prefixed | |
297 by @code{cl-}. Here is a complete list of functions @emph{not} | |
298 prefixed by @code{cl-} which were not taken from Common Lisp: | |
299 | |
300 @example | |
28039
c3a446101beb
Remove references to rassoc, delete, expt.
Dave Love <fx@gnu.org>
parents:
27511
diff
changeset
|
301 floatp-safe lexical-let lexical-let* |
25829 | 302 callf callf2 letf letf* |
31572
384603523f41
Remove obsolescences, especially hash tables and refs to Emacs 19.
Dave Love <fx@gnu.org>
parents:
30009
diff
changeset
|
303 defsubst* |
25829 | 304 @end example |
305 | |
306 The following simple functions and macros are defined in @file{cl.el}; | |
307 they do not cause other components like @file{cl-extra} to be loaded. | |
308 | |
309 @example | |
27225
ee242fddb1ab
Remove various old stuff and references thereto.
Dave Love <fx@gnu.org>
parents:
25829
diff
changeset
|
310 eql floatp-safe endp |
25829 | 311 evenp oddp plusp minusp |
35089 | 312 caaar .. cddddr |
25829 | 313 list* ldiff rest first .. tenth |
27225
ee242fddb1ab
Remove various old stuff and references thereto.
Dave Love <fx@gnu.org>
parents:
25829
diff
changeset
|
314 copy-list subst mapcar* [2] |
25829 | 315 adjoin [3] acons pairlis pop [4] |
316 push [4] pushnew [3,4] incf [4] decf [4] | |
317 proclaim declaim | |
318 @end example | |
319 | |
320 @noindent | |
321 [2] Only for one sequence argument or two list arguments. | |
322 | |
323 @noindent | |
324 [3] Only if @code{:test} is @code{eq}, @code{equal}, or unspecified, | |
325 and @code{:key} is not used. | |
326 | |
327 @noindent | |
328 [4] Only when @var{place} is a plain variable name. | |
329 | |
330 @iftex | |
331 @chapno=4 | |
332 @end iftex | |
333 | |
334 @node Program Structure, Predicates, Overview, Top | |
335 @chapter Program Structure | |
336 | |
337 @noindent | |
338 This section describes features of the @dfn{CL} package which have to | |
339 do with programs as a whole: advanced argument lists for functions, | |
340 and the @code{eval-when} construct. | |
341 | |
342 @menu | |
343 * Argument Lists:: `&key', `&aux', `defun*', `defmacro*'. | |
344 * Time of Evaluation:: The `eval-when' construct. | |
345 @end menu | |
346 | |
347 @iftex | |
348 @secno=1 | |
349 @end iftex | |
350 | |
351 @node Argument Lists, Time of Evaluation, Program Structure, Program Structure | |
352 @section Argument Lists | |
353 | |
354 @noindent | |
355 Emacs Lisp's notation for argument lists of functions is a subset of | |
356 the Common Lisp notation. As well as the familiar @code{&optional} | |
357 and @code{&rest} markers, Common Lisp allows you to specify default | |
358 values for optional arguments, and it provides the additional markers | |
359 @code{&key} and @code{&aux}. | |
360 | |
361 Since argument parsing is built-in to Emacs, there is no way for | |
362 this package to implement Common Lisp argument lists seamlessly. | |
363 Instead, this package defines alternates for several Lisp forms | |
364 which you must use if you need Common Lisp argument lists. | |
365 | |
366 @defspec defun* name arglist body... | |
367 This form is identical to the regular @code{defun} form, except | |
368 that @var{arglist} is allowed to be a full Common Lisp argument | |
369 list. Also, the function body is enclosed in an implicit block | |
370 called @var{name}; @pxref{Blocks and Exits}. | |
371 @end defspec | |
372 | |
373 @defspec defsubst* name arglist body... | |
374 This is just like @code{defun*}, except that the function that | |
375 is defined is automatically proclaimed @code{inline}, i.e., | |
376 calls to it may be expanded into in-line code by the byte compiler. | |
31572
384603523f41
Remove obsolescences, especially hash tables and refs to Emacs 19.
Dave Love <fx@gnu.org>
parents:
30009
diff
changeset
|
377 This is analogous to the @code{defsubst} form; |
25829 | 378 @code{defsubst*} uses a different method (compiler macros) which |
379 works in all version of Emacs, and also generates somewhat more | |
380 efficient inline expansions. In particular, @code{defsubst*} | |
381 arranges for the processing of keyword arguments, default values, | |
382 etc., to be done at compile-time whenever possible. | |
383 @end defspec | |
384 | |
385 @defspec defmacro* name arglist body... | |
386 This is identical to the regular @code{defmacro} form, | |
387 except that @var{arglist} is allowed to be a full Common Lisp | |
388 argument list. The @code{&environment} keyword is supported as | |
389 described in Steele. The @code{&whole} keyword is supported only | |
390 within destructured lists (see below); top-level @code{&whole} | |
391 cannot be implemented with the current Emacs Lisp interpreter. | |
392 The macro expander body is enclosed in an implicit block called | |
393 @var{name}. | |
394 @end defspec | |
395 | |
396 @defspec function* symbol-or-lambda | |
397 This is identical to the regular @code{function} form, | |
398 except that if the argument is a @code{lambda} form then that | |
399 form may use a full Common Lisp argument list. | |
400 @end defspec | |
401 | |
402 Also, all forms (such as @code{defsetf} and @code{flet}) defined | |
403 in this package that include @var{arglist}s in their syntax allow | |
404 full Common Lisp argument lists. | |
405 | |
406 Note that it is @emph{not} necessary to use @code{defun*} in | |
407 order to have access to most @dfn{CL} features in your function. | |
408 These features are always present; @code{defun*}'s only | |
409 difference from @code{defun} is its more flexible argument | |
410 lists and its implicit block. | |
411 | |
412 The full form of a Common Lisp argument list is | |
413 | |
414 @example | |
415 (@var{var}... | |
416 &optional (@var{var} @var{initform} @var{svar})... | |
417 &rest @var{var} | |
418 &key ((@var{keyword} @var{var}) @var{initform} @var{svar})... | |
419 &aux (@var{var} @var{initform})...) | |
420 @end example | |
421 | |
422 Each of the five argument list sections is optional. The @var{svar}, | |
423 @var{initform}, and @var{keyword} parts are optional; if they are | |
424 omitted, then @samp{(@var{var})} may be written simply @samp{@var{var}}. | |
425 | |
426 The first section consists of zero or more @dfn{required} arguments. | |
427 These arguments must always be specified in a call to the function; | |
428 there is no difference between Emacs Lisp and Common Lisp as far as | |
429 required arguments are concerned. | |
430 | |
431 The second section consists of @dfn{optional} arguments. These | |
432 arguments may be specified in the function call; if they are not, | |
433 @var{initform} specifies the default value used for the argument. | |
434 (No @var{initform} means to use @code{nil} as the default.) The | |
435 @var{initform} is evaluated with the bindings for the preceding | |
436 arguments already established; @code{(a &optional (b (1+ a)))} | |
437 matches one or two arguments, with the second argument defaulting | |
438 to one plus the first argument. If the @var{svar} is specified, | |
439 it is an auxiliary variable which is bound to @code{t} if the optional | |
440 argument was specified, or to @code{nil} if the argument was omitted. | |
441 If you don't use an @var{svar}, then there will be no way for your | |
442 function to tell whether it was called with no argument, or with | |
443 the default value passed explicitly as an argument. | |
444 | |
445 The third section consists of a single @dfn{rest} argument. If | |
446 more arguments were passed to the function than are accounted for | |
447 by the required and optional arguments, those extra arguments are | |
448 collected into a list and bound to the ``rest'' argument variable. | |
449 Common Lisp's @code{&rest} is equivalent to that of Emacs Lisp. | |
450 Common Lisp accepts @code{&body} as a synonym for @code{&rest} in | |
451 macro contexts; this package accepts it all the time. | |
452 | |
453 The fourth section consists of @dfn{keyword} arguments. These | |
454 are optional arguments which are specified by name rather than | |
455 positionally in the argument list. For example, | |
456 | |
457 @example | |
458 (defun* foo (a &optional b &key c d (e 17))) | |
459 @end example | |
460 | |
461 @noindent | |
462 defines a function which may be called with one, two, or more | |
463 arguments. The first two arguments are bound to @code{a} and | |
464 @code{b} in the usual way. The remaining arguments must be | |
465 pairs of the form @code{:c}, @code{:d}, or @code{:e} followed | |
466 by the value to be bound to the corresponding argument variable. | |
467 (Symbols whose names begin with a colon are called @dfn{keywords}, | |
468 and they are self-quoting in the same way as @code{nil} and | |
469 @code{t}.) | |
470 | |
471 For example, the call @code{(foo 1 2 :d 3 :c 4)} sets the five | |
472 arguments to 1, 2, 4, 3, and 17, respectively. If the same keyword | |
473 appears more than once in the function call, the first occurrence | |
474 takes precedence over the later ones. Note that it is not possible | |
475 to specify keyword arguments without specifying the optional | |
476 argument @code{b} as well, since @code{(foo 1 :c 2)} would bind | |
477 @code{b} to the keyword @code{:c}, then signal an error because | |
478 @code{2} is not a valid keyword. | |
479 | |
480 If a @var{keyword} symbol is explicitly specified in the argument | |
481 list as shown in the above diagram, then that keyword will be | |
482 used instead of just the variable name prefixed with a colon. | |
483 You can specify a @var{keyword} symbol which does not begin with | |
484 a colon at all, but such symbols will not be self-quoting; you | |
485 will have to quote them explicitly with an apostrophe in the | |
486 function call. | |
487 | |
488 Ordinarily it is an error to pass an unrecognized keyword to | |
489 a function, e.g., @code{(foo 1 2 :c 3 :goober 4)}. You can ask | |
490 Lisp to ignore unrecognized keywords, either by adding the | |
491 marker @code{&allow-other-keys} after the keyword section | |
492 of the argument list, or by specifying an @code{:allow-other-keys} | |
493 argument in the call whose value is non-@code{nil}. If the | |
494 function uses both @code{&rest} and @code{&key} at the same time, | |
495 the ``rest'' argument is bound to the keyword list as it appears | |
496 in the call. For example: | |
497 | |
498 @smallexample | |
499 (defun* find-thing (thing &rest rest &key need &allow-other-keys) | |
500 (or (apply 'member* thing thing-list :allow-other-keys t rest) | |
501 (if need (error "Thing not found")))) | |
502 @end smallexample | |
503 | |
504 @noindent | |
505 This function takes a @code{:need} keyword argument, but also | |
506 accepts other keyword arguments which are passed on to the | |
507 @code{member*} function. @code{allow-other-keys} is used to | |
508 keep both @code{find-thing} and @code{member*} from complaining | |
509 about each others' keywords in the arguments. | |
510 | |
511 The fifth section of the argument list consists of @dfn{auxiliary | |
512 variables}. These are not really arguments at all, but simply | |
513 variables which are bound to @code{nil} or to the specified | |
514 @var{initforms} during execution of the function. There is no | |
515 difference between the following two functions, except for a | |
516 matter of stylistic taste: | |
517 | |
518 @example | |
519 (defun* foo (a b &aux (c (+ a b)) d) | |
520 @var{body}) | |
521 | |
522 (defun* foo (a b) | |
523 (let ((c (+ a b)) d) | |
524 @var{body})) | |
525 @end example | |
526 | |
527 Argument lists support @dfn{destructuring}. In Common Lisp, | |
528 destructuring is only allowed with @code{defmacro}; this package | |
529 allows it with @code{defun*} and other argument lists as well. | |
530 In destructuring, any argument variable (@var{var} in the above | |
531 diagram) can be replaced by a list of variables, or more generally, | |
532 a recursive argument list. The corresponding argument value must | |
533 be a list whose elements match this recursive argument list. | |
534 For example: | |
535 | |
536 @example | |
537 (defmacro* dolist ((var listform &optional resultform) | |
538 &rest body) | |
539 ...) | |
540 @end example | |
541 | |
542 This says that the first argument of @code{dolist} must be a list | |
543 of two or three items; if there are other arguments as well as this | |
544 list, they are stored in @code{body}. All features allowed in | |
545 regular argument lists are allowed in these recursive argument lists. | |
546 In addition, the clause @samp{&whole @var{var}} is allowed at the | |
547 front of a recursive argument list. It binds @var{var} to the | |
548 whole list being matched; thus @code{(&whole all a b)} matches | |
549 a list of two things, with @code{a} bound to the first thing, | |
550 @code{b} bound to the second thing, and @code{all} bound to the | |
551 list itself. (Common Lisp allows @code{&whole} in top-level | |
552 @code{defmacro} argument lists as well, but Emacs Lisp does not | |
553 support this usage.) | |
554 | |
555 One last feature of destructuring is that the argument list may be | |
556 dotted, so that the argument list @code{(a b . c)} is functionally | |
557 equivalent to @code{(a b &rest c)}. | |
558 | |
559 If the optimization quality @code{safety} is set to 0 | |
560 (@pxref{Declarations}), error checking for wrong number of | |
561 arguments and invalid keyword arguments is disabled. By default, | |
562 argument lists are rigorously checked. | |
563 | |
31572
384603523f41
Remove obsolescences, especially hash tables and refs to Emacs 19.
Dave Love <fx@gnu.org>
parents:
30009
diff
changeset
|
564 @node Time of Evaluation, , Argument Lists, Program Structure |
25829 | 565 @section Time of Evaluation |
566 | |
567 @noindent | |
568 Normally, the byte-compiler does not actually execute the forms in | |
569 a file it compiles. For example, if a file contains @code{(setq foo t)}, | |
570 the act of compiling it will not actually set @code{foo} to @code{t}. | |
571 This is true even if the @code{setq} was a top-level form (i.e., not | |
572 enclosed in a @code{defun} or other form). Sometimes, though, you | |
573 would like to have certain top-level forms evaluated at compile-time. | |
574 For example, the compiler effectively evaluates @code{defmacro} forms | |
575 at compile-time so that later parts of the file can refer to the | |
576 macros that are defined. | |
577 | |
578 @defspec eval-when (situations...) forms... | |
579 This form controls when the body @var{forms} are evaluated. | |
580 The @var{situations} list may contain any set of the symbols | |
581 @code{compile}, @code{load}, and @code{eval} (or their long-winded | |
582 ANSI equivalents, @code{:compile-toplevel}, @code{:load-toplevel}, | |
583 and @code{:execute}). | |
584 | |
585 The @code{eval-when} form is handled differently depending on | |
586 whether or not it is being compiled as a top-level form. | |
587 Specifically, it gets special treatment if it is being compiled | |
588 by a command such as @code{byte-compile-file} which compiles files | |
589 or buffers of code, and it appears either literally at the | |
590 top level of the file or inside a top-level @code{progn}. | |
591 | |
592 For compiled top-level @code{eval-when}s, the body @var{forms} are | |
593 executed at compile-time if @code{compile} is in the @var{situations} | |
594 list, and the @var{forms} are written out to the file (to be executed | |
595 at load-time) if @code{load} is in the @var{situations} list. | |
596 | |
597 For non-compiled-top-level forms, only the @code{eval} situation is | |
598 relevant. (This includes forms executed by the interpreter, forms | |
599 compiled with @code{byte-compile} rather than @code{byte-compile-file}, | |
600 and non-top-level forms.) The @code{eval-when} acts like a | |
601 @code{progn} if @code{eval} is specified, and like @code{nil} | |
602 (ignoring the body @var{forms}) if not. | |
603 | |
604 The rules become more subtle when @code{eval-when}s are nested; | |
605 consult Steele (second edition) for the gruesome details (and | |
606 some gruesome examples). | |
607 | |
608 Some simple examples: | |
609 | |
610 @example | |
611 ;; Top-level forms in foo.el: | |
612 (eval-when (compile) (setq foo1 'bar)) | |
613 (eval-when (load) (setq foo2 'bar)) | |
614 (eval-when (compile load) (setq foo3 'bar)) | |
615 (eval-when (eval) (setq foo4 'bar)) | |
616 (eval-when (eval compile) (setq foo5 'bar)) | |
617 (eval-when (eval load) (setq foo6 'bar)) | |
618 (eval-when (eval compile load) (setq foo7 'bar)) | |
619 @end example | |
620 | |
621 When @file{foo.el} is compiled, these variables will be set during | |
622 the compilation itself: | |
623 | |
624 @example | |
625 foo1 foo3 foo5 foo7 ; `compile' | |
626 @end example | |
627 | |
628 When @file{foo.elc} is loaded, these variables will be set: | |
629 | |
630 @example | |
631 foo2 foo3 foo6 foo7 ; `load' | |
632 @end example | |
633 | |
634 And if @file{foo.el} is loaded uncompiled, these variables will | |
635 be set: | |
636 | |
637 @example | |
638 foo4 foo5 foo6 foo7 ; `eval' | |
639 @end example | |
640 | |
641 If these seven @code{eval-when}s had been, say, inside a @code{defun}, | |
642 then the first three would have been equivalent to @code{nil} and the | |
643 last four would have been equivalent to the corresponding @code{setq}s. | |
644 | |
645 Note that @code{(eval-when (load eval) @dots{})} is equivalent | |
646 to @code{(progn @dots{})} in all contexts. The compiler treats | |
647 certain top-level forms, like @code{defmacro} (sort-of) and | |
648 @code{require}, as if they were wrapped in @code{(eval-when | |
649 (compile load eval) @dots{})}. | |
650 @end defspec | |
651 | |
31572
384603523f41
Remove obsolescences, especially hash tables and refs to Emacs 19.
Dave Love <fx@gnu.org>
parents:
30009
diff
changeset
|
652 Emacs includes two special forms related to @code{eval-when}. |
25829 | 653 One of these, @code{eval-when-compile}, is not quite equivalent to |
27225
ee242fddb1ab
Remove various old stuff and references thereto.
Dave Love <fx@gnu.org>
parents:
25829
diff
changeset
|
654 any @code{eval-when} construct and is described below. |
25829 | 655 |
656 The other form, @code{(eval-and-compile @dots{})}, is exactly | |
657 equivalent to @samp{(eval-when (compile load eval) @dots{})} and | |
658 so is not itself defined by this package. | |
659 | |
660 @defspec eval-when-compile forms... | |
661 The @var{forms} are evaluated at compile-time; at execution time, | |
662 this form acts like a quoted constant of the resulting value. Used | |
663 at top-level, @code{eval-when-compile} is just like @samp{eval-when | |
664 (compile eval)}. In other contexts, @code{eval-when-compile} | |
665 allows code to be evaluated once at compile-time for efficiency | |
666 or other reasons. | |
667 | |
668 This form is similar to the @samp{#.} syntax of true Common Lisp. | |
669 @end defspec | |
670 | |
671 @defspec load-time-value form | |
672 The @var{form} is evaluated at load-time; at execution time, | |
673 this form acts like a quoted constant of the resulting value. | |
674 | |
675 Early Common Lisp had a @samp{#,} syntax that was similar to | |
676 this, but ANSI Common Lisp replaced it with @code{load-time-value} | |
677 and gave it more well-defined semantics. | |
678 | |
679 In a compiled file, @code{load-time-value} arranges for @var{form} | |
680 to be evaluated when the @file{.elc} file is loaded and then used | |
681 as if it were a quoted constant. In code compiled by | |
682 @code{byte-compile} rather than @code{byte-compile-file}, the | |
683 effect is identical to @code{eval-when-compile}. In uncompiled | |
684 code, both @code{eval-when-compile} and @code{load-time-value} | |
685 act exactly like @code{progn}. | |
686 | |
687 @example | |
688 (defun report () | |
689 (insert "This function was executed on: " | |
690 (current-time-string) | |
691 ", compiled on: " | |
692 (eval-when-compile (current-time-string)) | |
693 ;; or '#.(current-time-string) in real Common Lisp | |
694 ", and loaded on: " | |
695 (load-time-value (current-time-string)))) | |
696 @end example | |
697 | |
698 @noindent | |
699 Byte-compiled, the above defun will result in the following code | |
700 (or its compiled equivalent, of course) in the @file{.elc} file: | |
701 | |
702 @example | |
703 (setq --temp-- (current-time-string)) | |
704 (defun report () | |
705 (insert "This function was executed on: " | |
706 (current-time-string) | |
707 ", compiled on: " | |
708 '"Wed Jun 23 18:33:43 1993" | |
709 ", and loaded on: " | |
710 --temp--)) | |
711 @end example | |
712 @end defspec | |
713 | |
714 @node Predicates, Control Structure, Program Structure, Top | |
715 @chapter Predicates | |
716 | |
717 @noindent | |
718 This section describes functions for testing whether various | |
719 facts are true or false. | |
720 | |
721 @menu | |
722 * Type Predicates:: `typep', `deftype', and `coerce' | |
723 * Equality Predicates:: `eql' and `equalp' | |
724 @end menu | |
725 | |
726 @node Type Predicates, Equality Predicates, Predicates, Predicates | |
727 @section Type Predicates | |
728 | |
729 @noindent | |
730 The @dfn{CL} package defines a version of the Common Lisp @code{typep} | |
731 predicate. | |
732 | |
733 @defun typep object type | |
734 Check if @var{object} is of type @var{type}, where @var{type} is a | |
735 (quoted) type name of the sort used by Common Lisp. For example, | |
736 @code{(typep foo 'integer)} is equivalent to @code{(integerp foo)}. | |
737 @end defun | |
738 | |
739 The @var{type} argument to the above function is either a symbol | |
740 or a list beginning with a symbol. | |
741 | |
742 @itemize @bullet | |
743 @item | |
744 If the type name is a symbol, Emacs appends @samp{-p} to the | |
745 symbol name to form the name of a predicate function for testing | |
746 the type. (Built-in predicates whose names end in @samp{p} rather | |
747 than @samp{-p} are used when appropriate.) | |
748 | |
749 @item | |
750 The type symbol @code{t} stands for the union of all types. | |
751 @code{(typep @var{object} t)} is always true. Likewise, the | |
752 type symbol @code{nil} stands for nothing at all, and | |
753 @code{(typep @var{object} nil)} is always false. | |
754 | |
755 @item | |
756 The type symbol @code{null} represents the symbol @code{nil}. | |
757 Thus @code{(typep @var{object} 'null)} is equivalent to | |
758 @code{(null @var{object})}. | |
759 | |
760 @item | |
761 The type symbol @code{real} is a synonym for @code{number}, and | |
762 @code{fixnum} is a synonym for @code{integer}. | |
763 | |
764 @item | |
765 The type symbols @code{character} and @code{string-char} match | |
766 integers in the range from 0 to 255. | |
767 | |
768 @item | |
769 The type symbol @code{float} uses the @code{floatp-safe} predicate | |
770 defined by this package rather than @code{floatp}, so it will work | |
771 correctly even in Emacs versions without floating-point support. | |
772 | |
773 @item | |
774 The type list @code{(integer @var{low} @var{high})} represents all | |
775 integers between @var{low} and @var{high}, inclusive. Either bound | |
776 may be a list of a single integer to specify an exclusive limit, | |
777 or a @code{*} to specify no limit. The type @code{(integer * *)} | |
778 is thus equivalent to @code{integer}. | |
779 | |
780 @item | |
781 Likewise, lists beginning with @code{float}, @code{real}, or | |
782 @code{number} represent numbers of that type falling in a particular | |
783 range. | |
784 | |
785 @item | |
786 Lists beginning with @code{and}, @code{or}, and @code{not} form | |
787 combinations of types. For example, @code{(or integer (float 0 *))} | |
788 represents all objects that are integers or non-negative floats. | |
789 | |
790 @item | |
791 Lists beginning with @code{member} or @code{member*} represent | |
792 objects @code{eql} to any of the following values. For example, | |
793 @code{(member 1 2 3 4)} is equivalent to @code{(integer 1 4)}, | |
794 and @code{(member nil)} is equivalent to @code{null}. | |
795 | |
796 @item | |
797 Lists of the form @code{(satisfies @var{predicate})} represent | |
798 all objects for which @var{predicate} returns true when called | |
799 with that object as an argument. | |
800 @end itemize | |
801 | |
802 The following function and macro (not technically predicates) are | |
803 related to @code{typep}. | |
804 | |
805 @defun coerce object type | |
806 This function attempts to convert @var{object} to the specified | |
807 @var{type}. If @var{object} is already of that type as determined by | |
808 @code{typep}, it is simply returned. Otherwise, certain types of | |
809 conversions will be made: If @var{type} is any sequence type | |
810 (@code{string}, @code{list}, etc.) then @var{object} will be | |
811 converted to that type if possible. If @var{type} is | |
812 @code{character}, then strings of length one and symbols with | |
813 one-character names can be coerced. If @var{type} is @code{float}, | |
814 then integers can be coerced in versions of Emacs that support | |
815 floats. In all other circumstances, @code{coerce} signals an | |
816 error. | |
817 @end defun | |
818 | |
819 @defspec deftype name arglist forms... | |
820 This macro defines a new type called @var{name}. It is similar | |
821 to @code{defmacro} in many ways; when @var{name} is encountered | |
822 as a type name, the body @var{forms} are evaluated and should | |
823 return a type specifier that is equivalent to the type. The | |
824 @var{arglist} is a Common Lisp argument list of the sort accepted | |
825 by @code{defmacro*}. The type specifier @samp{(@var{name} @var{args}...)} | |
826 is expanded by calling the expander with those arguments; the type | |
827 symbol @samp{@var{name}} is expanded by calling the expander with | |
828 no arguments. The @var{arglist} is processed the same as for | |
829 @code{defmacro*} except that optional arguments without explicit | |
830 defaults use @code{*} instead of @code{nil} as the ``default'' | |
831 default. Some examples: | |
832 | |
833 @example | |
834 (deftype null () '(satisfies null)) ; predefined | |
835 (deftype list () '(or null cons)) ; predefined | |
836 (deftype unsigned-byte (&optional bits) | |
837 (list 'integer 0 (if (eq bits '*) bits (1- (lsh 1 bits))))) | |
838 (unsigned-byte 8) @equiv{} (integer 0 255) | |
839 (unsigned-byte) @equiv{} (integer 0 *) | |
840 unsigned-byte @equiv{} (integer 0 *) | |
841 @end example | |
842 | |
843 @noindent | |
844 The last example shows how the Common Lisp @code{unsigned-byte} | |
845 type specifier could be implemented if desired; this package does | |
846 not implement @code{unsigned-byte} by default. | |
847 @end defspec | |
848 | |
849 The @code{typecase} and @code{check-type} macros also use type | |
850 names. @xref{Conditionals}. @xref{Assertions}. The @code{map}, | |
851 @code{concatenate}, and @code{merge} functions take type-name | |
852 arguments to specify the type of sequence to return. @xref{Sequences}. | |
853 | |
31572
384603523f41
Remove obsolescences, especially hash tables and refs to Emacs 19.
Dave Love <fx@gnu.org>
parents:
30009
diff
changeset
|
854 @node Equality Predicates, , Type Predicates, Predicates |
25829 | 855 @section Equality Predicates |
856 | |
857 @noindent | |
858 This package defines two Common Lisp predicates, @code{eql} and | |
859 @code{equalp}. | |
860 | |
861 @defun eql a b | |
862 This function is almost the same as @code{eq}, except that if @var{a} | |
863 and @var{b} are numbers of the same type, it compares them for numeric | |
864 equality (as if by @code{equal} instead of @code{eq}). This makes a | |
865 difference only for versions of Emacs that are compiled with | |
31572
384603523f41
Remove obsolescences, especially hash tables and refs to Emacs 19.
Dave Love <fx@gnu.org>
parents:
30009
diff
changeset
|
866 floating-point support. Emacs floats are allocated |
25829 | 867 objects just like cons cells, which means that @code{(eq 3.0 3.0)} |
868 will not necessarily be true---if the two @code{3.0}s were allocated | |
869 separately, the pointers will be different even though the numbers are | |
870 the same. But @code{(eql 3.0 3.0)} will always be true. | |
871 | |
872 The types of the arguments must match, so @code{(eql 3 3.0)} is | |
873 still false. | |
874 | |
875 Note that Emacs integers are ``direct'' rather than allocated, which | |
876 basically means @code{(eq 3 3)} will always be true. Thus @code{eq} | |
877 and @code{eql} behave differently only if floating-point numbers are | |
878 involved, and are indistinguishable on Emacs versions that don't | |
879 support floats. | |
880 | |
881 There is a slight inconsistency with Common Lisp in the treatment of | |
882 positive and negative zeros. Some machines, notably those with IEEE | |
883 standard arithmetic, represent @code{+0} and @code{-0} as distinct | |
884 values. Normally this doesn't matter because the standard specifies | |
885 that @code{(= 0.0 -0.0)} should always be true, and this is indeed | |
886 what Emacs Lisp and Common Lisp do. But the Common Lisp standard | |
887 states that @code{(eql 0.0 -0.0)} and @code{(equal 0.0 -0.0)} should | |
888 be false on IEEE-like machines; Emacs Lisp does not do this, and in | |
889 fact the only known way to distinguish between the two zeros in Emacs | |
890 Lisp is to @code{format} them and check for a minus sign. | |
891 @end defun | |
892 | |
893 @defun equalp a b | |
894 This function is a more flexible version of @code{equal}. In | |
895 particular, it compares strings case-insensitively, and it compares | |
896 numbers without regard to type (so that @code{(equalp 3 3.0)} is | |
897 true). Vectors and conses are compared recursively. All other | |
898 objects are compared as if by @code{equal}. | |
899 | |
900 This function differs from Common Lisp @code{equalp} in several | |
901 respects. First, Common Lisp's @code{equalp} also compares | |
902 @emph{characters} case-insensitively, which would be impractical | |
903 in this package since Emacs does not distinguish between integers | |
904 and characters. In keeping with the idea that strings are less | |
905 vector-like in Emacs Lisp, this package's @code{equalp} also will | |
31572
384603523f41
Remove obsolescences, especially hash tables and refs to Emacs 19.
Dave Love <fx@gnu.org>
parents:
30009
diff
changeset
|
906 not compare strings against vectors of integers. |
25829 | 907 @end defun |
908 | |
909 Also note that the Common Lisp functions @code{member} and @code{assoc} | |
910 use @code{eql} to compare elements, whereas Emacs Lisp follows the | |
911 MacLisp tradition and uses @code{equal} for these two functions. | |
912 In Emacs, use @code{member*} and @code{assoc*} to get functions | |
913 which use @code{eql} for comparisons. | |
914 | |
915 @node Control Structure, Macros, Predicates, Top | |
916 @chapter Control Structure | |
917 | |
918 @noindent | |
919 The features described in the following sections implement | |
920 various advanced control structures, including the powerful | |
921 @code{setf} facility and a number of looping and conditional | |
922 constructs. | |
923 | |
924 @menu | |
925 * Assignment:: The `psetq' form | |
926 * Generalized Variables:: `setf', `incf', `push', etc. | |
927 * Variable Bindings:: `progv', `lexical-let', `flet', `macrolet' | |
928 * Conditionals:: `case', `typecase' | |
929 * Blocks and Exits:: `block', `return', `return-from' | |
27511
ead47a92d9d0
Replace dotimes, dolist, with comment.
Dave Love <fx@gnu.org>
parents:
27487
diff
changeset
|
930 * Iteration:: `do', `dotimes', `dolist', `do-symbols' |
25829 | 931 * Loop Facility:: The Common Lisp `loop' macro |
932 * Multiple Values:: `values', `multiple-value-bind', etc. | |
933 @end menu | |
934 | |
935 @node Assignment, Generalized Variables, Control Structure, Control Structure | |
936 @section Assignment | |
937 | |
938 @noindent | |
939 The @code{psetq} form is just like @code{setq}, except that multiple | |
940 assignments are done in parallel rather than sequentially. | |
941 | |
942 @defspec psetq [symbol form]@dots{} | |
943 This special form (actually a macro) is used to assign to several | |
944 variables simultaneously. Given only one @var{symbol} and @var{form}, | |
945 it has the same effect as @code{setq}. Given several @var{symbol} | |
946 and @var{form} pairs, it evaluates all the @var{form}s in advance | |
947 and then stores the corresponding variables afterwards. | |
948 | |
949 @example | |
950 (setq x 2 y 3) | |
951 (setq x (+ x y) y (* x y)) | |
952 x | |
953 @result{} 5 | |
954 y ; @r{@code{y} was computed after @code{x} was set.} | |
955 @result{} 15 | |
956 (setq x 2 y 3) | |
957 (psetq x (+ x y) y (* x y)) | |
958 x | |
959 @result{} 5 | |
960 y ; @r{@code{y} was computed before @code{x} was set.} | |
961 @result{} 6 | |
962 @end example | |
963 | |
964 The simplest use of @code{psetq} is @code{(psetq x y y x)}, which | |
965 exchanges the values of two variables. (The @code{rotatef} form | |
966 provides an even more convenient way to swap two variables; | |
967 @pxref{Modify Macros}.) | |
968 | |
969 @code{psetq} always returns @code{nil}. | |
970 @end defspec | |
971 | |
972 @node Generalized Variables, Variable Bindings, Assignment, Control Structure | |
973 @section Generalized Variables | |
974 | |
975 @noindent | |
976 A ``generalized variable'' or ``place form'' is one of the many places | |
977 in Lisp memory where values can be stored. The simplest place form is | |
978 a regular Lisp variable. But the cars and cdrs of lists, elements | |
979 of arrays, properties of symbols, and many other locations are also | |
980 places where Lisp values are stored. | |
981 | |
982 The @code{setf} form is like @code{setq}, except that it accepts | |
983 arbitrary place forms on the left side rather than just | |
984 symbols. For example, @code{(setf (car a) b)} sets the car of | |
985 @code{a} to @code{b}, doing the same operation as @code{(setcar a b)} | |
986 but without having to remember two separate functions for setting | |
987 and accessing every type of place. | |
988 | |
989 Generalized variables are analogous to ``lvalues'' in the C | |
990 language, where @samp{x = a[i]} gets an element from an array | |
991 and @samp{a[i] = x} stores an element using the same notation. | |
992 Just as certain forms like @code{a[i]} can be lvalues in C, there | |
993 is a set of forms that can be generalized variables in Lisp. | |
994 | |
995 @menu | |
996 * Basic Setf:: `setf' and place forms | |
997 * Modify Macros:: `incf', `push', `rotatef', `letf', `callf', etc. | |
998 * Customizing Setf:: `define-modify-macro', `defsetf', `define-setf-method' | |
999 @end menu | |
1000 | |
1001 @node Basic Setf, Modify Macros, Generalized Variables, Generalized Variables | |
1002 @subsection Basic Setf | |
1003 | |
1004 @noindent | |
1005 The @code{setf} macro is the most basic way to operate on generalized | |
1006 variables. | |
1007 | |
1008 @defspec setf [place form]@dots{} | |
1009 This macro evaluates @var{form} and stores it in @var{place}, which | |
1010 must be a valid generalized variable form. If there are several | |
1011 @var{place} and @var{form} pairs, the assignments are done sequentially | |
1012 just as with @code{setq}. @code{setf} returns the value of the last | |
1013 @var{form}. | |
1014 | |
1015 The following Lisp forms will work as generalized variables, and | |
1016 so may legally appear in the @var{place} argument of @code{setf}: | |
1017 | |
1018 @itemize @bullet | |
1019 @item | |
1020 A symbol naming a variable. In other words, @code{(setf x y)} is | |
1021 exactly equivalent to @code{(setq x y)}, and @code{setq} itself is | |
1022 strictly speaking redundant now that @code{setf} exists. Many | |
1023 programmers continue to prefer @code{setq} for setting simple | |
1024 variables, though, purely for stylistic or historical reasons. | |
1025 The macro @code{(setf x y)} actually expands to @code{(setq x y)}, | |
1026 so there is no performance penalty for using it in compiled code. | |
1027 | |
1028 @item | |
1029 A call to any of the following Lisp functions: | |
1030 | |
1031 @smallexample | |
1032 car cdr caar .. cddddr | |
1033 nth rest first .. tenth | |
1034 aref elt nthcdr | |
1035 symbol-function symbol-value symbol-plist | |
1036 get get* getf | |
1037 gethash subseq | |
1038 @end smallexample | |
1039 | |
1040 @noindent | |
1041 Note that for @code{nthcdr} and @code{getf}, the list argument | |
1042 of the function must itself be a valid @var{place} form. For | |
1043 example, @code{(setf (nthcdr 0 foo) 7)} will set @code{foo} itself | |
1044 to 7. Note that @code{push} and @code{pop} on an @code{nthcdr} | |
1045 place can be used to insert or delete at any position in a list. | |
1046 The use of @code{nthcdr} as a @var{place} form is an extension | |
1047 to standard Common Lisp. | |
1048 | |
1049 @item | |
1050 The following Emacs-specific functions are also @code{setf}-able. | |
1051 | |
1052 @smallexample | |
1053 buffer-file-name marker-position | |
1054 buffer-modified-p match-data | |
1055 buffer-name mouse-position | |
1056 buffer-string overlay-end | |
1057 buffer-substring overlay-get | |
1058 current-buffer overlay-start | |
1059 current-case-table point | |
1060 current-column point-marker | |
1061 current-global-map point-max | |
1062 current-input-mode point-min | |
1063 current-local-map process-buffer | |
1064 current-window-configuration process-filter | |
1065 default-file-modes process-sentinel | |
1066 default-value read-mouse-position | |
1067 documentation-property screen-height | |
1068 extent-data screen-menubar | |
1069 extent-end-position screen-width | |
1070 extent-start-position selected-window | |
1071 face-background selected-screen | |
1072 face-background-pixmap selected-frame | |
1073 face-font standard-case-table | |
1074 face-foreground syntax-table | |
1075 face-underline-p window-buffer | |
1076 file-modes window-dedicated-p | |
1077 frame-height window-display-table | |
1078 frame-parameters window-height | |
1079 frame-visible-p window-hscroll | |
1080 frame-width window-point | |
1081 get-register window-start | |
1082 getenv window-width | |
1083 global-key-binding x-get-cut-buffer | |
1084 keymap-parent x-get-cutbuffer | |
1085 local-key-binding x-get-secondary-selection | |
1086 mark x-get-selection | |
1087 mark-marker | |
1088 @end smallexample | |
1089 | |
1090 Most of these have directly corresponding ``set'' functions, like | |
1091 @code{use-local-map} for @code{current-local-map}, or @code{goto-char} | |
1092 for @code{point}. A few, like @code{point-min}, expand to longer | |
1093 sequences of code when they are @code{setf}'d (@code{(narrow-to-region | |
1094 x (point-max))} in this case). | |
1095 | |
1096 @item | |
1097 A call of the form @code{(substring @var{subplace} @var{n} [@var{m}])}, | |
1098 where @var{subplace} is itself a legal generalized variable whose | |
1099 current value is a string, and where the value stored is also a | |
1100 string. The new string is spliced into the specified part of the | |
1101 destination string. For example: | |
1102 | |
1103 @example | |
1104 (setq a (list "hello" "world")) | |
1105 @result{} ("hello" "world") | |
1106 (cadr a) | |
1107 @result{} "world" | |
1108 (substring (cadr a) 2 4) | |
1109 @result{} "rl" | |
1110 (setf (substring (cadr a) 2 4) "o") | |
1111 @result{} "o" | |
1112 (cadr a) | |
1113 @result{} "wood" | |
1114 a | |
1115 @result{} ("hello" "wood") | |
1116 @end example | |
1117 | |
1118 The generalized variable @code{buffer-substring}, listed above, | |
1119 also works in this way by replacing a portion of the current buffer. | |
1120 | |
1121 @item | |
1122 A call of the form @code{(apply '@var{func} @dots{})} or | |
1123 @code{(apply (function @var{func}) @dots{})}, where @var{func} | |
1124 is a @code{setf}-able function whose store function is ``suitable'' | |
1125 in the sense described in Steele's book; since none of the standard | |
1126 Emacs place functions are suitable in this sense, this feature is | |
1127 only interesting when used with places you define yourself with | |
1128 @code{define-setf-method} or the long form of @code{defsetf}. | |
1129 | |
1130 @item | |
1131 A macro call, in which case the macro is expanded and @code{setf} | |
1132 is applied to the resulting form. | |
1133 | |
1134 @item | |
1135 Any form for which a @code{defsetf} or @code{define-setf-method} | |
1136 has been made. | |
1137 @end itemize | |
1138 | |
1139 Using any forms other than these in the @var{place} argument to | |
1140 @code{setf} will signal an error. | |
1141 | |
1142 The @code{setf} macro takes care to evaluate all subforms in | |
1143 the proper left-to-right order; for example, | |
1144 | |
1145 @example | |
1146 (setf (aref vec (incf i)) i) | |
1147 @end example | |
1148 | |
1149 @noindent | |
1150 looks like it will evaluate @code{(incf i)} exactly once, before the | |
1151 following access to @code{i}; the @code{setf} expander will insert | |
1152 temporary variables as necessary to ensure that it does in fact work | |
1153 this way no matter what setf-method is defined for @code{aref}. | |
1154 (In this case, @code{aset} would be used and no such steps would | |
1155 be necessary since @code{aset} takes its arguments in a convenient | |
1156 order.) | |
1157 | |
1158 However, if the @var{place} form is a macro which explicitly | |
1159 evaluates its arguments in an unusual order, this unusual order | |
1160 will be preserved. Adapting an example from Steele, given | |
1161 | |
1162 @example | |
1163 (defmacro wrong-order (x y) (list 'aref y x)) | |
1164 @end example | |
1165 | |
1166 @noindent | |
1167 the form @code{(setf (wrong-order @var{a} @var{b}) 17)} will | |
1168 evaluate @var{b} first, then @var{a}, just as in an actual call | |
1169 to @code{wrong-order}. | |
1170 @end defspec | |
1171 | |
1172 @node Modify Macros, Customizing Setf, Basic Setf, Generalized Variables | |
1173 @subsection Modify Macros | |
1174 | |
1175 @noindent | |
1176 This package defines a number of other macros besides @code{setf} | |
1177 that operate on generalized variables. Many are interesting and | |
1178 useful even when the @var{place} is just a variable name. | |
1179 | |
1180 @defspec psetf [place form]@dots{} | |
1181 This macro is to @code{setf} what @code{psetq} is to @code{setq}: | |
1182 When several @var{place}s and @var{form}s are involved, the | |
1183 assignments take place in parallel rather than sequentially. | |
1184 Specifically, all subforms are evaluated from left to right, then | |
1185 all the assignments are done (in an undefined order). | |
1186 @end defspec | |
1187 | |
1188 @defspec incf place &optional x | |
1189 This macro increments the number stored in @var{place} by one, or | |
1190 by @var{x} if specified. The incremented value is returned. For | |
1191 example, @code{(incf i)} is equivalent to @code{(setq i (1+ i))}, and | |
1192 @code{(incf (car x) 2)} is equivalent to @code{(setcar x (+ (car x) 2))}. | |
1193 | |
1194 Once again, care is taken to preserve the ``apparent'' order of | |
1195 evaluation. For example, | |
1196 | |
1197 @example | |
1198 (incf (aref vec (incf i))) | |
1199 @end example | |
1200 | |
1201 @noindent | |
1202 appears to increment @code{i} once, then increment the element of | |
1203 @code{vec} addressed by @code{i}; this is indeed exactly what it | |
1204 does, which means the above form is @emph{not} equivalent to the | |
1205 ``obvious'' expansion, | |
1206 | |
1207 @example | |
1208 (setf (aref vec (incf i)) (1+ (aref vec (incf i)))) ; Wrong! | |
1209 @end example | |
1210 | |
1211 @noindent | |
1212 but rather to something more like | |
1213 | |
1214 @example | |
1215 (let ((temp (incf i))) | |
1216 (setf (aref vec temp) (1+ (aref vec temp)))) | |
1217 @end example | |
1218 | |
1219 @noindent | |
1220 Again, all of this is taken care of automatically by @code{incf} and | |
1221 the other generalized-variable macros. | |
1222 | |
1223 As a more Emacs-specific example of @code{incf}, the expression | |
1224 @code{(incf (point) @var{n})} is essentially equivalent to | |
1225 @code{(forward-char @var{n})}. | |
1226 @end defspec | |
1227 | |
1228 @defspec decf place &optional x | |
1229 This macro decrements the number stored in @var{place} by one, or | |
1230 by @var{x} if specified. | |
1231 @end defspec | |
1232 | |
1233 @defspec pop place | |
1234 This macro removes and returns the first element of the list stored | |
1235 in @var{place}. It is analogous to @code{(prog1 (car @var{place}) | |
1236 (setf @var{place} (cdr @var{place})))}, except that it takes care | |
1237 to evaluate all subforms only once. | |
1238 @end defspec | |
1239 | |
1240 @defspec push x place | |
1241 This macro inserts @var{x} at the front of the list stored in | |
1242 @var{place}. It is analogous to @code{(setf @var{place} (cons | |
1243 @var{x} @var{place}))}, except for evaluation of the subforms. | |
1244 @end defspec | |
1245 | |
1246 @defspec pushnew x place @t{&key :test :test-not :key} | |
1247 This macro inserts @var{x} at the front of the list stored in | |
1248 @var{place}, but only if @var{x} was not @code{eql} to any | |
1249 existing element of the list. The optional keyword arguments | |
1250 are interpreted in the same way as for @code{adjoin}. | |
1251 @xref{Lists as Sets}. | |
1252 @end defspec | |
1253 | |
1254 @defspec shiftf place@dots{} newvalue | |
1255 This macro shifts the @var{place}s left by one, shifting in the | |
1256 value of @var{newvalue} (which may be any Lisp expression, not just | |
1257 a generalized variable), and returning the value shifted out of | |
1258 the first @var{place}. Thus, @code{(shiftf @var{a} @var{b} @var{c} | |
1259 @var{d})} is equivalent to | |
1260 | |
1261 @example | |
1262 (prog1 | |
1263 @var{a} | |
1264 (psetf @var{a} @var{b} | |
1265 @var{b} @var{c} | |
1266 @var{c} @var{d})) | |
1267 @end example | |
1268 | |
1269 @noindent | |
1270 except that the subforms of @var{a}, @var{b}, and @var{c} are actually | |
1271 evaluated only once each and in the apparent order. | |
1272 @end defspec | |
1273 | |
1274 @defspec rotatef place@dots{} | |
1275 This macro rotates the @var{place}s left by one in circular fashion. | |
1276 Thus, @code{(rotatef @var{a} @var{b} @var{c} @var{d})} is equivalent to | |
1277 | |
1278 @example | |
1279 (psetf @var{a} @var{b} | |
1280 @var{b} @var{c} | |
1281 @var{c} @var{d} | |
1282 @var{d} @var{a}) | |
1283 @end example | |
1284 | |
1285 @noindent | |
1286 except for the evaluation of subforms. @code{rotatef} always | |
1287 returns @code{nil}. Note that @code{(rotatef @var{a} @var{b})} | |
1288 conveniently exchanges @var{a} and @var{b}. | |
1289 @end defspec | |
1290 | |
1291 The following macros were invented for this package; they have no | |
1292 analogues in Common Lisp. | |
1293 | |
1294 @defspec letf (bindings@dots{}) forms@dots{} | |
1295 This macro is analogous to @code{let}, but for generalized variables | |
1296 rather than just symbols. Each @var{binding} should be of the form | |
1297 @code{(@var{place} @var{value})}; the original contents of the | |
1298 @var{place}s are saved, the @var{value}s are stored in them, and | |
1299 then the body @var{form}s are executed. Afterwards, the @var{places} | |
1300 are set back to their original saved contents. This cleanup happens | |
1301 even if the @var{form}s exit irregularly due to a @code{throw} or an | |
1302 error. | |
1303 | |
1304 For example, | |
1305 | |
1306 @example | |
1307 (letf (((point) (point-min)) | |
1308 (a 17)) | |
1309 ...) | |
1310 @end example | |
1311 | |
1312 @noindent | |
1313 moves ``point'' in the current buffer to the beginning of the buffer, | |
1314 and also binds @code{a} to 17 (as if by a normal @code{let}, since | |
1315 @code{a} is just a regular variable). After the body exits, @code{a} | |
1316 is set back to its original value and point is moved back to its | |
1317 original position. | |
1318 | |
1319 Note that @code{letf} on @code{(point)} is not quite like a | |
1320 @code{save-excursion}, as the latter effectively saves a marker | |
1321 which tracks insertions and deletions in the buffer. Actually, | |
1322 a @code{letf} of @code{(point-marker)} is much closer to this | |
1323 behavior. (@code{point} and @code{point-marker} are equivalent | |
1324 as @code{setf} places; each will accept either an integer or a | |
1325 marker as the stored value.) | |
1326 | |
1327 Since generalized variables look like lists, @code{let}'s shorthand | |
1328 of using @samp{foo} for @samp{(foo nil)} as a @var{binding} would | |
1329 be ambiguous in @code{letf} and is not allowed. | |
1330 | |
1331 However, a @var{binding} specifier may be a one-element list | |
1332 @samp{(@var{place})}, which is similar to @samp{(@var{place} | |
1333 @var{place})}. In other words, the @var{place} is not disturbed | |
1334 on entry to the body, and the only effect of the @code{letf} is | |
1335 to restore the original value of @var{place} afterwards. (The | |
1336 redundant access-and-store suggested by the @code{(@var{place} | |
1337 @var{place})} example does not actually occur.) | |
1338 | |
1339 In most cases, the @var{place} must have a well-defined value on | |
1340 entry to the @code{letf} form. The only exceptions are plain | |
1341 variables and calls to @code{symbol-value} and @code{symbol-function}. | |
1342 If the symbol is not bound on entry, it is simply made unbound by | |
1343 @code{makunbound} or @code{fmakunbound} on exit. | |
1344 @end defspec | |
1345 | |
1346 @defspec letf* (bindings@dots{}) forms@dots{} | |
1347 This macro is to @code{letf} what @code{let*} is to @code{let}: | |
1348 It does the bindings in sequential rather than parallel order. | |
1349 @end defspec | |
1350 | |
1351 @defspec callf @var{function} @var{place} @var{args}@dots{} | |
1352 This is the ``generic'' modify macro. It calls @var{function}, | |
1353 which should be an unquoted function name, macro name, or lambda. | |
1354 It passes @var{place} and @var{args} as arguments, and assigns the | |
1355 result back to @var{place}. For example, @code{(incf @var{place} | |
1356 @var{n})} is the same as @code{(callf + @var{place} @var{n})}. | |
1357 Some more examples: | |
1358 | |
1359 @example | |
1360 (callf abs my-number) | |
1361 (callf concat (buffer-name) "<" (int-to-string n) ">") | |
1362 (callf union happy-people (list joe bob) :test 'same-person) | |
1363 @end example | |
1364 | |
1365 @xref{Customizing Setf}, for @code{define-modify-macro}, a way | |
1366 to create even more concise notations for modify macros. Note | |
1367 again that @code{callf} is an extension to standard Common Lisp. | |
1368 @end defspec | |
1369 | |
1370 @defspec callf2 @var{function} @var{arg1} @var{place} @var{args}@dots{} | |
1371 This macro is like @code{callf}, except that @var{place} is | |
1372 the @emph{second} argument of @var{function} rather than the | |
1373 first. For example, @code{(push @var{x} @var{place})} is | |
1374 equivalent to @code{(callf2 cons @var{x} @var{place})}. | |
1375 @end defspec | |
1376 | |
1377 The @code{callf} and @code{callf2} macros serve as building | |
1378 blocks for other macros like @code{incf}, @code{pushnew}, and | |
1379 @code{define-modify-macro}. The @code{letf} and @code{letf*} | |
1380 macros are used in the processing of symbol macros; | |
1381 @pxref{Macro Bindings}. | |
1382 | |
31572
384603523f41
Remove obsolescences, especially hash tables and refs to Emacs 19.
Dave Love <fx@gnu.org>
parents:
30009
diff
changeset
|
1383 @node Customizing Setf, , Modify Macros, Generalized Variables |
25829 | 1384 @subsection Customizing Setf |
1385 | |
1386 @noindent | |
1387 Common Lisp defines three macros, @code{define-modify-macro}, | |
1388 @code{defsetf}, and @code{define-setf-method}, that allow the | |
1389 user to extend generalized variables in various ways. | |
1390 | |
1391 @defspec define-modify-macro name arglist function [doc-string] | |
1392 This macro defines a ``read-modify-write'' macro similar to | |
1393 @code{incf} and @code{decf}. The macro @var{name} is defined | |
1394 to take a @var{place} argument followed by additional arguments | |
1395 described by @var{arglist}. The call | |
1396 | |
1397 @example | |
1398 (@var{name} @var{place} @var{args}...) | |
1399 @end example | |
1400 | |
1401 @noindent | |
1402 will be expanded to | |
1403 | |
1404 @example | |
1405 (callf @var{func} @var{place} @var{args}...) | |
1406 @end example | |
1407 | |
1408 @noindent | |
1409 which in turn is roughly equivalent to | |
1410 | |
1411 @example | |
1412 (setf @var{place} (@var{func} @var{place} @var{args}...)) | |
1413 @end example | |
1414 | |
1415 For example: | |
1416 | |
1417 @example | |
1418 (define-modify-macro incf (&optional (n 1)) +) | |
1419 (define-modify-macro concatf (&rest args) concat) | |
1420 @end example | |
1421 | |
1422 Note that @code{&key} is not allowed in @var{arglist}, but | |
1423 @code{&rest} is sufficient to pass keywords on to the function. | |
1424 | |
1425 Most of the modify macros defined by Common Lisp do not exactly | |
1426 follow the pattern of @code{define-modify-macro}. For example, | |
1427 @code{push} takes its arguments in the wrong order, and @code{pop} | |
1428 is completely irregular. You can define these macros ``by hand'' | |
1429 using @code{get-setf-method}, or consult the source file | |
1430 @file{cl-macs.el} to see how to use the internal @code{setf} | |
1431 building blocks. | |
1432 @end defspec | |
1433 | |
1434 @defspec defsetf access-fn update-fn | |
1435 This is the simpler of two @code{defsetf} forms. Where | |
1436 @var{access-fn} is the name of a function which accesses a place, | |
1437 this declares @var{update-fn} to be the corresponding store | |
1438 function. From now on, | |
1439 | |
1440 @example | |
1441 (setf (@var{access-fn} @var{arg1} @var{arg2} @var{arg3}) @var{value}) | |
1442 @end example | |
1443 | |
1444 @noindent | |
1445 will be expanded to | |
1446 | |
1447 @example | |
1448 (@var{update-fn} @var{arg1} @var{arg2} @var{arg3} @var{value}) | |
1449 @end example | |
1450 | |
1451 @noindent | |
1452 The @var{update-fn} is required to be either a true function, or | |
1453 a macro which evaluates its arguments in a function-like way. Also, | |
1454 the @var{update-fn} is expected to return @var{value} as its result. | |
1455 Otherwise, the above expansion would not obey the rules for the way | |
1456 @code{setf} is supposed to behave. | |
1457 | |
1458 As a special (non-Common-Lisp) extension, a third argument of @code{t} | |
1459 to @code{defsetf} says that the @code{update-fn}'s return value is | |
1460 not suitable, so that the above @code{setf} should be expanded to | |
1461 something more like | |
1462 | |
1463 @example | |
1464 (let ((temp @var{value})) | |
1465 (@var{update-fn} @var{arg1} @var{arg2} @var{arg3} temp) | |
1466 temp) | |
1467 @end example | |
1468 | |
1469 Some examples of the use of @code{defsetf}, drawn from the standard | |
1470 suite of setf methods, are: | |
1471 | |
1472 @example | |
1473 (defsetf car setcar) | |
1474 (defsetf symbol-value set) | |
1475 (defsetf buffer-name rename-buffer t) | |
1476 @end example | |
1477 @end defspec | |
1478 | |
1479 @defspec defsetf access-fn arglist (store-var) forms@dots{} | |
1480 This is the second, more complex, form of @code{defsetf}. It is | |
1481 rather like @code{defmacro} except for the additional @var{store-var} | |
1482 argument. The @var{forms} should return a Lisp form which stores | |
1483 the value of @var{store-var} into the generalized variable formed | |
1484 by a call to @var{access-fn} with arguments described by @var{arglist}. | |
1485 The @var{forms} may begin with a string which documents the @code{setf} | |
1486 method (analogous to the doc string that appears at the front of a | |
1487 function). | |
1488 | |
1489 For example, the simple form of @code{defsetf} is shorthand for | |
1490 | |
1491 @example | |
1492 (defsetf @var{access-fn} (&rest args) (store) | |
1493 (append '(@var{update-fn}) args (list store))) | |
1494 @end example | |
1495 | |
1496 The Lisp form that is returned can access the arguments from | |
1497 @var{arglist} and @var{store-var} in an unrestricted fashion; | |
1498 macros like @code{setf} and @code{incf} which invoke this | |
1499 setf-method will insert temporary variables as needed to make | |
1500 sure the apparent order of evaluation is preserved. | |
1501 | |
1502 Another example drawn from the standard package: | |
1503 | |
1504 @example | |
1505 (defsetf nth (n x) (store) | |
1506 (list 'setcar (list 'nthcdr n x) store)) | |
1507 @end example | |
1508 @end defspec | |
1509 | |
1510 @defspec define-setf-method access-fn arglist forms@dots{} | |
1511 This is the most general way to create new place forms. When | |
1512 a @code{setf} to @var{access-fn} with arguments described by | |
1513 @var{arglist} is expanded, the @var{forms} are evaluated and | |
1514 must return a list of five items: | |
1515 | |
1516 @enumerate | |
1517 @item | |
1518 A list of @dfn{temporary variables}. | |
1519 | |
1520 @item | |
1521 A list of @dfn{value forms} corresponding to the temporary variables | |
1522 above. The temporary variables will be bound to these value forms | |
1523 as the first step of any operation on the generalized variable. | |
1524 | |
1525 @item | |
1526 A list of exactly one @dfn{store variable} (generally obtained | |
1527 from a call to @code{gensym}). | |
1528 | |
1529 @item | |
1530 A Lisp form which stores the contents of the store variable into | |
1531 the generalized variable, assuming the temporaries have been | |
1532 bound as described above. | |
1533 | |
1534 @item | |
1535 A Lisp form which accesses the contents of the generalized variable, | |
1536 assuming the temporaries have been bound. | |
1537 @end enumerate | |
1538 | |
1539 This is exactly like the Common Lisp macro of the same name, | |
1540 except that the method returns a list of five values rather | |
1541 than the five values themselves, since Emacs Lisp does not | |
1542 support Common Lisp's notion of multiple return values. | |
1543 | |
1544 Once again, the @var{forms} may begin with a documentation string. | |
1545 | |
1546 A setf-method should be maximally conservative with regard to | |
1547 temporary variables. In the setf-methods generated by | |
1548 @code{defsetf}, the second return value is simply the list of | |
1549 arguments in the place form, and the first return value is a | |
1550 list of a corresponding number of temporary variables generated | |
1551 by @code{gensym}. Macros like @code{setf} and @code{incf} which | |
1552 use this setf-method will optimize away most temporaries that | |
1553 turn out to be unnecessary, so there is little reason for the | |
1554 setf-method itself to optimize. | |
1555 @end defspec | |
1556 | |
1557 @defun get-setf-method place &optional env | |
1558 This function returns the setf-method for @var{place}, by | |
1559 invoking the definition previously recorded by @code{defsetf} | |
1560 or @code{define-setf-method}. The result is a list of five | |
1561 values as described above. You can use this function to build | |
1562 your own @code{incf}-like modify macros. (Actually, it is | |
1563 better to use the internal functions @code{cl-setf-do-modify} | |
1564 and @code{cl-setf-do-store}, which are a bit easier to use and | |
1565 which also do a number of optimizations; consult the source | |
1566 code for the @code{incf} function for a simple example.) | |
1567 | |
1568 The argument @var{env} specifies the ``environment'' to be | |
1569 passed on to @code{macroexpand} if @code{get-setf-method} should | |
1570 need to expand a macro in @var{place}. It should come from | |
1571 an @code{&environment} argument to the macro or setf-method | |
1572 that called @code{get-setf-method}. | |
1573 | |
1574 See also the source code for the setf-methods for @code{apply} | |
1575 and @code{substring}, each of which works by calling | |
1576 @code{get-setf-method} on a simpler case, then massaging | |
1577 the result in various ways. | |
1578 @end defun | |
1579 | |
1580 Modern Common Lisp defines a second, independent way to specify | |
1581 the @code{setf} behavior of a function, namely ``@code{setf} | |
1582 functions'' whose names are lists @code{(setf @var{name})} | |
1583 rather than symbols. For example, @code{(defun (setf foo) @dots{})} | |
1584 defines the function that is used when @code{setf} is applied to | |
1585 @code{foo}. This package does not currently support @code{setf} | |
1586 functions. In particular, it is a compile-time error to use | |
1587 @code{setf} on a form which has not already been @code{defsetf}'d | |
1588 or otherwise declared; in newer Common Lisps, this would not be | |
1589 an error since the function @code{(setf @var{func})} might be | |
1590 defined later. | |
1591 | |
1592 @iftex | |
1593 @secno=4 | |
1594 @end iftex | |
1595 | |
1596 @node Variable Bindings, Conditionals, Generalized Variables, Control Structure | |
1597 @section Variable Bindings | |
1598 | |
1599 @noindent | |
1600 These Lisp forms make bindings to variables and function names, | |
1601 analogous to Lisp's built-in @code{let} form. | |
1602 | |
1603 @xref{Modify Macros}, for the @code{letf} and @code{letf*} forms which | |
1604 are also related to variable bindings. | |
1605 | |
1606 @menu | |
1607 * Dynamic Bindings:: The `progv' form | |
1608 * Lexical Bindings:: `lexical-let' and lexical closures | |
1609 * Function Bindings:: `flet' and `labels' | |
1610 * Macro Bindings:: `macrolet' and `symbol-macrolet' | |
1611 @end menu | |
1612 | |
1613 @node Dynamic Bindings, Lexical Bindings, Variable Bindings, Variable Bindings | |
1614 @subsection Dynamic Bindings | |
1615 | |
1616 @noindent | |
1617 The standard @code{let} form binds variables whose names are known | |
1618 at compile-time. The @code{progv} form provides an easy way to | |
1619 bind variables whose names are computed at run-time. | |
1620 | |
1621 @defspec progv symbols values forms@dots{} | |
1622 This form establishes @code{let}-style variable bindings on a | |
1623 set of variables computed at run-time. The expressions | |
1624 @var{symbols} and @var{values} are evaluated, and must return lists | |
1625 of symbols and values, respectively. The symbols are bound to the | |
1626 corresponding values for the duration of the body @var{form}s. | |
1627 If @var{values} is shorter than @var{symbols}, the last few symbols | |
1628 are made unbound (as if by @code{makunbound}) inside the body. | |
1629 If @var{symbols} is shorter than @var{values}, the excess values | |
1630 are ignored. | |
1631 @end defspec | |
1632 | |
1633 @node Lexical Bindings, Function Bindings, Dynamic Bindings, Variable Bindings | |
1634 @subsection Lexical Bindings | |
1635 | |
1636 @noindent | |
1637 The @dfn{CL} package defines the following macro which | |
1638 more closely follows the Common Lisp @code{let} form: | |
1639 | |
1640 @defspec lexical-let (bindings@dots{}) forms@dots{} | |
1641 This form is exactly like @code{let} except that the bindings it | |
1642 establishes are purely lexical. Lexical bindings are similar to | |
1643 local variables in a language like C: Only the code physically | |
1644 within the body of the @code{lexical-let} (after macro expansion) | |
1645 may refer to the bound variables. | |
1646 | |
1647 @example | |
1648 (setq a 5) | |
1649 (defun foo (b) (+ a b)) | |
1650 (let ((a 2)) (foo a)) | |
1651 @result{} 4 | |
1652 (lexical-let ((a 2)) (foo a)) | |
1653 @result{} 7 | |
1654 @end example | |
1655 | |
1656 @noindent | |
1657 In this example, a regular @code{let} binding of @code{a} actually | |
1658 makes a temporary change to the global variable @code{a}, so @code{foo} | |
1659 is able to see the binding of @code{a} to 2. But @code{lexical-let} | |
1660 actually creates a distinct local variable @code{a} for use within its | |
1661 body, without any effect on the global variable of the same name. | |
1662 | |
1663 The most important use of lexical bindings is to create @dfn{closures}. | |
1664 A closure is a function object that refers to an outside lexical | |
1665 variable. For example: | |
1666 | |
1667 @example | |
1668 (defun make-adder (n) | |
1669 (lexical-let ((n n)) | |
1670 (function (lambda (m) (+ n m))))) | |
1671 (setq add17 (make-adder 17)) | |
1672 (funcall add17 4) | |
1673 @result{} 21 | |
1674 @end example | |
1675 | |
1676 @noindent | |
1677 The call @code{(make-adder 17)} returns a function object which adds | |
1678 17 to its argument. If @code{let} had been used instead of | |
1679 @code{lexical-let}, the function object would have referred to the | |
1680 global @code{n}, which would have been bound to 17 only during the | |
1681 call to @code{make-adder} itself. | |
1682 | |
1683 @example | |
1684 (defun make-counter () | |
1685 (lexical-let ((n 0)) | |
1686 (function* (lambda (&optional (m 1)) (incf n m))))) | |
1687 (setq count-1 (make-counter)) | |
1688 (funcall count-1 3) | |
1689 @result{} 3 | |
1690 (funcall count-1 14) | |
1691 @result{} 17 | |
1692 (setq count-2 (make-counter)) | |
1693 (funcall count-2 5) | |
1694 @result{} 5 | |
1695 (funcall count-1 2) | |
1696 @result{} 19 | |
1697 (funcall count-2) | |
1698 @result{} 6 | |
1699 @end example | |
1700 | |
1701 @noindent | |
1702 Here we see that each call to @code{make-counter} creates a distinct | |
1703 local variable @code{n}, which serves as a private counter for the | |
1704 function object that is returned. | |
1705 | |
1706 Closed-over lexical variables persist until the last reference to | |
1707 them goes away, just like all other Lisp objects. For example, | |
1708 @code{count-2} refers to a function object which refers to an | |
1709 instance of the variable @code{n}; this is the only reference | |
1710 to that variable, so after @code{(setq count-2 nil)} the garbage | |
1711 collector would be able to delete this instance of @code{n}. | |
1712 Of course, if a @code{lexical-let} does not actually create any | |
1713 closures, then the lexical variables are free as soon as the | |
1714 @code{lexical-let} returns. | |
1715 | |
1716 Many closures are used only during the extent of the bindings they | |
1717 refer to; these are known as ``downward funargs'' in Lisp parlance. | |
1718 When a closure is used in this way, regular Emacs Lisp dynamic | |
1719 bindings suffice and will be more efficient than @code{lexical-let} | |
1720 closures: | |
1721 | |
1722 @example | |
1723 (defun add-to-list (x list) | |
28344
54fda0e8528a
Weed out redundant uses of `function'
Stefan Monnier <monnier@iro.umontreal.ca>
parents:
28039
diff
changeset
|
1724 (mapcar (lambda (y) (+ x y))) list) |
25829 | 1725 (add-to-list 7 '(1 2 5)) |
1726 @result{} (8 9 12) | |
1727 @end example | |
1728 | |
1729 @noindent | |
1730 Since this lambda is only used while @code{x} is still bound, | |
1731 it is not necessary to make a true closure out of it. | |
1732 | |
1733 You can use @code{defun} or @code{flet} inside a @code{lexical-let} | |
1734 to create a named closure. If several closures are created in the | |
1735 body of a single @code{lexical-let}, they all close over the same | |
1736 instance of the lexical variable. | |
1737 | |
1738 The @code{lexical-let} form is an extension to Common Lisp. In | |
1739 true Common Lisp, all bindings are lexical unless declared otherwise. | |
1740 @end defspec | |
1741 | |
1742 @defspec lexical-let* (bindings@dots{}) forms@dots{} | |
1743 This form is just like @code{lexical-let}, except that the bindings | |
1744 are made sequentially in the manner of @code{let*}. | |
1745 @end defspec | |
1746 | |
1747 @node Function Bindings, Macro Bindings, Lexical Bindings, Variable Bindings | |
1748 @subsection Function Bindings | |
1749 | |
1750 @noindent | |
1751 These forms make @code{let}-like bindings to functions instead | |
1752 of variables. | |
1753 | |
1754 @defspec flet (bindings@dots{}) forms@dots{} | |
1755 This form establishes @code{let}-style bindings on the function | |
1756 cells of symbols rather than on the value cells. Each @var{binding} | |
1757 must be a list of the form @samp{(@var{name} @var{arglist} | |
1758 @var{forms}@dots{})}, which defines a function exactly as if | |
1759 it were a @code{defun*} form. The function @var{name} is defined | |
1760 accordingly for the duration of the body of the @code{flet}; then | |
1761 the old function definition, or lack thereof, is restored. | |
1762 | |
1763 While @code{flet} in Common Lisp establishes a lexical binding of | |
1764 @var{name}, Emacs Lisp @code{flet} makes a dynamic binding. The | |
1765 result is that @code{flet} affects indirect calls to a function as | |
1766 well as calls directly inside the @code{flet} form itself. | |
1767 | |
1768 You can use @code{flet} to disable or modify the behavior of a | |
1769 function in a temporary fashion. This will even work on Emacs | |
1770 primitives, although note that some calls to primitive functions | |
1771 internal to Emacs are made without going through the symbol's | |
1772 function cell, and so will not be affected by @code{flet}. For | |
1773 example, | |
1774 | |
1775 @example | |
1776 (flet ((message (&rest args) (push args saved-msgs))) | |
1777 (do-something)) | |
1778 @end example | |
1779 | |
1780 This code attempts to replace the built-in function @code{message} | |
1781 with a function that simply saves the messages in a list rather | |
1782 than displaying them. The original definition of @code{message} | |
1783 will be restored after @code{do-something} exits. This code will | |
1784 work fine on messages generated by other Lisp code, but messages | |
1785 generated directly inside Emacs will not be caught since they make | |
1786 direct C-language calls to the message routines rather than going | |
1787 through the Lisp @code{message} function. | |
1788 | |
1789 Functions defined by @code{flet} may use the full Common Lisp | |
1790 argument notation supported by @code{defun*}; also, the function | |
1791 body is enclosed in an implicit block as if by @code{defun*}. | |
1792 @xref{Program Structure}. | |
1793 @end defspec | |
1794 | |
1795 @defspec labels (bindings@dots{}) forms@dots{} | |
1796 The @code{labels} form is like @code{flet}, except that it | |
1797 makes lexical bindings of the function names rather than | |
1798 dynamic bindings. (In true Common Lisp, both @code{flet} and | |
1799 @code{labels} make lexical bindings of slightly different sorts; | |
1800 since Emacs Lisp is dynamically bound by default, it seemed | |
1801 more appropriate for @code{flet} also to use dynamic binding. | |
1802 The @code{labels} form, with its lexical binding, is fully | |
1803 compatible with Common Lisp.) | |
1804 | |
1805 Lexical scoping means that all references to the named | |
1806 functions must appear physically within the body of the | |
1807 @code{labels} form. References may appear both in the body | |
1808 @var{forms} of @code{labels} itself, and in the bodies of | |
1809 the functions themselves. Thus, @code{labels} can define | |
1810 local recursive functions, or mutually-recursive sets of | |
1811 functions. | |
1812 | |
1813 A ``reference'' to a function name is either a call to that | |
1814 function, or a use of its name quoted by @code{quote} or | |
1815 @code{function} to be passed on to, say, @code{mapcar}. | |
1816 @end defspec | |
1817 | |
31572
384603523f41
Remove obsolescences, especially hash tables and refs to Emacs 19.
Dave Love <fx@gnu.org>
parents:
30009
diff
changeset
|
1818 @node Macro Bindings, , Function Bindings, Variable Bindings |
25829 | 1819 @subsection Macro Bindings |
1820 | |
1821 @noindent | |
1822 These forms create local macros and ``symbol macros.'' | |
1823 | |
1824 @defspec macrolet (bindings@dots{}) forms@dots{} | |
1825 This form is analogous to @code{flet}, but for macros instead of | |
1826 functions. Each @var{binding} is a list of the same form as the | |
1827 arguments to @code{defmacro*} (i.e., a macro name, argument list, | |
1828 and macro-expander forms). The macro is defined accordingly for | |
1829 use within the body of the @code{macrolet}. | |
1830 | |
1831 Because of the nature of macros, @code{macrolet} is lexically | |
1832 scoped even in Emacs Lisp: The @code{macrolet} binding will | |
1833 affect only calls that appear physically within the body | |
1834 @var{forms}, possibly after expansion of other macros in the | |
1835 body. | |
1836 @end defspec | |
1837 | |
1838 @defspec symbol-macrolet (bindings@dots{}) forms@dots{} | |
1839 This form creates @dfn{symbol macros}, which are macros that look | |
1840 like variable references rather than function calls. Each | |
1841 @var{binding} is a list @samp{(@var{var} @var{expansion})}; | |
1842 any reference to @var{var} within the body @var{forms} is | |
1843 replaced by @var{expansion}. | |
1844 | |
1845 @example | |
1846 (setq bar '(5 . 9)) | |
1847 (symbol-macrolet ((foo (car bar))) | |
1848 (incf foo)) | |
1849 bar | |
1850 @result{} (6 . 9) | |
1851 @end example | |
1852 | |
1853 A @code{setq} of a symbol macro is treated the same as a @code{setf}. | |
1854 I.e., @code{(setq foo 4)} in the above would be equivalent to | |
1855 @code{(setf foo 4)}, which in turn expands to @code{(setf (car bar) 4)}. | |
1856 | |
1857 Likewise, a @code{let} or @code{let*} binding a symbol macro is | |
1858 treated like a @code{letf} or @code{letf*}. This differs from true | |
1859 Common Lisp, where the rules of lexical scoping cause a @code{let} | |
1860 binding to shadow a @code{symbol-macrolet} binding. In this package, | |
1861 only @code{lexical-let} and @code{lexical-let*} will shadow a symbol | |
1862 macro. | |
1863 | |
1864 There is no analogue of @code{defmacro} for symbol macros; all symbol | |
1865 macros are local. A typical use of @code{symbol-macrolet} is in the | |
1866 expansion of another macro: | |
1867 | |
1868 @example | |
1869 (defmacro* my-dolist ((x list) &rest body) | |
1870 (let ((var (gensym))) | |
1871 (list 'loop 'for var 'on list 'do | |
1872 (list* 'symbol-macrolet (list (list x (list 'car var))) | |
1873 body)))) | |
1874 | |
1875 (setq mylist '(1 2 3 4)) | |
1876 (my-dolist (x mylist) (incf x)) | |
1877 mylist | |
1878 @result{} (2 3 4 5) | |
1879 @end example | |
1880 | |
1881 @noindent | |
1882 In this example, the @code{my-dolist} macro is similar to @code{dolist} | |
27511
ead47a92d9d0
Replace dotimes, dolist, with comment.
Dave Love <fx@gnu.org>
parents:
27487
diff
changeset
|
1883 (@pxref{Iteration}) except that the variable @code{x} becomes a true |
25829 | 1884 reference onto the elements of the list. The @code{my-dolist} call |
1885 shown here expands to | |
1886 | |
1887 @example | |
1888 (loop for G1234 on mylist do | |
1889 (symbol-macrolet ((x (car G1234))) | |
1890 (incf x))) | |
1891 @end example | |
1892 | |
1893 @noindent | |
1894 which in turn expands to | |
1895 | |
1896 @example | |
1897 (loop for G1234 on mylist do (incf (car G1234))) | |
1898 @end example | |
1899 | |
1900 @xref{Loop Facility}, for a description of the @code{loop} macro. | |
1901 This package defines a nonstandard @code{in-ref} loop clause that | |
1902 works much like @code{my-dolist}. | |
1903 @end defspec | |
1904 | |
1905 @node Conditionals, Blocks and Exits, Variable Bindings, Control Structure | |
1906 @section Conditionals | |
1907 | |
1908 @noindent | |
1909 These conditional forms augment Emacs Lisp's simple @code{if}, | |
1910 @code{and}, @code{or}, and @code{cond} forms. | |
1911 | |
1912 @defspec case keyform clause@dots{} | |
1913 This macro evaluates @var{keyform}, then compares it with the key | |
1914 values listed in the various @var{clause}s. Whichever clause matches | |
1915 the key is executed; comparison is done by @code{eql}. If no clause | |
1916 matches, the @code{case} form returns @code{nil}. The clauses are | |
1917 of the form | |
1918 | |
1919 @example | |
1920 (@var{keylist} @var{body-forms}@dots{}) | |
1921 @end example | |
1922 | |
1923 @noindent | |
1924 where @var{keylist} is a list of key values. If there is exactly | |
1925 one value, and it is not a cons cell or the symbol @code{nil} or | |
1926 @code{t}, then it can be used by itself as a @var{keylist} without | |
1927 being enclosed in a list. All key values in the @code{case} form | |
1928 must be distinct. The final clauses may use @code{t} in place of | |
1929 a @var{keylist} to indicate a default clause that should be taken | |
1930 if none of the other clauses match. (The symbol @code{otherwise} | |
1931 is also recognized in place of @code{t}. To make a clause that | |
1932 matches the actual symbol @code{t}, @code{nil}, or @code{otherwise}, | |
1933 enclose the symbol in a list.) | |
1934 | |
1935 For example, this expression reads a keystroke, then does one of | |
1936 four things depending on whether it is an @samp{a}, a @samp{b}, | |
1937 a @key{RET} or @kbd{C-j}, or anything else. | |
1938 | |
1939 @example | |
1940 (case (read-char) | |
1941 (?a (do-a-thing)) | |
1942 (?b (do-b-thing)) | |
1943 ((?\r ?\n) (do-ret-thing)) | |
1944 (t (do-other-thing))) | |
1945 @end example | |
1946 @end defspec | |
1947 | |
1948 @defspec ecase keyform clause@dots{} | |
1949 This macro is just like @code{case}, except that if the key does | |
1950 not match any of the clauses, an error is signaled rather than | |
1951 simply returning @code{nil}. | |
1952 @end defspec | |
1953 | |
1954 @defspec typecase keyform clause@dots{} | |
1955 This macro is a version of @code{case} that checks for types | |
1956 rather than values. Each @var{clause} is of the form | |
1957 @samp{(@var{type} @var{body}...)}. @xref{Type Predicates}, | |
1958 for a description of type specifiers. For example, | |
1959 | |
1960 @example | |
1961 (typecase x | |
1962 (integer (munch-integer x)) | |
1963 (float (munch-float x)) | |
1964 (string (munch-integer (string-to-int x))) | |
1965 (t (munch-anything x))) | |
1966 @end example | |
1967 | |
1968 The type specifier @code{t} matches any type of object; the word | |
1969 @code{otherwise} is also allowed. To make one clause match any of | |
1970 several types, use an @code{(or ...)} type specifier. | |
1971 @end defspec | |
1972 | |
1973 @defspec etypecase keyform clause@dots{} | |
1974 This macro is just like @code{typecase}, except that if the key does | |
1975 not match any of the clauses, an error is signaled rather than | |
1976 simply returning @code{nil}. | |
1977 @end defspec | |
1978 | |
1979 @node Blocks and Exits, Iteration, Conditionals, Control Structure | |
1980 @section Blocks and Exits | |
1981 | |
1982 @noindent | |
1983 Common Lisp @dfn{blocks} provide a non-local exit mechanism very | |
1984 similar to @code{catch} and @code{throw}, but lexically rather than | |
1985 dynamically scoped. This package actually implements @code{block} | |
1986 in terms of @code{catch}; however, the lexical scoping allows the | |
1987 optimizing byte-compiler to omit the costly @code{catch} step if the | |
1988 body of the block does not actually @code{return-from} the block. | |
1989 | |
1990 @defspec block name forms@dots{} | |
1991 The @var{forms} are evaluated as if by a @code{progn}. However, | |
1992 if any of the @var{forms} execute @code{(return-from @var{name})}, | |
1993 they will jump out and return directly from the @code{block} form. | |
1994 The @code{block} returns the result of the last @var{form} unless | |
1995 a @code{return-from} occurs. | |
1996 | |
1997 The @code{block}/@code{return-from} mechanism is quite similar to | |
1998 the @code{catch}/@code{throw} mechanism. The main differences are | |
1999 that block @var{name}s are unevaluated symbols, rather than forms | |
2000 (such as quoted symbols) which evaluate to a tag at run-time; and | |
2001 also that blocks are lexically scoped whereas @code{catch}/@code{throw} | |
2002 are dynamically scoped. This means that functions called from the | |
2003 body of a @code{catch} can also @code{throw} to the @code{catch}, | |
2004 but the @code{return-from} referring to a block name must appear | |
2005 physically within the @var{forms} that make up the body of the block. | |
2006 They may not appear within other called functions, although they may | |
2007 appear within macro expansions or @code{lambda}s in the body. Block | |
2008 names and @code{catch} names form independent name-spaces. | |
2009 | |
2010 In true Common Lisp, @code{defun} and @code{defmacro} surround | |
2011 the function or expander bodies with implicit blocks with the | |
2012 same name as the function or macro. This does not occur in Emacs | |
2013 Lisp, but this package provides @code{defun*} and @code{defmacro*} | |
2014 forms which do create the implicit block. | |
2015 | |
2016 The Common Lisp looping constructs defined by this package, | |
27511
ead47a92d9d0
Replace dotimes, dolist, with comment.
Dave Love <fx@gnu.org>
parents:
27487
diff
changeset
|
2017 such as @code{loop} and @code{dolist}, also create implicit blocks |
25829 | 2018 just as in Common Lisp. |
2019 | |
2020 Because they are implemented in terms of Emacs Lisp @code{catch} | |
2021 and @code{throw}, blocks have the same overhead as actual | |
2022 @code{catch} constructs (roughly two function calls). However, | |
31572
384603523f41
Remove obsolescences, especially hash tables and refs to Emacs 19.
Dave Love <fx@gnu.org>
parents:
30009
diff
changeset
|
2023 the optimizing byte compiler will optimize away the @code{catch} |
384603523f41
Remove obsolescences, especially hash tables and refs to Emacs 19.
Dave Love <fx@gnu.org>
parents:
30009
diff
changeset
|
2024 if the block does |
25829 | 2025 not in fact contain any @code{return} or @code{return-from} calls |
2026 that jump to it. This means that @code{do} loops and @code{defun*} | |
2027 functions which don't use @code{return} don't pay the overhead to | |
2028 support it. | |
2029 @end defspec | |
2030 | |
2031 @defspec return-from name [result] | |
2032 This macro returns from the block named @var{name}, which must be | |
2033 an (unevaluated) symbol. If a @var{result} form is specified, it | |
2034 is evaluated to produce the result returned from the @code{block}. | |
2035 Otherwise, @code{nil} is returned. | |
2036 @end defspec | |
2037 | |
2038 @defspec return [result] | |
2039 This macro is exactly like @code{(return-from nil @var{result})}. | |
27511
ead47a92d9d0
Replace dotimes, dolist, with comment.
Dave Love <fx@gnu.org>
parents:
27487
diff
changeset
|
2040 Common Lisp loops like @code{do} and @code{dolist} implicitly enclose |
25829 | 2041 themselves in @code{nil} blocks. |
2042 @end defspec | |
2043 | |
2044 @node Iteration, Loop Facility, Blocks and Exits, Control Structure | |
2045 @section Iteration | |
2046 | |
2047 @noindent | |
2048 The macros described here provide more sophisticated, high-level | |
2049 looping constructs to complement Emacs Lisp's basic @code{while} | |
2050 loop. | |
2051 | |
2052 @defspec loop forms@dots{} | |
2053 The @dfn{CL} package supports both the simple, old-style meaning of | |
2054 @code{loop} and the extremely powerful and flexible feature known as | |
2055 the @dfn{Loop Facility} or @dfn{Loop Macro}. This more advanced | |
2056 facility is discussed in the following section; @pxref{Loop Facility}. | |
2057 The simple form of @code{loop} is described here. | |
2058 | |
2059 If @code{loop} is followed by zero or more Lisp expressions, | |
2060 then @code{(loop @var{exprs}@dots{})} simply creates an infinite | |
2061 loop executing the expressions over and over. The loop is | |
2062 enclosed in an implicit @code{nil} block. Thus, | |
2063 | |
2064 @example | |
2065 (loop (foo) (if (no-more) (return 72)) (bar)) | |
2066 @end example | |
2067 | |
2068 @noindent | |
2069 is exactly equivalent to | |
2070 | |
2071 @example | |
2072 (block nil (while t (foo) (if (no-more) (return 72)) (bar))) | |
2073 @end example | |
2074 | |
2075 If any of the expressions are plain symbols, the loop is instead | |
2076 interpreted as a Loop Macro specification as described later. | |
2077 (This is not a restriction in practice, since a plain symbol | |
2078 in the above notation would simply access and throw away the | |
2079 value of a variable.) | |
2080 @end defspec | |
2081 | |
2082 @defspec do (spec@dots{}) (end-test [result@dots{}]) forms@dots{} | |
2083 This macro creates a general iterative loop. Each @var{spec} is | |
2084 of the form | |
2085 | |
2086 @example | |
2087 (@var{var} [@var{init} [@var{step}]]) | |
2088 @end example | |
2089 | |
2090 The loop works as follows: First, each @var{var} is bound to the | |
2091 associated @var{init} value as if by a @code{let} form. Then, in | |
2092 each iteration of the loop, the @var{end-test} is evaluated; if | |
2093 true, the loop is finished. Otherwise, the body @var{forms} are | |
2094 evaluated, then each @var{var} is set to the associated @var{step} | |
2095 expression (as if by a @code{psetq} form) and the next iteration | |
2096 begins. Once the @var{end-test} becomes true, the @var{result} | |
2097 forms are evaluated (with the @var{var}s still bound to their | |
2098 values) to produce the result returned by @code{do}. | |
2099 | |
2100 The entire @code{do} loop is enclosed in an implicit @code{nil} | |
2101 block, so that you can use @code{(return)} to break out of the | |
2102 loop at any time. | |
2103 | |
2104 If there are no @var{result} forms, the loop returns @code{nil}. | |
2105 If a given @var{var} has no @var{step} form, it is bound to its | |
2106 @var{init} value but not otherwise modified during the @code{do} | |
2107 loop (unless the code explicitly modifies it); this case is just | |
2108 a shorthand for putting a @code{(let ((@var{var} @var{init})) @dots{})} | |
2109 around the loop. If @var{init} is also omitted it defaults to | |
2110 @code{nil}, and in this case a plain @samp{@var{var}} can be used | |
2111 in place of @samp{(@var{var})}, again following the analogy with | |
2112 @code{let}. | |
2113 | |
2114 This example (from Steele) illustrates a loop which applies the | |
2115 function @code{f} to successive pairs of values from the lists | |
2116 @code{foo} and @code{bar}; it is equivalent to the call | |
2117 @code{(mapcar* 'f foo bar)}. Note that this loop has no body | |
2118 @var{forms} at all, performing all its work as side effects of | |
2119 the rest of the loop. | |
2120 | |
2121 @example | |
2122 (do ((x foo (cdr x)) | |
2123 (y bar (cdr y)) | |
2124 (z nil (cons (f (car x) (car y)) z))) | |
2125 ((or (null x) (null y)) | |
2126 (nreverse z))) | |
2127 @end example | |
2128 @end defspec | |
2129 | |
2130 @defspec do* (spec@dots{}) (end-test [result@dots{}]) forms@dots{} | |
2131 This is to @code{do} what @code{let*} is to @code{let}. In | |
2132 particular, the initial values are bound as if by @code{let*} | |
2133 rather than @code{let}, and the steps are assigned as if by | |
2134 @code{setq} rather than @code{psetq}. | |
2135 | |
2136 Here is another way to write the above loop: | |
2137 | |
2138 @example | |
2139 (do* ((xp foo (cdr xp)) | |
2140 (yp bar (cdr yp)) | |
2141 (x (car xp) (car xp)) | |
2142 (y (car yp) (car yp)) | |
2143 z) | |
2144 ((or (null xp) (null yp)) | |
2145 (nreverse z)) | |
2146 (push (f x y) z)) | |
2147 @end example | |
2148 @end defspec | |
2149 | |
27511
ead47a92d9d0
Replace dotimes, dolist, with comment.
Dave Love <fx@gnu.org>
parents:
27487
diff
changeset
|
2150 @defspec dolist (var list [result]) forms@dots{} |
ead47a92d9d0
Replace dotimes, dolist, with comment.
Dave Love <fx@gnu.org>
parents:
27487
diff
changeset
|
2151 This is a more specialized loop which iterates across the elements |
ead47a92d9d0
Replace dotimes, dolist, with comment.
Dave Love <fx@gnu.org>
parents:
27487
diff
changeset
|
2152 of a list. @var{list} should evaluate to a list; the body @var{forms} |
ead47a92d9d0
Replace dotimes, dolist, with comment.
Dave Love <fx@gnu.org>
parents:
27487
diff
changeset
|
2153 are executed with @var{var} bound to each element of the list in |
ead47a92d9d0
Replace dotimes, dolist, with comment.
Dave Love <fx@gnu.org>
parents:
27487
diff
changeset
|
2154 turn. Finally, the @var{result} form (or @code{nil}) is evaluated |
ead47a92d9d0
Replace dotimes, dolist, with comment.
Dave Love <fx@gnu.org>
parents:
27487
diff
changeset
|
2155 with @var{var} bound to @code{nil} to produce the result returned by |
ead47a92d9d0
Replace dotimes, dolist, with comment.
Dave Love <fx@gnu.org>
parents:
27487
diff
changeset
|
2156 the loop. Unlike with Emacs's built in @code{dolist}, the loop is |
ead47a92d9d0
Replace dotimes, dolist, with comment.
Dave Love <fx@gnu.org>
parents:
27487
diff
changeset
|
2157 surrounded by an implicit @code{nil} block. |
ead47a92d9d0
Replace dotimes, dolist, with comment.
Dave Love <fx@gnu.org>
parents:
27487
diff
changeset
|
2158 @end defspec |
ead47a92d9d0
Replace dotimes, dolist, with comment.
Dave Love <fx@gnu.org>
parents:
27487
diff
changeset
|
2159 |
ead47a92d9d0
Replace dotimes, dolist, with comment.
Dave Love <fx@gnu.org>
parents:
27487
diff
changeset
|
2160 @defspec dotimes (var count [result]) forms@dots{} |
ead47a92d9d0
Replace dotimes, dolist, with comment.
Dave Love <fx@gnu.org>
parents:
27487
diff
changeset
|
2161 This is a more specialized loop which iterates a specified number |
ead47a92d9d0
Replace dotimes, dolist, with comment.
Dave Love <fx@gnu.org>
parents:
27487
diff
changeset
|
2162 of times. The body is executed with @var{var} bound to the integers |
ead47a92d9d0
Replace dotimes, dolist, with comment.
Dave Love <fx@gnu.org>
parents:
27487
diff
changeset
|
2163 from zero (inclusive) to @var{count} (exclusive), in turn. Then |
ead47a92d9d0
Replace dotimes, dolist, with comment.
Dave Love <fx@gnu.org>
parents:
27487
diff
changeset
|
2164 the @code{result} form is evaluated with @var{var} bound to the total |
ead47a92d9d0
Replace dotimes, dolist, with comment.
Dave Love <fx@gnu.org>
parents:
27487
diff
changeset
|
2165 number of iterations that were done (i.e., @code{(max 0 @var{count})}) |
ead47a92d9d0
Replace dotimes, dolist, with comment.
Dave Love <fx@gnu.org>
parents:
27487
diff
changeset
|
2166 to get the return value for the loop form. Unlike with Emacs's built in |
ead47a92d9d0
Replace dotimes, dolist, with comment.
Dave Love <fx@gnu.org>
parents:
27487
diff
changeset
|
2167 @code{dolist}, the loop is surrounded by an implicit @code{nil} block. |
ead47a92d9d0
Replace dotimes, dolist, with comment.
Dave Love <fx@gnu.org>
parents:
27487
diff
changeset
|
2168 @end defspec |
ead47a92d9d0
Replace dotimes, dolist, with comment.
Dave Love <fx@gnu.org>
parents:
27487
diff
changeset
|
2169 |
25829 | 2170 @defspec do-symbols (var [obarray [result]]) forms@dots{} |
2171 This loop iterates over all interned symbols. If @var{obarray} | |
2172 is specified and is not @code{nil}, it loops over all symbols in | |
2173 that obarray. For each symbol, the body @var{forms} are evaluated | |
2174 with @var{var} bound to that symbol. The symbols are visited in | |
2175 an unspecified order. Afterward the @var{result} form, if any, | |
2176 is evaluated (with @var{var} bound to @code{nil}) to get the return | |
2177 value. The loop is surrounded by an implicit @code{nil} block. | |
2178 @end defspec | |
2179 | |
2180 @defspec do-all-symbols (var [result]) forms@dots{} | |
2181 This is identical to @code{do-symbols} except that the @var{obarray} | |
2182 argument is omitted; it always iterates over the default obarray. | |
2183 @end defspec | |
2184 | |
2185 @xref{Mapping over Sequences}, for some more functions for | |
2186 iterating over vectors or lists. | |
2187 | |
2188 @node Loop Facility, Multiple Values, Iteration, Control Structure | |
2189 @section Loop Facility | |
2190 | |
2191 @noindent | |
2192 A common complaint with Lisp's traditional looping constructs is | |
2193 that they are either too simple and limited, such as Common Lisp's | |
2194 @code{dotimes} or Emacs Lisp's @code{while}, or too unreadable and | |
2195 obscure, like Common Lisp's @code{do} loop. | |
2196 | |
2197 To remedy this, recent versions of Common Lisp have added a new | |
2198 construct called the ``Loop Facility'' or ``@code{loop} macro,'' | |
2199 with an easy-to-use but very powerful and expressive syntax. | |
2200 | |
2201 @menu | |
2202 * Loop Basics:: `loop' macro, basic clause structure | |
2203 * Loop Examples:: Working examples of `loop' macro | |
2204 * For Clauses:: Clauses introduced by `for' or `as' | |
2205 * Iteration Clauses:: `repeat', `while', `thereis', etc. | |
2206 * Accumulation Clauses:: `collect', `sum', `maximize', etc. | |
2207 * Other Clauses:: `with', `if', `initially', `finally' | |
2208 @end menu | |
2209 | |
2210 @node Loop Basics, Loop Examples, Loop Facility, Loop Facility | |
2211 @subsection Loop Basics | |
2212 | |
2213 @noindent | |
2214 The @code{loop} macro essentially creates a mini-language within | |
2215 Lisp that is specially tailored for describing loops. While this | |
2216 language is a little strange-looking by the standards of regular Lisp, | |
2217 it turns out to be very easy to learn and well-suited to its purpose. | |
2218 | |
2219 Since @code{loop} is a macro, all parsing of the loop language | |
2220 takes place at byte-compile time; compiled @code{loop}s are just | |
2221 as efficient as the equivalent @code{while} loops written longhand. | |
2222 | |
2223 @defspec loop clauses@dots{} | |
2224 A loop construct consists of a series of @var{clause}s, each | |
2225 introduced by a symbol like @code{for} or @code{do}. Clauses | |
2226 are simply strung together in the argument list of @code{loop}, | |
2227 with minimal extra parentheses. The various types of clauses | |
2228 specify initializations, such as the binding of temporary | |
2229 variables, actions to be taken in the loop, stepping actions, | |
2230 and final cleanup. | |
2231 | |
2232 Common Lisp specifies a certain general order of clauses in a | |
2233 loop: | |
2234 | |
2235 @example | |
2236 (loop @var{name-clause} | |
2237 @var{var-clauses}@dots{} | |
2238 @var{action-clauses}@dots{}) | |
2239 @end example | |
2240 | |
2241 The @var{name-clause} optionally gives a name to the implicit | |
2242 block that surrounds the loop. By default, the implicit block | |
2243 is named @code{nil}. The @var{var-clauses} specify what | |
2244 variables should be bound during the loop, and how they should | |
2245 be modified or iterated throughout the course of the loop. The | |
2246 @var{action-clauses} are things to be done during the loop, such | |
2247 as computing, collecting, and returning values. | |
2248 | |
2249 The Emacs version of the @code{loop} macro is less restrictive about | |
2250 the order of clauses, but things will behave most predictably if | |
2251 you put the variable-binding clauses @code{with}, @code{for}, and | |
2252 @code{repeat} before the action clauses. As in Common Lisp, | |
2253 @code{initially} and @code{finally} clauses can go anywhere. | |
2254 | |
2255 Loops generally return @code{nil} by default, but you can cause | |
2256 them to return a value by using an accumulation clause like | |
2257 @code{collect}, an end-test clause like @code{always}, or an | |
2258 explicit @code{return} clause to jump out of the implicit block. | |
2259 (Because the loop body is enclosed in an implicit block, you can | |
2260 also use regular Lisp @code{return} or @code{return-from} to | |
2261 break out of the loop.) | |
2262 @end defspec | |
2263 | |
2264 The following sections give some examples of the Loop Macro in | |
2265 action, and describe the particular loop clauses in great detail. | |
2266 Consult the second edition of Steele's @dfn{Common Lisp, the Language}, | |
2267 for additional discussion and examples of the @code{loop} macro. | |
2268 | |
2269 @node Loop Examples, For Clauses, Loop Basics, Loop Facility | |
2270 @subsection Loop Examples | |
2271 | |
2272 @noindent | |
2273 Before listing the full set of clauses that are allowed, let's | |
2274 look at a few example loops just to get a feel for the @code{loop} | |
2275 language. | |
2276 | |
2277 @example | |
2278 (loop for buf in (buffer-list) | |
2279 collect (buffer-file-name buf)) | |
2280 @end example | |
2281 | |
2282 @noindent | |
2283 This loop iterates over all Emacs buffers, using the list | |
2284 returned by @code{buffer-list}. For each buffer @code{buf}, | |
2285 it calls @code{buffer-file-name} and collects the results into | |
2286 a list, which is then returned from the @code{loop} construct. | |
2287 The result is a list of the file names of all the buffers in | |
2288 Emacs' memory. The words @code{for}, @code{in}, and @code{collect} | |
2289 are reserved words in the @code{loop} language. | |
2290 | |
2291 @example | |
2292 (loop repeat 20 do (insert "Yowsa\n")) | |
2293 @end example | |
2294 | |
2295 @noindent | |
2296 This loop inserts the phrase ``Yowsa'' twenty times in the | |
2297 current buffer. | |
2298 | |
2299 @example | |
2300 (loop until (eobp) do (munch-line) (forward-line 1)) | |
2301 @end example | |
2302 | |
2303 @noindent | |
2304 This loop calls @code{munch-line} on every line until the end | |
2305 of the buffer. If point is already at the end of the buffer, | |
2306 the loop exits immediately. | |
2307 | |
2308 @example | |
2309 (loop do (munch-line) until (eobp) do (forward-line 1)) | |
2310 @end example | |
2311 | |
2312 @noindent | |
2313 This loop is similar to the above one, except that @code{munch-line} | |
2314 is always called at least once. | |
2315 | |
2316 @example | |
2317 (loop for x from 1 to 100 | |
2318 for y = (* x x) | |
2319 until (>= y 729) | |
2320 finally return (list x (= y 729))) | |
2321 @end example | |
2322 | |
2323 @noindent | |
2324 This more complicated loop searches for a number @code{x} whose | |
2325 square is 729. For safety's sake it only examines @code{x} | |
2326 values up to 100; dropping the phrase @samp{to 100} would | |
2327 cause the loop to count upwards with no limit. The second | |
2328 @code{for} clause defines @code{y} to be the square of @code{x} | |
2329 within the loop; the expression after the @code{=} sign is | |
2330 reevaluated each time through the loop. The @code{until} | |
2331 clause gives a condition for terminating the loop, and the | |
2332 @code{finally} clause says what to do when the loop finishes. | |
2333 (This particular example was written less concisely than it | |
2334 could have been, just for the sake of illustration.) | |
2335 | |
2336 Note that even though this loop contains three clauses (two | |
2337 @code{for}s and an @code{until}) that would have been enough to | |
2338 define loops all by themselves, it still creates a single loop | |
2339 rather than some sort of triple-nested loop. You must explicitly | |
2340 nest your @code{loop} constructs if you want nested loops. | |
2341 | |
2342 @node For Clauses, Iteration Clauses, Loop Examples, Loop Facility | |
2343 @subsection For Clauses | |
2344 | |
2345 @noindent | |
2346 Most loops are governed by one or more @code{for} clauses. | |
2347 A @code{for} clause simultaneously describes variables to be | |
2348 bound, how those variables are to be stepped during the loop, | |
2349 and usually an end condition based on those variables. | |
2350 | |
2351 The word @code{as} is a synonym for the word @code{for}. This | |
2352 word is followed by a variable name, then a word like @code{from} | |
2353 or @code{across} that describes the kind of iteration desired. | |
2354 In Common Lisp, the phrase @code{being the} sometimes precedes | |
2355 the type of iteration; in this package both @code{being} and | |
2356 @code{the} are optional. The word @code{each} is a synonym | |
2357 for @code{the}, and the word that follows it may be singular | |
2358 or plural: @samp{for x being the elements of y} or | |
2359 @samp{for x being each element of y}. Which form you use | |
2360 is purely a matter of style. | |
2361 | |
2362 The variable is bound around the loop as if by @code{let}: | |
2363 | |
2364 @example | |
2365 (setq i 'happy) | |
2366 (loop for i from 1 to 10 do (do-something-with i)) | |
2367 i | |
2368 @result{} happy | |
2369 @end example | |
2370 | |
2371 @table @code | |
2372 @item for @var{var} from @var{expr1} to @var{expr2} by @var{expr3} | |
2373 This type of @code{for} clause creates a counting loop. Each of | |
2374 the three sub-terms is optional, though there must be at least one | |
2375 term so that the clause is marked as a counting clause. | |
2376 | |
2377 The three expressions are the starting value, the ending value, and | |
2378 the step value, respectively, of the variable. The loop counts | |
2379 upwards by default (@var{expr3} must be positive), from @var{expr1} | |
2380 to @var{expr2} inclusively. If you omit the @code{from} term, the | |
2381 loop counts from zero; if you omit the @code{to} term, the loop | |
2382 counts forever without stopping (unless stopped by some other | |
2383 loop clause, of course); if you omit the @code{by} term, the loop | |
2384 counts in steps of one. | |
2385 | |
2386 You can replace the word @code{from} with @code{upfrom} or | |
2387 @code{downfrom} to indicate the direction of the loop. Likewise, | |
2388 you can replace @code{to} with @code{upto} or @code{downto}. | |
2389 For example, @samp{for x from 5 downto 1} executes five times | |
2390 with @code{x} taking on the integers from 5 down to 1 in turn. | |
2391 Also, you can replace @code{to} with @code{below} or @code{above}, | |
2392 which are like @code{upto} and @code{downto} respectively except | |
2393 that they are exclusive rather than inclusive limits: | |
2394 | |
2395 @example | |
2396 (loop for x to 10 collect x) | |
2397 @result{} (0 1 2 3 4 5 6 7 8 9 10) | |
2398 (loop for x below 10 collect x) | |
2399 @result{} (0 1 2 3 4 5 6 7 8 9) | |
2400 @end example | |
2401 | |
2402 The @code{by} value is always positive, even for downward-counting | |
2403 loops. Some sort of @code{from} value is required for downward | |
2404 loops; @samp{for x downto 5} is not a legal loop clause all by | |
2405 itself. | |
2406 | |
2407 @item for @var{var} in @var{list} by @var{function} | |
2408 This clause iterates @var{var} over all the elements of @var{list}, | |
2409 in turn. If you specify the @code{by} term, then @var{function} | |
2410 is used to traverse the list instead of @code{cdr}; it must be a | |
2411 function taking one argument. For example: | |
2412 | |
2413 @example | |
2414 (loop for x in '(1 2 3 4 5 6) collect (* x x)) | |
2415 @result{} (1 4 9 16 25 36) | |
2416 (loop for x in '(1 2 3 4 5 6) by 'cddr collect (* x x)) | |
2417 @result{} (1 9 25) | |
2418 @end example | |
2419 | |
2420 @item for @var{var} on @var{list} by @var{function} | |
2421 This clause iterates @var{var} over all the cons cells of @var{list}. | |
2422 | |
2423 @example | |
2424 (loop for x on '(1 2 3 4) collect x) | |
2425 @result{} ((1 2 3 4) (2 3 4) (3 4) (4)) | |
2426 @end example | |
2427 | |
2428 With @code{by}, there is no real reason that the @code{on} expression | |
2429 must be a list. For example: | |
2430 | |
2431 @example | |
2432 (loop for x on first-animal by 'next-animal collect x) | |
2433 @end example | |
2434 | |
2435 @noindent | |
2436 where @code{(next-animal x)} takes an ``animal'' @var{x} and returns | |
2437 the next in the (assumed) sequence of animals, or @code{nil} if | |
2438 @var{x} was the last animal in the sequence. | |
2439 | |
2440 @item for @var{var} in-ref @var{list} by @var{function} | |
2441 This is like a regular @code{in} clause, but @var{var} becomes | |
2442 a @code{setf}-able ``reference'' onto the elements of the list | |
2443 rather than just a temporary variable. For example, | |
2444 | |
2445 @example | |
2446 (loop for x in-ref my-list do (incf x)) | |
2447 @end example | |
2448 | |
2449 @noindent | |
2450 increments every element of @code{my-list} in place. This clause | |
2451 is an extension to standard Common Lisp. | |
2452 | |
2453 @item for @var{var} across @var{array} | |
2454 This clause iterates @var{var} over all the elements of @var{array}, | |
2455 which may be a vector or a string. | |
2456 | |
2457 @example | |
2458 (loop for x across "aeiou" | |
2459 do (use-vowel (char-to-string x))) | |
2460 @end example | |
2461 | |
2462 @item for @var{var} across-ref @var{array} | |
2463 This clause iterates over an array, with @var{var} a @code{setf}-able | |
2464 reference onto the elements; see @code{in-ref} above. | |
2465 | |
2466 @item for @var{var} being the elements of @var{sequence} | |
2467 This clause iterates over the elements of @var{sequence}, which may | |
2468 be a list, vector, or string. Since the type must be determined | |
2469 at run-time, this is somewhat less efficient than @code{in} or | |
2470 @code{across}. The clause may be followed by the additional term | |
2471 @samp{using (index @var{var2})} to cause @var{var2} to be bound to | |
2472 the successive indices (starting at 0) of the elements. | |
2473 | |
2474 This clause type is taken from older versions of the @code{loop} macro, | |
2475 and is not present in modern Common Lisp. The @samp{using (sequence ...)} | |
2476 term of the older macros is not supported. | |
2477 | |
2478 @item for @var{var} being the elements of-ref @var{sequence} | |
2479 This clause iterates over a sequence, with @var{var} a @code{setf}-able | |
2480 reference onto the elements; see @code{in-ref} above. | |
2481 | |
2482 @item for @var{var} being the symbols [of @var{obarray}] | |
2483 This clause iterates over symbols, either over all interned symbols | |
2484 or over all symbols in @var{obarray}. The loop is executed with | |
2485 @var{var} bound to each symbol in turn. The symbols are visited in | |
2486 an unspecified order. | |
2487 | |
2488 As an example, | |
2489 | |
2490 @example | |
2491 (loop for sym being the symbols | |
2492 when (fboundp sym) | |
2493 when (string-match "^map" (symbol-name sym)) | |
2494 collect sym) | |
2495 @end example | |
2496 | |
2497 @noindent | |
2498 returns a list of all the functions whose names begin with @samp{map}. | |
2499 | |
2500 The Common Lisp words @code{external-symbols} and @code{present-symbols} | |
2501 are also recognized but are equivalent to @code{symbols} in Emacs Lisp. | |
2502 | |
2503 Due to a minor implementation restriction, it will not work to have | |
2504 more than one @code{for} clause iterating over symbols, hash tables, | |
2505 keymaps, overlays, or intervals in a given @code{loop}. Fortunately, | |
2506 it would rarely if ever be useful to do so. It @emph{is} legal to mix | |
2507 one of these types of clauses with other clauses like @code{for ... to} | |
2508 or @code{while}. | |
2509 | |
2510 @item for @var{var} being the hash-keys of @var{hash-table} | |
2511 This clause iterates over the entries in @var{hash-table}. For each | |
2512 hash table entry, @var{var} is bound to the entry's key. If you write | |
2513 @samp{the hash-values} instead, @var{var} is bound to the values | |
2514 of the entries. The clause may be followed by the additional | |
2515 term @samp{using (hash-values @var{var2})} (where @code{hash-values} | |
2516 is the opposite word of the word following @code{the}) to cause | |
2517 @var{var} and @var{var2} to be bound to the two parts of each | |
2518 hash table entry. | |
2519 | |
2520 @item for @var{var} being the key-codes of @var{keymap} | |
31572
384603523f41
Remove obsolescences, especially hash tables and refs to Emacs 19.
Dave Love <fx@gnu.org>
parents:
30009
diff
changeset
|
2521 This clause iterates over the entries in @var{keymap}. |
384603523f41
Remove obsolescences, especially hash tables and refs to Emacs 19.
Dave Love <fx@gnu.org>
parents:
30009
diff
changeset
|
2522 The iteration does not enter nested keymaps or inherited (parent) keymaps. |
25829 | 2523 You can use @samp{the key-bindings} to access the commands bound to |
2524 the keys rather than the key codes, and you can add a @code{using} | |
2525 clause to access both the codes and the bindings together. | |
2526 | |
2527 @item for @var{var} being the key-seqs of @var{keymap} | |
2528 This clause iterates over all key sequences defined by @var{keymap} | |
2529 and its nested keymaps, where @var{var} takes on values which are | |
27225
ee242fddb1ab
Remove various old stuff and references thereto.
Dave Love <fx@gnu.org>
parents:
25829
diff
changeset
|
2530 vectors. The strings or vectors |
25829 | 2531 are reused for each iteration, so you must copy them if you wish to keep |
2532 them permanently. You can add a @samp{using (key-bindings ...)} | |
2533 clause to get the command bindings as well. | |
2534 | |
2535 @item for @var{var} being the overlays [of @var{buffer}] @dots{} | |
31572
384603523f41
Remove obsolescences, especially hash tables and refs to Emacs 19.
Dave Love <fx@gnu.org>
parents:
30009
diff
changeset
|
2536 This clause iterates over the ``overlays'' of a buffer |
384603523f41
Remove obsolescences, especially hash tables and refs to Emacs 19.
Dave Love <fx@gnu.org>
parents:
30009
diff
changeset
|
2537 (the clause @code{extents} is synonymous |
27225
ee242fddb1ab
Remove various old stuff and references thereto.
Dave Love <fx@gnu.org>
parents:
25829
diff
changeset
|
2538 with @code{overlays}). If the @code{of} term is omitted, the current |
ee242fddb1ab
Remove various old stuff and references thereto.
Dave Love <fx@gnu.org>
parents:
25829
diff
changeset
|
2539 buffer is used. |
25829 | 2540 This clause also accepts optional @samp{from @var{pos}} and |
2541 @samp{to @var{pos}} terms, limiting the clause to overlays which | |
2542 overlap the specified region. | |
2543 | |
2544 @item for @var{var} being the intervals [of @var{buffer}] @dots{} | |
2545 This clause iterates over all intervals of a buffer with constant | |
2546 text properties. The variable @var{var} will be bound to conses | |
2547 of start and end positions, where one start position is always equal | |
2548 to the previous end position. The clause allows @code{of}, | |
2549 @code{from}, @code{to}, and @code{property} terms, where the latter | |
2550 term restricts the search to just the specified property. The | |
31572
384603523f41
Remove obsolescences, especially hash tables and refs to Emacs 19.
Dave Love <fx@gnu.org>
parents:
30009
diff
changeset
|
2551 @code{of} term may specify either a buffer or a string. |
25829 | 2552 |
2553 @item for @var{var} being the frames | |
2554 This clause iterates over all frames, i.e., X window system windows | |
31572
384603523f41
Remove obsolescences, especially hash tables and refs to Emacs 19.
Dave Love <fx@gnu.org>
parents:
30009
diff
changeset
|
2555 open on Emacs files. The |
25829 | 2556 clause @code{screens} is a synonym for @code{frames}. The frames |
2557 are visited in @code{next-frame} order starting from | |
2558 @code{selected-frame}. | |
2559 | |
2560 @item for @var{var} being the windows [of @var{frame}] | |
2561 This clause iterates over the windows (in the Emacs sense) of | |
27225
ee242fddb1ab
Remove various old stuff and references thereto.
Dave Love <fx@gnu.org>
parents:
25829
diff
changeset
|
2562 the current frame, or of the specified @var{frame}. |
25829 | 2563 |
2564 @item for @var{var} being the buffers | |
2565 This clause iterates over all buffers in Emacs. It is equivalent | |
2566 to @samp{for @var{var} in (buffer-list)}. | |
2567 | |
2568 @item for @var{var} = @var{expr1} then @var{expr2} | |
2569 This clause does a general iteration. The first time through | |
2570 the loop, @var{var} will be bound to @var{expr1}. On the second | |
2571 and successive iterations it will be set by evaluating @var{expr2} | |
2572 (which may refer to the old value of @var{var}). For example, | |
2573 these two loops are effectively the same: | |
2574 | |
2575 @example | |
2576 (loop for x on my-list by 'cddr do ...) | |
2577 (loop for x = my-list then (cddr x) while x do ...) | |
2578 @end example | |
2579 | |
2580 Note that this type of @code{for} clause does not imply any sort | |
2581 of terminating condition; the above example combines it with a | |
2582 @code{while} clause to tell when to end the loop. | |
2583 | |
2584 If you omit the @code{then} term, @var{expr1} is used both for | |
2585 the initial setting and for successive settings: | |
2586 | |
2587 @example | |
2588 (loop for x = (random) when (> x 0) return x) | |
2589 @end example | |
2590 | |
2591 @noindent | |
2592 This loop keeps taking random numbers from the @code{(random)} | |
2593 function until it gets a positive one, which it then returns. | |
2594 @end table | |
2595 | |
2596 If you include several @code{for} clauses in a row, they are | |
2597 treated sequentially (as if by @code{let*} and @code{setq}). | |
2598 You can instead use the word @code{and} to link the clauses, | |
2599 in which case they are processed in parallel (as if by @code{let} | |
2600 and @code{psetq}). | |
2601 | |
2602 @example | |
2603 (loop for x below 5 for y = nil then x collect (list x y)) | |
2604 @result{} ((0 nil) (1 1) (2 2) (3 3) (4 4)) | |
2605 (loop for x below 5 and y = nil then x collect (list x y)) | |
2606 @result{} ((0 nil) (1 0) (2 1) (3 2) (4 3)) | |
2607 @end example | |
2608 | |
2609 @noindent | |
2610 In the first loop, @code{y} is set based on the value of @code{x} | |
2611 that was just set by the previous clause; in the second loop, | |
2612 @code{x} and @code{y} are set simultaneously so @code{y} is set | |
2613 based on the value of @code{x} left over from the previous time | |
2614 through the loop. | |
2615 | |
2616 Another feature of the @code{loop} macro is @dfn{destructuring}, | |
2617 similar in concept to the destructuring provided by @code{defmacro}. | |
2618 The @var{var} part of any @code{for} clause can be given as a list | |
2619 of variables instead of a single variable. The values produced | |
2620 during loop execution must be lists; the values in the lists are | |
2621 stored in the corresponding variables. | |
2622 | |
2623 @example | |
2624 (loop for (x y) in '((2 3) (4 5) (6 7)) collect (+ x y)) | |
2625 @result{} (5 9 13) | |
2626 @end example | |
2627 | |
2628 In loop destructuring, if there are more values than variables | |
2629 the trailing values are ignored, and if there are more variables | |
2630 than values the trailing variables get the value @code{nil}. | |
2631 If @code{nil} is used as a variable name, the corresponding | |
2632 values are ignored. Destructuring may be nested, and dotted | |
2633 lists of variables like @code{(x . y)} are allowed. | |
2634 | |
2635 @node Iteration Clauses, Accumulation Clauses, For Clauses, Loop Facility | |
2636 @subsection Iteration Clauses | |
2637 | |
2638 @noindent | |
2639 Aside from @code{for} clauses, there are several other loop clauses | |
2640 that control the way the loop operates. They might be used by | |
2641 themselves, or in conjunction with one or more @code{for} clauses. | |
2642 | |
2643 @table @code | |
2644 @item repeat @var{integer} | |
2645 This clause simply counts up to the specified number using an | |
2646 internal temporary variable. The loops | |
2647 | |
2648 @example | |
2649 (loop repeat n do ...) | |
2650 (loop for temp to n do ...) | |
2651 @end example | |
2652 | |
2653 @noindent | |
2654 are identical except that the second one forces you to choose | |
2655 a name for a variable you aren't actually going to use. | |
2656 | |
2657 @item while @var{condition} | |
2658 This clause stops the loop when the specified condition (any Lisp | |
2659 expression) becomes @code{nil}. For example, the following two | |
2660 loops are equivalent, except for the implicit @code{nil} block | |
2661 that surrounds the second one: | |
2662 | |
2663 @example | |
2664 (while @var{cond} @var{forms}@dots{}) | |
2665 (loop while @var{cond} do @var{forms}@dots{}) | |
2666 @end example | |
2667 | |
2668 @item until @var{condition} | |
2669 This clause stops the loop when the specified condition is true, | |
2670 i.e., non-@code{nil}. | |
2671 | |
2672 @item always @var{condition} | |
2673 This clause stops the loop when the specified condition is @code{nil}. | |
2674 Unlike @code{while}, it stops the loop using @code{return nil} so that | |
2675 the @code{finally} clauses are not executed. If all the conditions | |
2676 were non-@code{nil}, the loop returns @code{t}: | |
2677 | |
2678 @example | |
2679 (if (loop for size in size-list always (> size 10)) | |
2680 (some-big-sizes) | |
2681 (no-big-sizes)) | |
2682 @end example | |
2683 | |
2684 @item never @var{condition} | |
2685 This clause is like @code{always}, except that the loop returns | |
2686 @code{t} if any conditions were false, or @code{nil} otherwise. | |
2687 | |
2688 @item thereis @var{condition} | |
2689 This clause stops the loop when the specified form is non-@code{nil}; | |
2690 in this case, it returns that non-@code{nil} value. If all the | |
2691 values were @code{nil}, the loop returns @code{nil}. | |
2692 @end table | |
2693 | |
2694 @node Accumulation Clauses, Other Clauses, Iteration Clauses, Loop Facility | |
2695 @subsection Accumulation Clauses | |
2696 | |
2697 @noindent | |
2698 These clauses cause the loop to accumulate information about the | |
2699 specified Lisp @var{form}. The accumulated result is returned | |
2700 from the loop unless overridden, say, by a @code{return} clause. | |
2701 | |
2702 @table @code | |
2703 @item collect @var{form} | |
2704 This clause collects the values of @var{form} into a list. Several | |
2705 examples of @code{collect} appear elsewhere in this manual. | |
2706 | |
2707 The word @code{collecting} is a synonym for @code{collect}, and | |
2708 likewise for the other accumulation clauses. | |
2709 | |
2710 @item append @var{form} | |
2711 This clause collects lists of values into a result list using | |
2712 @code{append}. | |
2713 | |
2714 @item nconc @var{form} | |
2715 This clause collects lists of values into a result list by | |
2716 destructively modifying the lists rather than copying them. | |
2717 | |
2718 @item concat @var{form} | |
2719 This clause concatenates the values of the specified @var{form} | |
2720 into a string. (It and the following clause are extensions to | |
2721 standard Common Lisp.) | |
2722 | |
2723 @item vconcat @var{form} | |
2724 This clause concatenates the values of the specified @var{form} | |
2725 into a vector. | |
2726 | |
2727 @item count @var{form} | |
2728 This clause counts the number of times the specified @var{form} | |
2729 evaluates to a non-@code{nil} value. | |
2730 | |
2731 @item sum @var{form} | |
2732 This clause accumulates the sum of the values of the specified | |
2733 @var{form}, which must evaluate to a number. | |
2734 | |
2735 @item maximize @var{form} | |
2736 This clause accumulates the maximum value of the specified @var{form}, | |
2737 which must evaluate to a number. The return value is undefined if | |
2738 @code{maximize} is executed zero times. | |
2739 | |
2740 @item minimize @var{form} | |
2741 This clause accumulates the minimum value of the specified @var{form}. | |
2742 @end table | |
2743 | |
2744 Accumulation clauses can be followed by @samp{into @var{var}} to | |
2745 cause the data to be collected into variable @var{var} (which is | |
2746 automatically @code{let}-bound during the loop) rather than an | |
2747 unnamed temporary variable. Also, @code{into} accumulations do | |
2748 not automatically imply a return value. The loop must use some | |
2749 explicit mechanism, such as @code{finally return}, to return | |
2750 the accumulated result. | |
2751 | |
2752 It is legal for several accumulation clauses of the same type to | |
2753 accumulate into the same place. From Steele: | |
2754 | |
2755 @example | |
2756 (loop for name in '(fred sue alice joe june) | |
2757 for kids in '((bob ken) () () (kris sunshine) ()) | |
2758 collect name | |
2759 append kids) | |
2760 @result{} (fred bob ken sue alice joe kris sunshine june) | |
2761 @end example | |
2762 | |
31572
384603523f41
Remove obsolescences, especially hash tables and refs to Emacs 19.
Dave Love <fx@gnu.org>
parents:
30009
diff
changeset
|
2763 @node Other Clauses, , Accumulation Clauses, Loop Facility |
25829 | 2764 @subsection Other Clauses |
2765 | |
2766 @noindent | |
2767 This section describes the remaining loop clauses. | |
2768 | |
2769 @table @code | |
2770 @item with @var{var} = @var{value} | |
2771 This clause binds a variable to a value around the loop, but | |
2772 otherwise leaves the variable alone during the loop. The following | |
2773 loops are basically equivalent: | |
2774 | |
2775 @example | |
2776 (loop with x = 17 do ...) | |
2777 (let ((x 17)) (loop do ...)) | |
2778 (loop for x = 17 then x do ...) | |
2779 @end example | |
2780 | |
2781 Naturally, the variable @var{var} might be used for some purpose | |
2782 in the rest of the loop. For example: | |
2783 | |
2784 @example | |
2785 (loop for x in my-list with res = nil do (push x res) | |
2786 finally return res) | |
2787 @end example | |
2788 | |
2789 This loop inserts the elements of @code{my-list} at the front of | |
2790 a new list being accumulated in @code{res}, then returns the | |
2791 list @code{res} at the end of the loop. The effect is similar | |
2792 to that of a @code{collect} clause, but the list gets reversed | |
2793 by virtue of the fact that elements are being pushed onto the | |
2794 front of @code{res} rather than the end. | |
2795 | |
2796 If you omit the @code{=} term, the variable is initialized to | |
2797 @code{nil}. (Thus the @samp{= nil} in the above example is | |
2798 unnecessary.) | |
2799 | |
2800 Bindings made by @code{with} are sequential by default, as if | |
2801 by @code{let*}. Just like @code{for} clauses, @code{with} clauses | |
2802 can be linked with @code{and} to cause the bindings to be made by | |
2803 @code{let} instead. | |
2804 | |
2805 @item if @var{condition} @var{clause} | |
2806 This clause executes the following loop clause only if the specified | |
2807 condition is true. The following @var{clause} should be an accumulation, | |
2808 @code{do}, @code{return}, @code{if}, or @code{unless} clause. | |
2809 Several clauses may be linked by separating them with @code{and}. | |
2810 These clauses may be followed by @code{else} and a clause or clauses | |
2811 to execute if the condition was false. The whole construct may | |
2812 optionally be followed by the word @code{end} (which may be used to | |
2813 disambiguate an @code{else} or @code{and} in a nested @code{if}). | |
2814 | |
2815 The actual non-@code{nil} value of the condition form is available | |
2816 by the name @code{it} in the ``then'' part. For example: | |
2817 | |
2818 @example | |
2819 (setq funny-numbers '(6 13 -1)) | |
2820 @result{} (6 13 -1) | |
2821 (loop for x below 10 | |
2822 if (oddp x) | |
2823 collect x into odds | |
2824 and if (memq x funny-numbers) return (cdr it) end | |
2825 else | |
2826 collect x into evens | |
2827 finally return (vector odds evens)) | |
2828 @result{} [(1 3 5 7 9) (0 2 4 6 8)] | |
2829 (setq funny-numbers '(6 7 13 -1)) | |
2830 @result{} (6 7 13 -1) | |
2831 (loop <@r{same thing again}>) | |
2832 @result{} (13 -1) | |
2833 @end example | |
2834 | |
2835 Note the use of @code{and} to put two clauses into the ``then'' | |
2836 part, one of which is itself an @code{if} clause. Note also that | |
2837 @code{end}, while normally optional, was necessary here to make | |
2838 it clear that the @code{else} refers to the outermost @code{if} | |
2839 clause. In the first case, the loop returns a vector of lists | |
2840 of the odd and even values of @var{x}. In the second case, the | |
2841 odd number 7 is one of the @code{funny-numbers} so the loop | |
2842 returns early; the actual returned value is based on the result | |
2843 of the @code{memq} call. | |
2844 | |
2845 @item when @var{condition} @var{clause} | |
2846 This clause is just a synonym for @code{if}. | |
2847 | |
2848 @item unless @var{condition} @var{clause} | |
2849 The @code{unless} clause is just like @code{if} except that the | |
2850 sense of the condition is reversed. | |
2851 | |
2852 @item named @var{name} | |
2853 This clause gives a name other than @code{nil} to the implicit | |
2854 block surrounding the loop. The @var{name} is the symbol to be | |
2855 used as the block name. | |
2856 | |
2857 @item initially [do] @var{forms}... | |
2858 This keyword introduces one or more Lisp forms which will be | |
2859 executed before the loop itself begins (but after any variables | |
2860 requested by @code{for} or @code{with} have been bound to their | |
2861 initial values). @code{initially} clauses can appear anywhere; | |
2862 if there are several, they are executed in the order they appear | |
2863 in the loop. The keyword @code{do} is optional. | |
2864 | |
2865 @item finally [do] @var{forms}... | |
2866 This introduces Lisp forms which will be executed after the loop | |
2867 finishes (say, on request of a @code{for} or @code{while}). | |
2868 @code{initially} and @code{finally} clauses may appear anywhere | |
2869 in the loop construct, but they are executed (in the specified | |
2870 order) at the beginning or end, respectively, of the loop. | |
2871 | |
2872 @item finally return @var{form} | |
2873 This says that @var{form} should be executed after the loop | |
2874 is done to obtain a return value. (Without this, or some other | |
2875 clause like @code{collect} or @code{return}, the loop will simply | |
2876 return @code{nil}.) Variables bound by @code{for}, @code{with}, | |
2877 or @code{into} will still contain their final values when @var{form} | |
2878 is executed. | |
2879 | |
2880 @item do @var{forms}... | |
2881 The word @code{do} may be followed by any number of Lisp expressions | |
2882 which are executed as an implicit @code{progn} in the body of the | |
2883 loop. Many of the examples in this section illustrate the use of | |
2884 @code{do}. | |
2885 | |
2886 @item return @var{form} | |
2887 This clause causes the loop to return immediately. The following | |
2888 Lisp form is evaluated to give the return value of the @code{loop} | |
2889 form. The @code{finally} clauses, if any, are not executed. | |
2890 Of course, @code{return} is generally used inside an @code{if} or | |
2891 @code{unless}, as its use in a top-level loop clause would mean | |
2892 the loop would never get to ``loop'' more than once. | |
2893 | |
2894 The clause @samp{return @var{form}} is equivalent to | |
2895 @samp{do (return @var{form})} (or @code{return-from} if the loop | |
2896 was named). The @code{return} clause is implemented a bit more | |
2897 efficiently, though. | |
2898 @end table | |
2899 | |
2900 While there is no high-level way to add user extensions to @code{loop} | |
2901 (comparable to @code{defsetf} for @code{setf}, say), this package | |
2902 does offer two properties called @code{cl-loop-handler} and | |
2903 @code{cl-loop-for-handler} which are functions to be called when | |
2904 a given symbol is encountered as a top-level loop clause or | |
2905 @code{for} clause, respectively. Consult the source code in | |
2906 file @file{cl-macs.el} for details. | |
2907 | |
2908 This package's @code{loop} macro is compatible with that of Common | |
2909 Lisp, except that a few features are not implemented: @code{loop-finish} | |
2910 and data-type specifiers. Naturally, the @code{for} clauses which | |
2911 iterate over keymaps, overlays, intervals, frames, windows, and | |
2912 buffers are Emacs-specific extensions. | |
2913 | |
31572
384603523f41
Remove obsolescences, especially hash tables and refs to Emacs 19.
Dave Love <fx@gnu.org>
parents:
30009
diff
changeset
|
2914 @node Multiple Values, , Loop Facility, Control Structure |
25829 | 2915 @section Multiple Values |
2916 | |
2917 @noindent | |
2918 Common Lisp functions can return zero or more results. Emacs Lisp | |
2919 functions, by contrast, always return exactly one result. This | |
2920 package makes no attempt to emulate Common Lisp multiple return | |
2921 values; Emacs versions of Common Lisp functions that return more | |
2922 than one value either return just the first value (as in | |
2923 @code{compiler-macroexpand}) or return a list of values (as in | |
2924 @code{get-setf-method}). This package @emph{does} define placeholders | |
2925 for the Common Lisp functions that work with multiple values, but | |
2926 in Emacs Lisp these functions simply operate on lists instead. | |
2927 The @code{values} form, for example, is a synonym for @code{list} | |
2928 in Emacs. | |
2929 | |
2930 @defspec multiple-value-bind (var@dots{}) values-form forms@dots{} | |
2931 This form evaluates @var{values-form}, which must return a list of | |
2932 values. It then binds the @var{var}s to these respective values, | |
2933 as if by @code{let}, and then executes the body @var{forms}. | |
2934 If there are more @var{var}s than values, the extra @var{var}s | |
2935 are bound to @code{nil}. If there are fewer @var{var}s than | |
2936 values, the excess values are ignored. | |
2937 @end defspec | |
2938 | |
2939 @defspec multiple-value-setq (var@dots{}) form | |
2940 This form evaluates @var{form}, which must return a list of values. | |
2941 It then sets the @var{var}s to these respective values, as if by | |
2942 @code{setq}. Extra @var{var}s or values are treated the same as | |
2943 in @code{multiple-value-bind}. | |
2944 @end defspec | |
2945 | |
2946 The older Quiroz package attempted a more faithful (but still | |
2947 imperfect) emulation of Common Lisp multiple values. The old | |
2948 method ``usually'' simulated true multiple values quite well, | |
2949 but under certain circumstances would leave spurious return | |
2950 values in memory where a later, unrelated @code{multiple-value-bind} | |
2951 form would see them. | |
2952 | |
2953 Since a perfect emulation is not feasible in Emacs Lisp, this | |
2954 package opts to keep it as simple and predictable as possible. | |
2955 | |
2956 @node Macros, Declarations, Control Structure, Top | |
2957 @chapter Macros | |
2958 | |
2959 @noindent | |
2960 This package implements the various Common Lisp features of | |
2961 @code{defmacro}, such as destructuring, @code{&environment}, | |
2962 and @code{&body}. Top-level @code{&whole} is not implemented | |
2963 for @code{defmacro} due to technical difficulties. | |
2964 @xref{Argument Lists}. | |
2965 | |
2966 Destructuring is made available to the user by way of the | |
2967 following macro: | |
2968 | |
2969 @defspec destructuring-bind arglist expr forms@dots{} | |
2970 This macro expands to code which executes @var{forms}, with | |
2971 the variables in @var{arglist} bound to the list of values | |
2972 returned by @var{expr}. The @var{arglist} can include all | |
2973 the features allowed for @code{defmacro} argument lists, | |
2974 including destructuring. (The @code{&environment} keyword | |
2975 is not allowed.) The macro expansion will signal an error | |
2976 if @var{expr} returns a list of the wrong number of arguments | |
2977 or with incorrect keyword arguments. | |
2978 @end defspec | |
2979 | |
2980 This package also includes the Common Lisp @code{define-compiler-macro} | |
2981 facility, which allows you to define compile-time expansions and | |
2982 optimizations for your functions. | |
2983 | |
2984 @defspec define-compiler-macro name arglist forms@dots{} | |
2985 This form is similar to @code{defmacro}, except that it only expands | |
2986 calls to @var{name} at compile-time; calls processed by the Lisp | |
2987 interpreter are not expanded, nor are they expanded by the | |
2988 @code{macroexpand} function. | |
2989 | |
2990 The argument list may begin with a @code{&whole} keyword and a | |
2991 variable. This variable is bound to the macro-call form itself, | |
2992 i.e., to a list of the form @samp{(@var{name} @var{args}@dots{})}. | |
2993 If the macro expander returns this form unchanged, then the | |
2994 compiler treats it as a normal function call. This allows | |
2995 compiler macros to work as optimizers for special cases of a | |
2996 function, leaving complicated cases alone. | |
2997 | |
2998 For example, here is a simplified version of a definition that | |
2999 appears as a standard part of this package: | |
3000 | |
3001 @example | |
3002 (define-compiler-macro member* (&whole form a list &rest keys) | |
3003 (if (and (null keys) | |
3004 (eq (car-safe a) 'quote) | |
3005 (not (floatp-safe (cadr a)))) | |
3006 (list 'memq a list) | |
3007 form)) | |
3008 @end example | |
3009 | |
3010 @noindent | |
3011 This definition causes @code{(member* @var{a} @var{list})} to change | |
3012 to a call to the faster @code{memq} in the common case where @var{a} | |
3013 is a non-floating-point constant; if @var{a} is anything else, or | |
3014 if there are any keyword arguments in the call, then the original | |
3015 @code{member*} call is left intact. (The actual compiler macro | |
3016 for @code{member*} optimizes a number of other cases, including | |
3017 common @code{:test} predicates.) | |
3018 @end defspec | |
3019 | |
3020 @defun compiler-macroexpand form | |
3021 This function is analogous to @code{macroexpand}, except that it | |
3022 expands compiler macros rather than regular macros. It returns | |
3023 @var{form} unchanged if it is not a call to a function for which | |
3024 a compiler macro has been defined, or if that compiler macro | |
3025 decided to punt by returning its @code{&whole} argument. Like | |
3026 @code{macroexpand}, it expands repeatedly until it reaches a form | |
3027 for which no further expansion is possible. | |
3028 @end defun | |
3029 | |
3030 @xref{Macro Bindings}, for descriptions of the @code{macrolet} | |
3031 and @code{symbol-macrolet} forms for making ``local'' macro | |
3032 definitions. | |
3033 | |
3034 @node Declarations, Symbols, Macros, Top | |
3035 @chapter Declarations | |
3036 | |
3037 @noindent | |
3038 Common Lisp includes a complex and powerful ``declaration'' | |
3039 mechanism that allows you to give the compiler special hints | |
3040 about the types of data that will be stored in particular variables, | |
3041 and about the ways those variables and functions will be used. This | |
3042 package defines versions of all the Common Lisp declaration forms: | |
3043 @code{declare}, @code{locally}, @code{proclaim}, @code{declaim}, | |
3044 and @code{the}. | |
3045 | |
3046 Most of the Common Lisp declarations are not currently useful in | |
3047 Emacs Lisp, as the byte-code system provides little opportunity | |
3048 to benefit from type information, and @code{special} declarations | |
3049 are redundant in a fully dynamically-scoped Lisp. A few | |
31572
384603523f41
Remove obsolescences, especially hash tables and refs to Emacs 19.
Dave Love <fx@gnu.org>
parents:
30009
diff
changeset
|
3050 declarations are meaningful when the optimizing byte |
25829 | 3051 compiler is being used, however. Under the earlier non-optimizing |
3052 compiler, these declarations will effectively be ignored. | |
3053 | |
3054 @defun proclaim decl-spec | |
3055 This function records a ``global'' declaration specified by | |
3056 @var{decl-spec}. Since @code{proclaim} is a function, @var{decl-spec} | |
3057 is evaluated and thus should normally be quoted. | |
3058 @end defun | |
3059 | |
3060 @defspec declaim decl-specs@dots{} | |
3061 This macro is like @code{proclaim}, except that it takes any number | |
3062 of @var{decl-spec} arguments, and the arguments are unevaluated and | |
3063 unquoted. The @code{declaim} macro also puts an @code{(eval-when | |
3064 (compile load eval) ...)} around the declarations so that they will | |
3065 be registered at compile-time as well as at run-time. (This is vital, | |
3066 since normally the declarations are meant to influence the way the | |
3067 compiler treats the rest of the file that contains the @code{declaim} | |
3068 form.) | |
3069 @end defspec | |
3070 | |
3071 @defspec declare decl-specs@dots{} | |
3072 This macro is used to make declarations within functions and other | |
3073 code. Common Lisp allows declarations in various locations, generally | |
3074 at the beginning of any of the many ``implicit @code{progn}s'' | |
3075 throughout Lisp syntax, such as function bodies, @code{let} bodies, | |
3076 etc. Currently the only declaration understood by @code{declare} | |
3077 is @code{special}. | |
3078 @end defspec | |
3079 | |
3080 @defspec locally declarations@dots{} forms@dots{} | |
3081 In this package, @code{locally} is no different from @code{progn}. | |
3082 @end defspec | |
3083 | |
3084 @defspec the type form | |
3085 Type information provided by @code{the} is ignored in this package; | |
3086 in other words, @code{(the @var{type} @var{form})} is equivalent | |
3087 to @var{form}. Future versions of the optimizing byte-compiler may | |
3088 make use of this information. | |
3089 | |
3090 For example, @code{mapcar} can map over both lists and arrays. It is | |
3091 hard for the compiler to expand @code{mapcar} into an in-line loop | |
3092 unless it knows whether the sequence will be a list or an array ahead | |
3093 of time. With @code{(mapcar 'car (the vector foo))}, a future | |
3094 compiler would have enough information to expand the loop in-line. | |
3095 For now, Emacs Lisp will treat the above code as exactly equivalent | |
3096 to @code{(mapcar 'car foo)}. | |
3097 @end defspec | |
3098 | |
3099 Each @var{decl-spec} in a @code{proclaim}, @code{declaim}, or | |
3100 @code{declare} should be a list beginning with a symbol that says | |
3101 what kind of declaration it is. This package currently understands | |
3102 @code{special}, @code{inline}, @code{notinline}, @code{optimize}, | |
3103 and @code{warn} declarations. (The @code{warn} declaration is an | |
3104 extension of standard Common Lisp.) Other Common Lisp declarations, | |
3105 such as @code{type} and @code{ftype}, are silently ignored. | |
3106 | |
3107 @table @code | |
3108 @item special | |
3109 Since all variables in Emacs Lisp are ``special'' (in the Common | |
3110 Lisp sense), @code{special} declarations are only advisory. They | |
3111 simply tell the optimizing byte compiler that the specified | |
3112 variables are intentionally being referred to without being | |
3113 bound in the body of the function. The compiler normally emits | |
3114 warnings for such references, since they could be typographical | |
3115 errors for references to local variables. | |
3116 | |
3117 The declaration @code{(declare (special @var{var1} @var{var2}))} is | |
3118 equivalent to @code{(defvar @var{var1}) (defvar @var{var2})} in the | |
3119 optimizing compiler, or to nothing at all in older compilers (which | |
3120 do not warn for non-local references). | |
3121 | |
3122 In top-level contexts, it is generally better to write | |
3123 @code{(defvar @var{var})} than @code{(declaim (special @var{var}))}, | |
3124 since @code{defvar} makes your intentions clearer. But the older | |
3125 byte compilers can not handle @code{defvar}s appearing inside of | |
3126 functions, while @code{(declare (special @var{var}))} takes care | |
3127 to work correctly with all compilers. | |
3128 | |
3129 @item inline | |
3130 The @code{inline} @var{decl-spec} lists one or more functions | |
3131 whose bodies should be expanded ``in-line'' into calling functions | |
3132 whenever the compiler is able to arrange for it. For example, | |
3133 the Common Lisp function @code{cadr} is declared @code{inline} | |
3134 by this package so that the form @code{(cadr @var{x})} will | |
3135 expand directly into @code{(car (cdr @var{x}))} when it is called | |
3136 in user functions, for a savings of one (relatively expensive) | |
3137 function call. | |
3138 | |
3139 The following declarations are all equivalent. Note that the | |
3140 @code{defsubst} form is a convenient way to define a function | |
31572
384603523f41
Remove obsolescences, especially hash tables and refs to Emacs 19.
Dave Love <fx@gnu.org>
parents:
30009
diff
changeset
|
3141 and declare it inline all at once. |
25829 | 3142 |
3143 @example | |
3144 (declaim (inline foo bar)) | |
3145 (eval-when (compile load eval) (proclaim '(inline foo bar))) | |
31572
384603523f41
Remove obsolescences, especially hash tables and refs to Emacs 19.
Dave Love <fx@gnu.org>
parents:
30009
diff
changeset
|
3146 (defsubst foo (...) ...) ; instead of defun |
25829 | 3147 @end example |
3148 | |
3149 @strong{Note:} This declaration remains in effect after the | |
3150 containing source file is done. It is correct to use it to | |
3151 request that a function you have defined should be inlined, | |
3152 but it is impolite to use it to request inlining of an external | |
3153 function. | |
3154 | |
3155 In Common Lisp, it is possible to use @code{(declare (inline @dots{}))} | |
3156 before a particular call to a function to cause just that call to | |
3157 be inlined; the current byte compilers provide no way to implement | |
3158 this, so @code{(declare (inline @dots{}))} is currently ignored by | |
3159 this package. | |
3160 | |
3161 @item notinline | |
3162 The @code{notinline} declaration lists functions which should | |
3163 not be inlined after all; it cancels a previous @code{inline} | |
3164 declaration. | |
3165 | |
3166 @item optimize | |
3167 This declaration controls how much optimization is performed by | |
3168 the compiler. Naturally, it is ignored by the earlier non-optimizing | |
3169 compilers. | |
3170 | |
3171 The word @code{optimize} is followed by any number of lists like | |
3172 @code{(speed 3)} or @code{(safety 2)}. Common Lisp defines several | |
3173 optimization ``qualities''; this package ignores all but @code{speed} | |
3174 and @code{safety}. The value of a quality should be an integer from | |
3175 0 to 3, with 0 meaning ``unimportant'' and 3 meaning ``very important.'' | |
3176 The default level for both qualities is 1. | |
3177 | |
31572
384603523f41
Remove obsolescences, especially hash tables and refs to Emacs 19.
Dave Love <fx@gnu.org>
parents:
30009
diff
changeset
|
3178 In this package, with the optimizing compiler, the |
25829 | 3179 @code{speed} quality is tied to the @code{byte-compile-optimize} |
3180 flag, which is set to @code{nil} for @code{(speed 0)} and to | |
3181 @code{t} for higher settings; and the @code{safety} quality is | |
3182 tied to the @code{byte-compile-delete-errors} flag, which is | |
3183 set to @code{t} for @code{(safety 3)} and to @code{nil} for all | |
3184 lower settings. (The latter flag controls whether the compiler | |
3185 is allowed to optimize out code whose only side-effect could | |
3186 be to signal an error, e.g., rewriting @code{(progn foo bar)} to | |
3187 @code{bar} when it is not known whether @code{foo} will be bound | |
3188 at run-time.) | |
3189 | |
3190 Note that even compiling with @code{(safety 0)}, the Emacs | |
3191 byte-code system provides sufficient checking to prevent real | |
3192 harm from being done. For example, barring serious bugs in | |
3193 Emacs itself, Emacs will not crash with a segmentation fault | |
3194 just because of an error in a fully-optimized Lisp program. | |
3195 | |
3196 The @code{optimize} declaration is normally used in a top-level | |
3197 @code{proclaim} or @code{declaim} in a file; Common Lisp allows | |
3198 it to be used with @code{declare} to set the level of optimization | |
3199 locally for a given form, but this will not work correctly with the | |
3200 current version of the optimizing compiler. (The @code{declare} | |
3201 will set the new optimization level, but that level will not | |
3202 automatically be unset after the enclosing form is done.) | |
3203 | |
3204 @item warn | |
3205 This declaration controls what sorts of warnings are generated | |
3206 by the byte compiler. Again, only the optimizing compiler | |
3207 generates warnings. The word @code{warn} is followed by any | |
3208 number of ``warning qualities,'' similar in form to optimization | |
3209 qualities. The currently supported warning types are | |
3210 @code{redefine}, @code{callargs}, @code{unresolved}, and | |
3211 @code{free-vars}; in the current system, a value of 0 will | |
3212 disable these warnings and any higher value will enable them. | |
3213 See the documentation for the optimizing byte compiler for details. | |
3214 @end table | |
3215 | |
3216 @node Symbols, Numbers, Declarations, Top | |
3217 @chapter Symbols | |
3218 | |
3219 @noindent | |
3220 This package defines several symbol-related features that were | |
3221 missing from Emacs Lisp. | |
3222 | |
3223 @menu | |
3224 * Property Lists:: `get*', `remprop', `getf', `remf' | |
3225 * Creating Symbols:: `gensym', `gentemp' | |
3226 @end menu | |
3227 | |
3228 @node Property Lists, Creating Symbols, Symbols, Symbols | |
3229 @section Property Lists | |
3230 | |
3231 @noindent | |
3232 These functions augment the standard Emacs Lisp functions @code{get} | |
3233 and @code{put} for operating on properties attached to symbols. | |
3234 There are also functions for working with property lists as | |
3235 first-class data structures not attached to particular symbols. | |
3236 | |
3237 @defun get* symbol property &optional default | |
3238 This function is like @code{get}, except that if the property is | |
3239 not found, the @var{default} argument provides the return value. | |
3240 (The Emacs Lisp @code{get} function always uses @code{nil} as | |
3241 the default; this package's @code{get*} is equivalent to Common | |
3242 Lisp's @code{get}.) | |
3243 | |
3244 The @code{get*} function is @code{setf}-able; when used in this | |
3245 fashion, the @var{default} argument is allowed but ignored. | |
3246 @end defun | |
3247 | |
3248 @defun remprop symbol property | |
3249 This function removes the entry for @var{property} from the property | |
3250 list of @var{symbol}. It returns a true value if the property was | |
3251 indeed found and removed, or @code{nil} if there was no such property. | |
3252 (This function was probably omitted from Emacs originally because, | |
3253 since @code{get} did not allow a @var{default}, it was very difficult | |
3254 to distinguish between a missing property and a property whose value | |
3255 was @code{nil}; thus, setting a property to @code{nil} was close | |
3256 enough to @code{remprop} for most purposes.) | |
3257 @end defun | |
3258 | |
3259 @defun getf place property &optional default | |
3260 This function scans the list @var{place} as if it were a property | |
3261 list, i.e., a list of alternating property names and values. If | |
3262 an even-numbered element of @var{place} is found which is @code{eq} | |
3263 to @var{property}, the following odd-numbered element is returned. | |
3264 Otherwise, @var{default} is returned (or @code{nil} if no default | |
3265 is given). | |
3266 | |
3267 In particular, | |
3268 | |
3269 @example | |
3270 (get sym prop) @equiv{} (getf (symbol-plist sym) prop) | |
3271 @end example | |
3272 | |
3273 It is legal to use @code{getf} as a @code{setf} place, in which case | |
3274 its @var{place} argument must itself be a legal @code{setf} place. | |
3275 The @var{default} argument, if any, is ignored in this context. | |
3276 The effect is to change (via @code{setcar}) the value cell in the | |
3277 list that corresponds to @var{property}, or to cons a new property-value | |
3278 pair onto the list if the property is not yet present. | |
3279 | |
3280 @example | |
3281 (put sym prop val) @equiv{} (setf (getf (symbol-plist sym) prop) val) | |
3282 @end example | |
3283 | |
3284 The @code{get} and @code{get*} functions are also @code{setf}-able. | |
3285 The fact that @code{default} is ignored can sometimes be useful: | |
3286 | |
3287 @example | |
3288 (incf (get* 'foo 'usage-count 0)) | |
3289 @end example | |
3290 | |
3291 Here, symbol @code{foo}'s @code{usage-count} property is incremented | |
3292 if it exists, or set to 1 (an incremented 0) otherwise. | |
3293 | |
3294 When not used as a @code{setf} form, @code{getf} is just a regular | |
3295 function and its @var{place} argument can actually be any Lisp | |
3296 expression. | |
3297 @end defun | |
3298 | |
3299 @defspec remf place property | |
3300 This macro removes the property-value pair for @var{property} from | |
3301 the property list stored at @var{place}, which is any @code{setf}-able | |
3302 place expression. It returns true if the property was found. Note | |
3303 that if @var{property} happens to be first on the list, this will | |
3304 effectively do a @code{(setf @var{place} (cddr @var{place}))}, | |
3305 whereas if it occurs later, this simply uses @code{setcdr} to splice | |
3306 out the property and value cells. | |
3307 @end defspec | |
3308 | |
3309 @iftex | |
3310 @secno=2 | |
3311 @end iftex | |
3312 | |
31572
384603523f41
Remove obsolescences, especially hash tables and refs to Emacs 19.
Dave Love <fx@gnu.org>
parents:
30009
diff
changeset
|
3313 @node Creating Symbols, , Property Lists, Symbols |
25829 | 3314 @section Creating Symbols |
3315 | |
3316 @noindent | |
3317 These functions create unique symbols, typically for use as | |
3318 temporary variables. | |
3319 | |
3320 @defun gensym &optional x | |
3321 This function creates a new, uninterned symbol (using @code{make-symbol}) | |
3322 with a unique name. (The name of an uninterned symbol is relevant | |
3323 only if the symbol is printed.) By default, the name is generated | |
3324 from an increasing sequence of numbers, @samp{G1000}, @samp{G1001}, | |
3325 @samp{G1002}, etc. If the optional argument @var{x} is a string, that | |
3326 string is used as a prefix instead of @samp{G}. Uninterned symbols | |
3327 are used in macro expansions for temporary variables, to ensure that | |
3328 their names will not conflict with ``real'' variables in the user's | |
3329 code. | |
3330 @end defun | |
3331 | |
3332 @defvar *gensym-counter* | |
3333 This variable holds the counter used to generate @code{gensym} names. | |
3334 It is incremented after each use by @code{gensym}. In Common Lisp | |
3335 this is initialized with 0, but this package initializes it with a | |
3336 random (time-dependent) value to avoid trouble when two files that | |
3337 each used @code{gensym} in their compilation are loaded together. | |
3338 (Uninterned symbols become interned when the compiler writes them | |
3339 out to a file and the Emacs loader loads them, so their names have to | |
3340 be treated a bit more carefully than in Common Lisp where uninterned | |
3341 symbols remain uninterned after loading.) | |
3342 @end defvar | |
3343 | |
3344 @defun gentemp &optional x | |
3345 This function is like @code{gensym}, except that it produces a new | |
3346 @emph{interned} symbol. If the symbol that is generated already | |
3347 exists, the function keeps incrementing the counter and trying | |
3348 again until a new symbol is generated. | |
3349 @end defun | |
3350 | |
3351 The Quiroz @file{cl.el} package also defined a @code{defkeyword} | |
3352 form for creating self-quoting keyword symbols. This package | |
3353 automatically creates all keywords that are called for by | |
3354 @code{&key} argument specifiers, and discourages the use of | |
3355 keywords as data unrelated to keyword arguments, so the | |
3356 @code{defkeyword} form has been discontinued. | |
3357 | |
3358 @iftex | |
3359 @chapno=11 | |
3360 @end iftex | |
3361 | |
3362 @node Numbers, Sequences, Symbols, Top | |
3363 @chapter Numbers | |
3364 | |
3365 @noindent | |
3366 This section defines a few simple Common Lisp operations on numbers | |
3367 which were left out of Emacs Lisp. | |
3368 | |
3369 @menu | |
3370 * Predicates on Numbers:: `plusp', `oddp', `floatp-safe', etc. | |
28039
c3a446101beb
Remove references to rassoc, delete, expt.
Dave Love <fx@gnu.org>
parents:
27511
diff
changeset
|
3371 * Numerical Functions:: `abs', `floor*', etc. |
25829 | 3372 * Random Numbers:: `random*', `make-random-state' |
42136
5b63c25571a8
Remove most-positive-fixnum, most-negative-fixnum.
Dave Love <fx@gnu.org>
parents:
40590
diff
changeset
|
3373 * Implementation Parameters:: `most-positive-float' |
25829 | 3374 @end menu |
3375 | |
3376 @iftex | |
3377 @secno=1 | |
3378 @end iftex | |
3379 | |
3380 @node Predicates on Numbers, Numerical Functions, Numbers, Numbers | |
3381 @section Predicates on Numbers | |
3382 | |
3383 @noindent | |
3384 These functions return @code{t} if the specified condition is | |
3385 true of the numerical argument, or @code{nil} otherwise. | |
3386 | |
3387 @defun plusp number | |
3388 This predicate tests whether @var{number} is positive. It is an | |
3389 error if the argument is not a number. | |
3390 @end defun | |
3391 | |
3392 @defun minusp number | |
3393 This predicate tests whether @var{number} is negative. It is an | |
3394 error if the argument is not a number. | |
3395 @end defun | |
3396 | |
3397 @defun oddp integer | |
3398 This predicate tests whether @var{integer} is odd. It is an | |
3399 error if the argument is not an integer. | |
3400 @end defun | |
3401 | |
3402 @defun evenp integer | |
3403 This predicate tests whether @var{integer} is even. It is an | |
3404 error if the argument is not an integer. | |
3405 @end defun | |
3406 | |
3407 @defun floatp-safe object | |
3408 This predicate tests whether @var{object} is a floating-point | |
3409 number. On systems that support floating-point, this is equivalent | |
3410 to @code{floatp}. On other systems, this always returns @code{nil}. | |
3411 @end defun | |
3412 | |
3413 @iftex | |
3414 @secno=3 | |
3415 @end iftex | |
3416 | |
3417 @node Numerical Functions, Random Numbers, Predicates on Numbers, Numbers | |
3418 @section Numerical Functions | |
3419 | |
3420 @noindent | |
3421 These functions perform various arithmetic operations on numbers. | |
3422 | |
3423 @defun gcd &rest integers | |
3424 This function returns the Greatest Common Divisor of the arguments. | |
3425 For one argument, it returns the absolute value of that argument. | |
3426 For zero arguments, it returns zero. | |
3427 @end defun | |
3428 | |
3429 @defun lcm &rest integers | |
3430 This function returns the Least Common Multiple of the arguments. | |
3431 For one argument, it returns the absolute value of that argument. | |
3432 For zero arguments, it returns one. | |
3433 @end defun | |
3434 | |
3435 @defun isqrt integer | |
3436 This function computes the ``integer square root'' of its integer | |
3437 argument, i.e., the greatest integer less than or equal to the true | |
3438 square root of the argument. | |
3439 @end defun | |
3440 | |
3441 @defun floor* number &optional divisor | |
3442 This function implements the Common Lisp @code{floor} function. | |
3443 It is called @code{floor*} to avoid name conflicts with the | |
31572
384603523f41
Remove obsolescences, especially hash tables and refs to Emacs 19.
Dave Love <fx@gnu.org>
parents:
30009
diff
changeset
|
3444 simpler @code{floor} function built-in to Emacs. |
25829 | 3445 |
3446 With one argument, @code{floor*} returns a list of two numbers: | |
3447 The argument rounded down (toward minus infinity) to an integer, | |
3448 and the ``remainder'' which would have to be added back to the | |
3449 first return value to yield the argument again. If the argument | |
3450 is an integer @var{x}, the result is always the list @code{(@var{x} 0)}. | |
31572
384603523f41
Remove obsolescences, especially hash tables and refs to Emacs 19.
Dave Love <fx@gnu.org>
parents:
30009
diff
changeset
|
3451 If the argument is a floating-point number, the first |
25829 | 3452 result is a Lisp integer and the second is a Lisp float between |
3453 0 (inclusive) and 1 (exclusive). | |
3454 | |
3455 With two arguments, @code{floor*} divides @var{number} by | |
3456 @var{divisor}, and returns the floor of the quotient and the | |
3457 corresponding remainder as a list of two numbers. If | |
3458 @code{(floor* @var{x} @var{y})} returns @code{(@var{q} @var{r})}, | |
3459 then @code{@var{q}*@var{y} + @var{r} = @var{x}}, with @var{r} | |
3460 between 0 (inclusive) and @var{r} (exclusive). Also, note | |
3461 that @code{(floor* @var{x})} is exactly equivalent to | |
3462 @code{(floor* @var{x} 1)}. | |
3463 | |
3464 This function is entirely compatible with Common Lisp's @code{floor} | |
3465 function, except that it returns the two results in a list since | |
3466 Emacs Lisp does not support multiple-valued functions. | |
3467 @end defun | |
3468 | |
3469 @defun ceiling* number &optional divisor | |
3470 This function implements the Common Lisp @code{ceiling} function, | |
3471 which is analogous to @code{floor} except that it rounds the | |
3472 argument or quotient of the arguments up toward plus infinity. | |
3473 The remainder will be between 0 and minus @var{r}. | |
3474 @end defun | |
3475 | |
3476 @defun truncate* number &optional divisor | |
3477 This function implements the Common Lisp @code{truncate} function, | |
3478 which is analogous to @code{floor} except that it rounds the | |
3479 argument or quotient of the arguments toward zero. Thus it is | |
3480 equivalent to @code{floor*} if the argument or quotient is | |
3481 positive, or to @code{ceiling*} otherwise. The remainder has | |
3482 the same sign as @var{number}. | |
3483 @end defun | |
3484 | |
3485 @defun round* number &optional divisor | |
3486 This function implements the Common Lisp @code{round} function, | |
3487 which is analogous to @code{floor} except that it rounds the | |
3488 argument or quotient of the arguments to the nearest integer. | |
3489 In the case of a tie (the argument or quotient is exactly | |
3490 halfway between two integers), it rounds to the even integer. | |
3491 @end defun | |
3492 | |
3493 @defun mod* number divisor | |
3494 This function returns the same value as the second return value | |
3495 of @code{floor}. | |
3496 @end defun | |
3497 | |
3498 @defun rem* number divisor | |
3499 This function returns the same value as the second return value | |
3500 of @code{truncate}. | |
3501 @end defun | |
3502 | |
3503 These definitions are compatible with those in the Quiroz | |
3504 @file{cl.el} package, except that this package appends @samp{*} | |
3505 to certain function names to avoid conflicts with existing | |
31572
384603523f41
Remove obsolescences, especially hash tables and refs to Emacs 19.
Dave Love <fx@gnu.org>
parents:
30009
diff
changeset
|
3506 Emacs functions, and that the mechanism for returning |
25829 | 3507 multiple values is different. |
3508 | |
3509 @iftex | |
3510 @secno=8 | |
3511 @end iftex | |
3512 | |
3513 @node Random Numbers, Implementation Parameters, Numerical Functions, Numbers | |
3514 @section Random Numbers | |
3515 | |
3516 @noindent | |
3517 This package also provides an implementation of the Common Lisp | |
3518 random number generator. It uses its own additive-congruential | |
3519 algorithm, which is much more likely to give statistically clean | |
3520 random numbers than the simple generators supplied by many | |
3521 operating systems. | |
3522 | |
3523 @defun random* number &optional state | |
3524 This function returns a random nonnegative number less than | |
3525 @var{number}, and of the same type (either integer or floating-point). | |
3526 The @var{state} argument should be a @code{random-state} object | |
3527 which holds the state of the random number generator. The | |
3528 function modifies this state object as a side effect. If | |
3529 @var{state} is omitted, it defaults to the variable | |
3530 @code{*random-state*}, which contains a pre-initialized | |
3531 @code{random-state} object. | |
3532 @end defun | |
3533 | |
3534 @defvar *random-state* | |
3535 This variable contains the system ``default'' @code{random-state} | |
3536 object, used for calls to @code{random*} that do not specify an | |
3537 alternative state object. Since any number of programs in the | |
3538 Emacs process may be accessing @code{*random-state*} in interleaved | |
3539 fashion, the sequence generated from this variable will be | |
3540 irreproducible for all intents and purposes. | |
3541 @end defvar | |
3542 | |
3543 @defun make-random-state &optional state | |
3544 This function creates or copies a @code{random-state} object. | |
3545 If @var{state} is omitted or @code{nil}, it returns a new copy of | |
3546 @code{*random-state*}. This is a copy in the sense that future | |
3547 sequences of calls to @code{(random* @var{n})} and | |
3548 @code{(random* @var{n} @var{s})} (where @var{s} is the new | |
3549 random-state object) will return identical sequences of random | |
3550 numbers. | |
3551 | |
3552 If @var{state} is a @code{random-state} object, this function | |
3553 returns a copy of that object. If @var{state} is @code{t}, this | |
3554 function returns a new @code{random-state} object seeded from the | |
3555 date and time. As an extension to Common Lisp, @var{state} may also | |
3556 be an integer in which case the new object is seeded from that | |
3557 integer; each different integer seed will result in a completely | |
3558 different sequence of random numbers. | |
3559 | |
3560 It is legal to print a @code{random-state} object to a buffer or | |
3561 file and later read it back with @code{read}. If a program wishes | |
3562 to use a sequence of pseudo-random numbers which can be reproduced | |
3563 later for debugging, it can call @code{(make-random-state t)} to | |
3564 get a new sequence, then print this sequence to a file. When the | |
3565 program is later rerun, it can read the original run's random-state | |
3566 from the file. | |
3567 @end defun | |
3568 | |
3569 @defun random-state-p object | |
3570 This predicate returns @code{t} if @var{object} is a | |
3571 @code{random-state} object, or @code{nil} otherwise. | |
3572 @end defun | |
3573 | |
31572
384603523f41
Remove obsolescences, especially hash tables and refs to Emacs 19.
Dave Love <fx@gnu.org>
parents:
30009
diff
changeset
|
3574 @node Implementation Parameters, , Random Numbers, Numbers |
25829 | 3575 @section Implementation Parameters |
3576 | |
3577 @noindent | |
3578 This package defines several useful constants having to with numbers. | |
3579 | |
3580 The following parameters have to do with floating-point numbers. | |
3581 This package determines their values by exercising the computer's | |
3582 floating-point arithmetic in various ways. Because this operation | |
3583 might be slow, the code for initializing them is kept in a separate | |
3584 function that must be called before the parameters can be used. | |
3585 | |
3586 @defun cl-float-limits | |
31572
384603523f41
Remove obsolescences, especially hash tables and refs to Emacs 19.
Dave Love <fx@gnu.org>
parents:
30009
diff
changeset
|
3587 This function makes sure that the Common Lisp floating-point parameters |
384603523f41
Remove obsolescences, especially hash tables and refs to Emacs 19.
Dave Love <fx@gnu.org>
parents:
30009
diff
changeset
|
3588 like @code{most-positive-float} have been initialized. Until it is |
384603523f41
Remove obsolescences, especially hash tables and refs to Emacs 19.
Dave Love <fx@gnu.org>
parents:
30009
diff
changeset
|
3589 called, these parameters will be @code{nil}. If this version of Emacs |
384603523f41
Remove obsolescences, especially hash tables and refs to Emacs 19.
Dave Love <fx@gnu.org>
parents:
30009
diff
changeset
|
3590 does not support floats, the parameters will remain @code{nil}. If the |
384603523f41
Remove obsolescences, especially hash tables and refs to Emacs 19.
Dave Love <fx@gnu.org>
parents:
30009
diff
changeset
|
3591 parameters have already been initialized, the function returns |
384603523f41
Remove obsolescences, especially hash tables and refs to Emacs 19.
Dave Love <fx@gnu.org>
parents:
30009
diff
changeset
|
3592 immediately. |
25829 | 3593 |
3594 The algorithm makes assumptions that will be valid for most modern | |
3595 machines, but will fail if the machine's arithmetic is extremely | |
3596 unusual, e.g., decimal. | |
3597 @end defun | |
3598 | |
3599 Since true Common Lisp supports up to four different floating-point | |
3600 precisions, it has families of constants like | |
3601 @code{most-positive-single-float}, @code{most-positive-double-float}, | |
3602 @code{most-positive-long-float}, and so on. Emacs has only one | |
3603 floating-point precision, so this package omits the precision word | |
3604 from the constants' names. | |
3605 | |
3606 @defvar most-positive-float | |
3607 This constant equals the largest value a Lisp float can hold. | |
3608 For those systems whose arithmetic supports infinities, this is | |
3609 the largest @emph{finite} value. For IEEE machines, the value | |
3610 is approximately @code{1.79e+308}. | |
3611 @end defvar | |
3612 | |
3613 @defvar most-negative-float | |
3614 This constant equals the most-negative value a Lisp float can hold. | |
3615 (It is assumed to be equal to @code{(- most-positive-float)}.) | |
3616 @end defvar | |
3617 | |
3618 @defvar least-positive-float | |
3619 This constant equals the smallest Lisp float value greater than zero. | |
3620 For IEEE machines, it is about @code{4.94e-324} if denormals are | |
3621 supported or @code{2.22e-308} if not. | |
3622 @end defvar | |
3623 | |
3624 @defvar least-positive-normalized-float | |
3625 This constant equals the smallest @emph{normalized} Lisp float greater | |
3626 than zero, i.e., the smallest value for which IEEE denormalization | |
3627 will not result in a loss of precision. For IEEE machines, this | |
3628 value is about @code{2.22e-308}. For machines that do not support | |
3629 the concept of denormalization and gradual underflow, this constant | |
3630 will always equal @code{least-positive-float}. | |
3631 @end defvar | |
3632 | |
3633 @defvar least-negative-float | |
3634 This constant is the negative counterpart of @code{least-positive-float}. | |
3635 @end defvar | |
3636 | |
3637 @defvar least-negative-normalized-float | |
3638 This constant is the negative counterpart of | |
3639 @code{least-positive-normalized-float}. | |
3640 @end defvar | |
3641 | |
3642 @defvar float-epsilon | |
3643 This constant is the smallest positive Lisp float that can be added | |
3644 to 1.0 to produce a distinct value. Adding a smaller number to 1.0 | |
3645 will yield 1.0 again due to roundoff. For IEEE machines, epsilon | |
3646 is about @code{2.22e-16}. | |
3647 @end defvar | |
3648 | |
3649 @defvar float-negative-epsilon | |
3650 This is the smallest positive value that can be subtracted from | |
3651 1.0 to produce a distinct value. For IEEE machines, it is about | |
3652 @code{1.11e-16}. | |
3653 @end defvar | |
3654 | |
3655 @iftex | |
3656 @chapno=13 | |
3657 @end iftex | |
3658 | |
3659 @node Sequences, Lists, Numbers, Top | |
3660 @chapter Sequences | |
3661 | |
3662 @noindent | |
3663 Common Lisp defines a number of functions that operate on | |
3664 @dfn{sequences}, which are either lists, strings, or vectors. | |
3665 Emacs Lisp includes a few of these, notably @code{elt} and | |
3666 @code{length}; this package defines most of the rest. | |
3667 | |
3668 @menu | |
3669 * Sequence Basics:: Arguments shared by all sequence functions | |
3670 * Mapping over Sequences:: `mapcar*', `mapcan', `map', `every', etc. | |
3671 * Sequence Functions:: `subseq', `remove*', `substitute', etc. | |
3672 * Searching Sequences:: `find', `position', `count', `search', etc. | |
3673 * Sorting Sequences:: `sort*', `stable-sort', `merge' | |
3674 @end menu | |
3675 | |
3676 @node Sequence Basics, Mapping over Sequences, Sequences, Sequences | |
3677 @section Sequence Basics | |
3678 | |
3679 @noindent | |
3680 Many of the sequence functions take keyword arguments; @pxref{Argument | |
3681 Lists}. All keyword arguments are optional and, if specified, | |
3682 may appear in any order. | |
3683 | |
3684 The @code{:key} argument should be passed either @code{nil}, or a | |
3685 function of one argument. This key function is used as a filter | |
3686 through which the elements of the sequence are seen; for example, | |
3687 @code{(find x y :key 'car)} is similar to @code{(assoc* x y)}: | |
3688 It searches for an element of the list whose @code{car} equals | |
3689 @code{x}, rather than for an element which equals @code{x} itself. | |
3690 If @code{:key} is omitted or @code{nil}, the filter is effectively | |
3691 the identity function. | |
3692 | |
3693 The @code{:test} and @code{:test-not} arguments should be either | |
3694 @code{nil}, or functions of two arguments. The test function is | |
3695 used to compare two sequence elements, or to compare a search value | |
3696 with sequence elements. (The two values are passed to the test | |
3697 function in the same order as the original sequence function | |
3698 arguments from which they are derived, or, if they both come from | |
3699 the same sequence, in the same order as they appear in that sequence.) | |
3700 The @code{:test} argument specifies a function which must return | |
3701 true (non-@code{nil}) to indicate a match; instead, you may use | |
3702 @code{:test-not} to give a function which returns @emph{false} to | |
3703 indicate a match. The default test function is @code{:test 'eql}. | |
3704 | |
3705 Many functions which take @var{item} and @code{:test} or @code{:test-not} | |
3706 arguments also come in @code{-if} and @code{-if-not} varieties, | |
3707 where a @var{predicate} function is passed instead of @var{item}, | |
3708 and sequence elements match if the predicate returns true on them | |
3709 (or false in the case of @code{-if-not}). For example: | |
3710 | |
3711 @example | |
3712 (remove* 0 seq :test '=) @equiv{} (remove-if 'zerop seq) | |
3713 @end example | |
3714 | |
3715 @noindent | |
3716 to remove all zeros from sequence @code{seq}. | |
3717 | |
3718 Some operations can work on a subsequence of the argument sequence; | |
3719 these function take @code{:start} and @code{:end} arguments which | |
3720 default to zero and the length of the sequence, respectively. | |
3721 Only elements between @var{start} (inclusive) and @var{end} | |
3722 (exclusive) are affected by the operation. The @var{end} argument | |
3723 may be passed @code{nil} to signify the length of the sequence; | |
3724 otherwise, both @var{start} and @var{end} must be integers, with | |
3725 @code{0 <= @var{start} <= @var{end} <= (length @var{seq})}. | |
3726 If the function takes two sequence arguments, the limits are | |
3727 defined by keywords @code{:start1} and @code{:end1} for the first, | |
3728 and @code{:start2} and @code{:end2} for the second. | |
3729 | |
3730 A few functions accept a @code{:from-end} argument, which, if | |
3731 non-@code{nil}, causes the operation to go from right-to-left | |
3732 through the sequence instead of left-to-right, and a @code{:count} | |
3733 argument, which specifies an integer maximum number of elements | |
3734 to be removed or otherwise processed. | |
3735 | |
3736 The sequence functions make no guarantees about the order in | |
3737 which the @code{:test}, @code{:test-not}, and @code{:key} functions | |
3738 are called on various elements. Therefore, it is a bad idea to depend | |
3739 on side effects of these functions. For example, @code{:from-end} | |
3740 may cause the sequence to be scanned actually in reverse, or it may | |
3741 be scanned forwards but computing a result ``as if'' it were scanned | |
3742 backwards. (Some functions, like @code{mapcar*} and @code{every}, | |
3743 @emph{do} specify exactly the order in which the function is called | |
3744 so side effects are perfectly acceptable in those cases.) | |
3745 | |
31572
384603523f41
Remove obsolescences, especially hash tables and refs to Emacs 19.
Dave Love <fx@gnu.org>
parents:
30009
diff
changeset
|
3746 Strings may contain ``text properties'' as well |
25829 | 3747 as character data. Except as noted, it is undefined whether or |
3748 not text properties are preserved by sequence functions. For | |
3749 example, @code{(remove* ?A @var{str})} may or may not preserve | |
3750 the properties of the characters copied from @var{str} into the | |
3751 result. | |
3752 | |
3753 @node Mapping over Sequences, Sequence Functions, Sequence Basics, Sequences | |
3754 @section Mapping over Sequences | |
3755 | |
3756 @noindent | |
3757 These functions ``map'' the function you specify over the elements | |
3758 of lists or arrays. They are all variations on the theme of the | |
3759 built-in function @code{mapcar}. | |
3760 | |
3761 @defun mapcar* function seq &rest more-seqs | |
3762 This function calls @var{function} on successive parallel sets of | |
3763 elements from its argument sequences. Given a single @var{seq} | |
3764 argument it is equivalent to @code{mapcar}; given @var{n} sequences, | |
3765 it calls the function with the first elements of each of the sequences | |
3766 as the @var{n} arguments to yield the first element of the result | |
3767 list, then with the second elements, and so on. The mapping stops as | |
3768 soon as the shortest sequence runs out. The argument sequences may | |
3769 be any mixture of lists, strings, and vectors; the return sequence | |
3770 is always a list. | |
3771 | |
3772 Common Lisp's @code{mapcar} accepts multiple arguments but works | |
3773 only on lists; Emacs Lisp's @code{mapcar} accepts a single sequence | |
3774 argument. This package's @code{mapcar*} works as a compatible | |
3775 superset of both. | |
3776 @end defun | |
3777 | |
3778 @defun map result-type function seq &rest more-seqs | |
3779 This function maps @var{function} over the argument sequences, | |
3780 just like @code{mapcar*}, but it returns a sequence of type | |
3781 @var{result-type} rather than a list. @var{result-type} must | |
3782 be one of the following symbols: @code{vector}, @code{string}, | |
3783 @code{list} (in which case the effect is the same as for | |
3784 @code{mapcar*}), or @code{nil} (in which case the results are | |
3785 thrown away and @code{map} returns @code{nil}). | |
3786 @end defun | |
3787 | |
3788 @defun maplist function list &rest more-lists | |
3789 This function calls @var{function} on each of its argument lists, | |
3790 then on the @code{cdr}s of those lists, and so on, until the | |
3791 shortest list runs out. The results are returned in the form | |
3792 of a list. Thus, @code{maplist} is like @code{mapcar*} except | |
3793 that it passes in the list pointers themselves rather than the | |
3794 @code{car}s of the advancing pointers. | |
3795 @end defun | |
3796 | |
3797 @defun mapc function seq &rest more-seqs | |
36931 | 3798 This function is like @code{mapcar*}, except that the values returned |
3799 by @var{function} are ignored and thrown away rather than being | |
3800 collected into a list. The return value of @code{mapc} is @var{seq}, | |
3801 the first sequence. This function is more general than the Emacs | |
3802 primitive @code{mapc}. | |
25829 | 3803 @end defun |
3804 | |
3805 @defun mapl function list &rest more-lists | |
3806 This function is like @code{maplist}, except that it throws away | |
3807 the values returned by @var{function}. | |
3808 @end defun | |
3809 | |
3810 @defun mapcan function seq &rest more-seqs | |
3811 This function is like @code{mapcar*}, except that it concatenates | |
3812 the return values (which must be lists) using @code{nconc}, | |
3813 rather than simply collecting them into a list. | |
3814 @end defun | |
3815 | |
3816 @defun mapcon function list &rest more-lists | |
3817 This function is like @code{maplist}, except that it concatenates | |
3818 the return values using @code{nconc}. | |
3819 @end defun | |
3820 | |
3821 @defun some predicate seq &rest more-seqs | |
3822 This function calls @var{predicate} on each element of @var{seq} | |
3823 in turn; if @var{predicate} returns a non-@code{nil} value, | |
3824 @code{some} returns that value, otherwise it returns @code{nil}. | |
3825 Given several sequence arguments, it steps through the sequences | |
3826 in parallel until the shortest one runs out, just as in | |
3827 @code{mapcar*}. You can rely on the left-to-right order in which | |
3828 the elements are visited, and on the fact that mapping stops | |
3829 immediately as soon as @var{predicate} returns non-@code{nil}. | |
3830 @end defun | |
3831 | |
3832 @defun every predicate seq &rest more-seqs | |
3833 This function calls @var{predicate} on each element of the sequence(s) | |
3834 in turn; it returns @code{nil} as soon as @var{predicate} returns | |
3835 @code{nil} for any element, or @code{t} if the predicate was true | |
3836 for all elements. | |
3837 @end defun | |
3838 | |
3839 @defun notany predicate seq &rest more-seqs | |
3840 This function calls @var{predicate} on each element of the sequence(s) | |
3841 in turn; it returns @code{nil} as soon as @var{predicate} returns | |
3842 a non-@code{nil} value for any element, or @code{t} if the predicate | |
3843 was @code{nil} for all elements. | |
3844 @end defun | |
3845 | |
3846 @defun notevery predicate seq &rest more-seqs | |
3847 This function calls @var{predicate} on each element of the sequence(s) | |
3848 in turn; it returns a non-@code{nil} value as soon as @var{predicate} | |
3849 returns @code{nil} for any element, or @code{t} if the predicate was | |
3850 true for all elements. | |
3851 @end defun | |
3852 | |
3853 @defun reduce function seq @t{&key :from-end :start :end :initial-value :key} | |
3854 This function combines the elements of @var{seq} using an associative | |
3855 binary operation. Suppose @var{function} is @code{*} and @var{seq} is | |
3856 the list @code{(2 3 4 5)}. The first two elements of the list are | |
3857 combined with @code{(* 2 3) = 6}; this is combined with the next | |
3858 element, @code{(* 6 4) = 24}, and that is combined with the final | |
3859 element: @code{(* 24 5) = 120}. Note that the @code{*} function happens | |
3860 to be self-reducing, so that @code{(* 2 3 4 5)} has the same effect as | |
3861 an explicit call to @code{reduce}. | |
3862 | |
3863 If @code{:from-end} is true, the reduction is right-associative instead | |
3864 of left-associative: | |
3865 | |
3866 @example | |
3867 (reduce '- '(1 2 3 4)) | |
3868 @equiv{} (- (- (- 1 2) 3) 4) @result{} -8 | |
3869 (reduce '- '(1 2 3 4) :from-end t) | |
3870 @equiv{} (- 1 (- 2 (- 3 4))) @result{} -2 | |
3871 @end example | |
3872 | |
3873 If @code{:key} is specified, it is a function of one argument which | |
3874 is called on each of the sequence elements in turn. | |
3875 | |
3876 If @code{:initial-value} is specified, it is effectively added to the | |
3877 front (or rear in the case of @code{:from-end}) of the sequence. | |
3878 The @code{:key} function is @emph{not} applied to the initial value. | |
3879 | |
3880 If the sequence, including the initial value, has exactly one element | |
3881 then that element is returned without ever calling @var{function}. | |
3882 If the sequence is empty (and there is no initial value), then | |
3883 @var{function} is called with no arguments to obtain the return value. | |
3884 @end defun | |
3885 | |
3886 All of these mapping operations can be expressed conveniently in | |
3887 terms of the @code{loop} macro. In compiled code, @code{loop} will | |
3888 be faster since it generates the loop as in-line code with no | |
3889 function calls. | |
3890 | |
3891 @node Sequence Functions, Searching Sequences, Mapping over Sequences, Sequences | |
3892 @section Sequence Functions | |
3893 | |
3894 @noindent | |
3895 This section describes a number of Common Lisp functions for | |
3896 operating on sequences. | |
3897 | |
3898 @defun subseq sequence start &optional end | |
3899 This function returns a given subsequence of the argument | |
3900 @var{sequence}, which may be a list, string, or vector. | |
3901 The indices @var{start} and @var{end} must be in range, and | |
3902 @var{start} must be no greater than @var{end}. If @var{end} | |
3903 is omitted, it defaults to the length of the sequence. The | |
3904 return value is always a copy; it does not share structure | |
3905 with @var{sequence}. | |
3906 | |
3907 As an extension to Common Lisp, @var{start} and/or @var{end} | |
3908 may be negative, in which case they represent a distance back | |
3909 from the end of the sequence. This is for compatibility with | |
3910 Emacs' @code{substring} function. Note that @code{subseq} is | |
3911 the @emph{only} sequence function that allows negative | |
3912 @var{start} and @var{end}. | |
3913 | |
3914 You can use @code{setf} on a @code{subseq} form to replace a | |
3915 specified range of elements with elements from another sequence. | |
3916 The replacement is done as if by @code{replace}, described below. | |
3917 @end defun | |
3918 | |
3919 @defun concatenate result-type &rest seqs | |
3920 This function concatenates the argument sequences together to | |
3921 form a result sequence of type @var{result-type}, one of the | |
3922 symbols @code{vector}, @code{string}, or @code{list}. The | |
3923 arguments are always copied, even in cases such as | |
3924 @code{(concatenate 'list '(1 2 3))} where the result is | |
3925 identical to an argument. | |
3926 @end defun | |
3927 | |
3928 @defun fill seq item @t{&key :start :end} | |
3929 This function fills the elements of the sequence (or the specified | |
3930 part of the sequence) with the value @var{item}. | |
3931 @end defun | |
3932 | |
3933 @defun replace seq1 seq2 @t{&key :start1 :end1 :start2 :end2} | |
3934 This function copies part of @var{seq2} into part of @var{seq1}. | |
3935 The sequence @var{seq1} is not stretched or resized; the amount | |
3936 of data copied is simply the shorter of the source and destination | |
3937 (sub)sequences. The function returns @var{seq1}. | |
3938 | |
3939 If @var{seq1} and @var{seq2} are @code{eq}, then the replacement | |
3940 will work correctly even if the regions indicated by the start | |
3941 and end arguments overlap. However, if @var{seq1} and @var{seq2} | |
3942 are lists which share storage but are not @code{eq}, and the | |
3943 start and end arguments specify overlapping regions, the effect | |
3944 is undefined. | |
3945 @end defun | |
3946 | |
3947 @defun remove* item seq @t{&key :test :test-not :key :count :start :end :from-end} | |
3948 This returns a copy of @var{seq} with all elements matching | |
3949 @var{item} removed. The result may share storage with or be | |
3950 @code{eq} to @var{seq} in some circumstances, but the original | |
3951 @var{seq} will not be modified. The @code{:test}, @code{:test-not}, | |
3952 and @code{:key} arguments define the matching test that is used; | |
3953 by default, elements @code{eql} to @var{item} are removed. The | |
3954 @code{:count} argument specifies the maximum number of matching | |
3955 elements that can be removed (only the leftmost @var{count} matches | |
3956 are removed). The @code{:start} and @code{:end} arguments specify | |
3957 a region in @var{seq} in which elements will be removed; elements | |
3958 outside that region are not matched or removed. The @code{:from-end} | |
3959 argument, if true, says that elements should be deleted from the | |
3960 end of the sequence rather than the beginning (this matters only | |
3961 if @var{count} was also specified). | |
3962 @end defun | |
3963 | |
3964 @defun delete* item seq @t{&key :test :test-not :key :count :start :end :from-end} | |
3965 This deletes all elements of @var{seq} which match @var{item}. | |
3966 It is a destructive operation. Since Emacs Lisp does not support | |
3967 stretchable strings or vectors, this is the same as @code{remove*} | |
3968 for those sequence types. On lists, @code{remove*} will copy the | |
3969 list if necessary to preserve the original list, whereas | |
3970 @code{delete*} will splice out parts of the argument list. | |
3971 Compare @code{append} and @code{nconc}, which are analogous | |
3972 non-destructive and destructive list operations in Emacs Lisp. | |
3973 @end defun | |
3974 | |
3975 @findex remove-if | |
3976 @findex remove-if-not | |
3977 @findex delete-if | |
3978 @findex delete-if-not | |
3979 The predicate-oriented functions @code{remove-if}, @code{remove-if-not}, | |
3980 @code{delete-if}, and @code{delete-if-not} are defined similarly. | |
3981 | |
3982 @defun remove-duplicates seq @t{&key :test :test-not :key :start :end :from-end} | |
3983 This function returns a copy of @var{seq} with duplicate elements | |
3984 removed. Specifically, if two elements from the sequence match | |
3985 according to the @code{:test}, @code{:test-not}, and @code{:key} | |
3986 arguments, only the rightmost one is retained. If @code{:from-end} | |
3987 is true, the leftmost one is retained instead. If @code{:start} or | |
3988 @code{:end} is specified, only elements within that subsequence are | |
3989 examined or removed. | |
3990 @end defun | |
3991 | |
3992 @defun delete-duplicates seq @t{&key :test :test-not :key :start :end :from-end} | |
3993 This function deletes duplicate elements from @var{seq}. It is | |
3994 a destructive version of @code{remove-duplicates}. | |
3995 @end defun | |
3996 | |
3997 @defun substitute new old seq @t{&key :test :test-not :key :count :start :end :from-end} | |
3998 This function returns a copy of @var{seq}, with all elements | |
3999 matching @var{old} replaced with @var{new}. The @code{:count}, | |
4000 @code{:start}, @code{:end}, and @code{:from-end} arguments may be | |
4001 used to limit the number of substitutions made. | |
4002 @end defun | |
4003 | |
4004 @defun nsubstitute new old seq @t{&key :test :test-not :key :count :start :end :from-end} | |
4005 This is a destructive version of @code{substitute}; it performs | |
4006 the substitution using @code{setcar} or @code{aset} rather than | |
4007 by returning a changed copy of the sequence. | |
4008 @end defun | |
4009 | |
4010 @findex substitute-if | |
4011 @findex substitute-if-not | |
4012 @findex nsubstitute-if | |
4013 @findex nsubstitute-if-not | |
4014 The @code{substitute-if}, @code{substitute-if-not}, @code{nsubstitute-if}, | |
4015 and @code{nsubstitute-if-not} functions are defined similarly. For | |
4016 these, a @var{predicate} is given in place of the @var{old} argument. | |
4017 | |
4018 @node Searching Sequences, Sorting Sequences, Sequence Functions, Sequences | |
4019 @section Searching Sequences | |
4020 | |
4021 @noindent | |
4022 These functions search for elements or subsequences in a sequence. | |
4023 (See also @code{member*} and @code{assoc*}; @pxref{Lists}.) | |
4024 | |
4025 @defun find item seq @t{&key :test :test-not :key :start :end :from-end} | |
4026 This function searches @var{seq} for an element matching @var{item}. | |
4027 If it finds a match, it returns the matching element. Otherwise, | |
4028 it returns @code{nil}. It returns the leftmost match, unless | |
4029 @code{:from-end} is true, in which case it returns the rightmost | |
4030 match. The @code{:start} and @code{:end} arguments may be used to | |
4031 limit the range of elements that are searched. | |
4032 @end defun | |
4033 | |
4034 @defun position item seq @t{&key :test :test-not :key :start :end :from-end} | |
4035 This function is like @code{find}, except that it returns the | |
4036 integer position in the sequence of the matching item rather than | |
4037 the item itself. The position is relative to the start of the | |
4038 sequence as a whole, even if @code{:start} is non-zero. The function | |
4039 returns @code{nil} if no matching element was found. | |
4040 @end defun | |
4041 | |
4042 @defun count item seq @t{&key :test :test-not :key :start :end} | |
4043 This function returns the number of elements of @var{seq} which | |
4044 match @var{item}. The result is always a nonnegative integer. | |
4045 @end defun | |
4046 | |
4047 @findex find-if | |
4048 @findex find-if-not | |
4049 @findex position-if | |
4050 @findex position-if-not | |
4051 @findex count-if | |
4052 @findex count-if-not | |
4053 The @code{find-if}, @code{find-if-not}, @code{position-if}, | |
4054 @code{position-if-not}, @code{count-if}, and @code{count-if-not} | |
4055 functions are defined similarly. | |
4056 | |
4057 @defun mismatch seq1 seq2 @t{&key :test :test-not :key :start1 :end1 :start2 :end2 :from-end} | |
4058 This function compares the specified parts of @var{seq1} and | |
4059 @var{seq2}. If they are the same length and the corresponding | |
4060 elements match (according to @code{:test}, @code{:test-not}, | |
4061 and @code{:key}), the function returns @code{nil}. If there is | |
4062 a mismatch, the function returns the index (relative to @var{seq1}) | |
4063 of the first mismatching element. This will be the leftmost pair of | |
4064 elements which do not match, or the position at which the shorter of | |
4065 the two otherwise-matching sequences runs out. | |
4066 | |
4067 If @code{:from-end} is true, then the elements are compared from right | |
4068 to left starting at @code{(1- @var{end1})} and @code{(1- @var{end2})}. | |
4069 If the sequences differ, then one plus the index of the rightmost | |
4070 difference (relative to @var{seq1}) is returned. | |
4071 | |
4072 An interesting example is @code{(mismatch str1 str2 :key 'upcase)}, | |
4073 which compares two strings case-insensitively. | |
4074 @end defun | |
4075 | |
4076 @defun search seq1 seq2 @t{&key :test :test-not :key :from-end :start1 :end1 :start2 :end2} | |
4077 This function searches @var{seq2} for a subsequence that matches | |
4078 @var{seq1} (or part of it specified by @code{:start1} and | |
4079 @code{:end1}.) Only matches which fall entirely within the region | |
4080 defined by @code{:start2} and @code{:end2} will be considered. | |
4081 The return value is the index of the leftmost element of the | |
4082 leftmost match, relative to the start of @var{seq2}, or @code{nil} | |
4083 if no matches were found. If @code{:from-end} is true, the | |
4084 function finds the @emph{rightmost} matching subsequence. | |
4085 @end defun | |
4086 | |
31572
384603523f41
Remove obsolescences, especially hash tables and refs to Emacs 19.
Dave Love <fx@gnu.org>
parents:
30009
diff
changeset
|
4087 @node Sorting Sequences, , Searching Sequences, Sequences |
25829 | 4088 @section Sorting Sequences |
4089 | |
4090 @defun sort* seq predicate @t{&key :key} | |
4091 This function sorts @var{seq} into increasing order as determined | |
4092 by using @var{predicate} to compare pairs of elements. @var{predicate} | |
4093 should return true (non-@code{nil}) if and only if its first argument | |
4094 is less than (not equal to) its second argument. For example, | |
4095 @code{<} and @code{string-lessp} are suitable predicate functions | |
4096 for sorting numbers and strings, respectively; @code{>} would sort | |
4097 numbers into decreasing rather than increasing order. | |
4098 | |
4099 This function differs from Emacs' built-in @code{sort} in that it | |
4100 can operate on any type of sequence, not just lists. Also, it | |
4101 accepts a @code{:key} argument which is used to preprocess data | |
4102 fed to the @var{predicate} function. For example, | |
4103 | |
4104 @example | |
4105 (setq data (sort data 'string-lessp :key 'downcase)) | |
4106 @end example | |
4107 | |
4108 @noindent | |
4109 sorts @var{data}, a sequence of strings, into increasing alphabetical | |
4110 order without regard to case. A @code{:key} function of @code{car} | |
4111 would be useful for sorting association lists. | |
4112 | |
4113 The @code{sort*} function is destructive; it sorts lists by actually | |
4114 rearranging the @code{cdr} pointers in suitable fashion. | |
4115 @end defun | |
4116 | |
4117 @defun stable-sort seq predicate @t{&key :key} | |
4118 This function sorts @var{seq} @dfn{stably}, meaning two elements | |
4119 which are equal in terms of @var{predicate} are guaranteed not to | |
4120 be rearranged out of their original order by the sort. | |
4121 | |
4122 In practice, @code{sort*} and @code{stable-sort} are equivalent | |
4123 in Emacs Lisp because the underlying @code{sort} function is | |
4124 stable by default. However, this package reserves the right to | |
4125 use non-stable methods for @code{sort*} in the future. | |
4126 @end defun | |
4127 | |
4128 @defun merge type seq1 seq2 predicate @t{&key :key} | |
4129 This function merges two sequences @var{seq1} and @var{seq2} by | |
4130 interleaving their elements. The result sequence, of type @var{type} | |
4131 (in the sense of @code{concatenate}), has length equal to the sum | |
4132 of the lengths of the two input sequences. The sequences may be | |
4133 modified destructively. Order of elements within @var{seq1} and | |
4134 @var{seq2} is preserved in the interleaving; elements of the two | |
4135 sequences are compared by @var{predicate} (in the sense of | |
4136 @code{sort}) and the lesser element goes first in the result. | |
4137 When elements are equal, those from @var{seq1} precede those from | |
4138 @var{seq2} in the result. Thus, if @var{seq1} and @var{seq2} are | |
4139 both sorted according to @var{predicate}, then the result will be | |
4140 a merged sequence which is (stably) sorted according to | |
4141 @var{predicate}. | |
4142 @end defun | |
4143 | |
31572
384603523f41
Remove obsolescences, especially hash tables and refs to Emacs 19.
Dave Love <fx@gnu.org>
parents:
30009
diff
changeset
|
4144 @node Lists, Structures, Sequences, Top |
25829 | 4145 @chapter Lists |
4146 | |
4147 @noindent | |
4148 The functions described here operate on lists. | |
4149 | |
4150 @menu | |
36931 | 4151 * List Functions:: `caddr', `first', `list*', etc. |
25829 | 4152 * Substitution of Expressions:: `subst', `sublis', etc. |
4153 * Lists as Sets:: `member*', `adjoin', `union', etc. | |
4154 * Association Lists:: `assoc*', `rassoc*', `acons', `pairlis' | |
4155 @end menu | |
4156 | |
4157 @node List Functions, Substitution of Expressions, Lists, Lists | |
4158 @section List Functions | |
4159 | |
4160 @noindent | |
4161 This section describes a number of simple operations on lists, | |
4162 i.e., chains of cons cells. | |
4163 | |
4164 @defun caddr x | |
4165 This function is equivalent to @code{(car (cdr (cdr @var{x})))}. | |
4166 Likewise, this package defines all 28 @code{c@var{xxx}r} functions | |
4167 where @var{xxx} is up to four @samp{a}s and/or @samp{d}s. | |
4168 All of these functions are @code{setf}-able, and calls to them | |
4169 are expanded inline by the byte-compiler for maximum efficiency. | |
4170 @end defun | |
4171 | |
4172 @defun first x | |
4173 This function is a synonym for @code{(car @var{x})}. Likewise, | |
4174 the functions @code{second}, @code{third}, @dots{}, through | |
4175 @code{tenth} return the given element of the list @var{x}. | |
4176 @end defun | |
4177 | |
4178 @defun rest x | |
4179 This function is a synonym for @code{(cdr @var{x})}. | |
4180 @end defun | |
4181 | |
4182 @defun endp x | |
4183 Common Lisp defines this function to act like @code{null}, but | |
4184 signaling an error if @code{x} is neither a @code{nil} nor a | |
4185 cons cell. This package simply defines @code{endp} as a synonym | |
4186 for @code{null}. | |
4187 @end defun | |
4188 | |
4189 @defun list-length x | |
4190 This function returns the length of list @var{x}, exactly like | |
4191 @code{(length @var{x})}, except that if @var{x} is a circular | |
4192 list (where the cdr-chain forms a loop rather than terminating | |
4193 with @code{nil}), this function returns @code{nil}. (The regular | |
4194 @code{length} function would get stuck if given a circular list.) | |
4195 @end defun | |
4196 | |
4197 @defun list* arg &rest others | |
4198 This function constructs a list of its arguments. The final | |
4199 argument becomes the @code{cdr} of the last cell constructed. | |
4200 Thus, @code{(list* @var{a} @var{b} @var{c})} is equivalent to | |
4201 @code{(cons @var{a} (cons @var{b} @var{c}))}, and | |
4202 @code{(list* @var{a} @var{b} nil)} is equivalent to | |
4203 @code{(list @var{a} @var{b})}. | |
4204 | |
4205 (Note that this function really is called @code{list*} in Common | |
4206 Lisp; it is not a name invented for this package like @code{member*} | |
4207 or @code{defun*}.) | |
4208 @end defun | |
4209 | |
4210 @defun ldiff list sublist | |
4211 If @var{sublist} is a sublist of @var{list}, i.e., is @code{eq} to | |
4212 one of the cons cells of @var{list}, then this function returns | |
4213 a copy of the part of @var{list} up to but not including | |
4214 @var{sublist}. For example, @code{(ldiff x (cddr x))} returns | |
4215 the first two elements of the list @code{x}. The result is a | |
4216 copy; the original @var{list} is not modified. If @var{sublist} | |
4217 is not a sublist of @var{list}, a copy of the entire @var{list} | |
4218 is returned. | |
4219 @end defun | |
4220 | |
4221 @defun copy-list list | |
4222 This function returns a copy of the list @var{list}. It copies | |
4223 dotted lists like @code{(1 2 . 3)} correctly. | |
4224 @end defun | |
4225 | |
4226 @defun copy-tree x &optional vecp | |
4227 This function returns a copy of the tree of cons cells @var{x}. | |
4228 Unlike @code{copy-sequence} (and its alias @code{copy-list}), | |
4229 which copies only along the @code{cdr} direction, this function | |
4230 copies (recursively) along both the @code{car} and the @code{cdr} | |
4231 directions. If @var{x} is not a cons cell, the function simply | |
4232 returns @var{x} unchanged. If the optional @var{vecp} argument | |
4233 is true, this function copies vectors (recursively) as well as | |
4234 cons cells. | |
4235 @end defun | |
4236 | |
4237 @defun tree-equal x y @t{&key :test :test-not :key} | |
4238 This function compares two trees of cons cells. If @var{x} and | |
4239 @var{y} are both cons cells, their @code{car}s and @code{cdr}s are | |
4240 compared recursively. If neither @var{x} nor @var{y} is a cons | |
4241 cell, they are compared by @code{eql}, or according to the | |
4242 specified test. The @code{:key} function, if specified, is | |
4243 applied to the elements of both trees. @xref{Sequences}. | |
4244 @end defun | |
4245 | |
4246 @iftex | |
4247 @secno=3 | |
4248 @end iftex | |
4249 | |
4250 @node Substitution of Expressions, Lists as Sets, List Functions, Lists | |
4251 @section Substitution of Expressions | |
4252 | |
4253 @noindent | |
4254 These functions substitute elements throughout a tree of cons | |
4255 cells. (@xref{Sequence Functions}, for the @code{substitute} | |
4256 function, which works on just the top-level elements of a list.) | |
4257 | |
4258 @defun subst new old tree @t{&key :test :test-not :key} | |
4259 This function substitutes occurrences of @var{old} with @var{new} | |
4260 in @var{tree}, a tree of cons cells. It returns a substituted | |
4261 tree, which will be a copy except that it may share storage with | |
4262 the argument @var{tree} in parts where no substitutions occurred. | |
4263 The original @var{tree} is not modified. This function recurses | |
4264 on, and compares against @var{old}, both @code{car}s and @code{cdr}s | |
4265 of the component cons cells. If @var{old} is itself a cons cell, | |
4266 then matching cells in the tree are substituted as usual without | |
4267 recursively substituting in that cell. Comparisons with @var{old} | |
4268 are done according to the specified test (@code{eql} by default). | |
4269 The @code{:key} function is applied to the elements of the tree | |
4270 but not to @var{old}. | |
4271 @end defun | |
4272 | |
4273 @defun nsubst new old tree @t{&key :test :test-not :key} | |
4274 This function is like @code{subst}, except that it works by | |
4275 destructive modification (by @code{setcar} or @code{setcdr}) | |
4276 rather than copying. | |
4277 @end defun | |
4278 | |
4279 @findex subst-if | |
4280 @findex subst-if-not | |
4281 @findex nsubst-if | |
4282 @findex nsubst-if-not | |
4283 The @code{subst-if}, @code{subst-if-not}, @code{nsubst-if}, and | |
4284 @code{nsubst-if-not} functions are defined similarly. | |
4285 | |
4286 @defun sublis alist tree @t{&key :test :test-not :key} | |
4287 This function is like @code{subst}, except that it takes an | |
4288 association list @var{alist} of @var{old}-@var{new} pairs. | |
4289 Each element of the tree (after applying the @code{:key} | |
4290 function, if any), is compared with the @code{car}s of | |
4291 @var{alist}; if it matches, it is replaced by the corresponding | |
4292 @code{cdr}. | |
4293 @end defun | |
4294 | |
4295 @defun nsublis alist tree @t{&key :test :test-not :key} | |
4296 This is a destructive version of @code{sublis}. | |
4297 @end defun | |
4298 | |
4299 @node Lists as Sets, Association Lists, Substitution of Expressions, Lists | |
4300 @section Lists as Sets | |
4301 | |
4302 @noindent | |
4303 These functions perform operations on lists which represent sets | |
4304 of elements. | |
4305 | |
4306 @defun member* item list @t{&key :test :test-not :key} | |
4307 This function searches @var{list} for an element matching @var{item}. | |
4308 If a match is found, it returns the cons cell whose @code{car} was | |
4309 the matching element. Otherwise, it returns @code{nil}. Elements | |
4310 are compared by @code{eql} by default; you can use the @code{:test}, | |
4311 @code{:test-not}, and @code{:key} arguments to modify this behavior. | |
4312 @xref{Sequences}. | |
4313 | |
4314 Note that this function's name is suffixed by @samp{*} to avoid | |
31572
384603523f41
Remove obsolescences, especially hash tables and refs to Emacs 19.
Dave Love <fx@gnu.org>
parents:
30009
diff
changeset
|
4315 the incompatible @code{member} function defined in Emacs. |
25829 | 4316 (That function uses @code{equal} for comparisons; it is equivalent |
4317 to @code{(member* @var{item} @var{list} :test 'equal)}.) | |
4318 @end defun | |
4319 | |
4320 @findex member-if | |
4321 @findex member-if-not | |
4322 The @code{member-if} and @code{member-if-not} functions | |
4323 analogously search for elements which satisfy a given predicate. | |
4324 | |
4325 @defun tailp sublist list | |
4326 This function returns @code{t} if @var{sublist} is a sublist of | |
4327 @var{list}, i.e., if @var{sublist} is @code{eql} to @var{list} or to | |
4328 any of its @code{cdr}s. | |
4329 @end defun | |
4330 | |
4331 @defun adjoin item list @t{&key :test :test-not :key} | |
4332 This function conses @var{item} onto the front of @var{list}, | |
4333 like @code{(cons @var{item} @var{list})}, but only if @var{item} | |
4334 is not already present on the list (as determined by @code{member*}). | |
4335 If a @code{:key} argument is specified, it is applied to | |
4336 @var{item} as well as to the elements of @var{list} during | |
4337 the search, on the reasoning that @var{item} is ``about'' to | |
4338 become part of the list. | |
4339 @end defun | |
4340 | |
4341 @defun union list1 list2 @t{&key :test :test-not :key} | |
4342 This function combines two lists which represent sets of items, | |
4343 returning a list that represents the union of those two sets. | |
4344 The result list will contain all items which appear in @var{list1} | |
4345 or @var{list2}, and no others. If an item appears in both | |
4346 @var{list1} and @var{list2} it will be copied only once. If | |
4347 an item is duplicated in @var{list1} or @var{list2}, it is | |
4348 undefined whether or not that duplication will survive in the | |
4349 result list. The order of elements in the result list is also | |
4350 undefined. | |
4351 @end defun | |
4352 | |
4353 @defun nunion list1 list2 @t{&key :test :test-not :key} | |
4354 This is a destructive version of @code{union}; rather than copying, | |
4355 it tries to reuse the storage of the argument lists if possible. | |
4356 @end defun | |
4357 | |
4358 @defun intersection list1 list2 @t{&key :test :test-not :key} | |
4359 This function computes the intersection of the sets represented | |
4360 by @var{list1} and @var{list2}. It returns the list of items | |
4361 which appear in both @var{list1} and @var{list2}. | |
4362 @end defun | |
4363 | |
4364 @defun nintersection list1 list2 @t{&key :test :test-not :key} | |
4365 This is a destructive version of @code{intersection}. It | |
4366 tries to reuse storage of @var{list1} rather than copying. | |
4367 It does @emph{not} reuse the storage of @var{list2}. | |
4368 @end defun | |
4369 | |
4370 @defun set-difference list1 list2 @t{&key :test :test-not :key} | |
4371 This function computes the ``set difference'' of @var{list1} | |
4372 and @var{list2}, i.e., the set of elements that appear in | |
4373 @var{list1} but @emph{not} in @var{list2}. | |
4374 @end defun | |
4375 | |
4376 @defun nset-difference list1 list2 @t{&key :test :test-not :key} | |
4377 This is a destructive @code{set-difference}, which will try | |
4378 to reuse @var{list1} if possible. | |
4379 @end defun | |
4380 | |
4381 @defun set-exclusive-or list1 list2 @t{&key :test :test-not :key} | |
4382 This function computes the ``set exclusive or'' of @var{list1} | |
4383 and @var{list2}, i.e., the set of elements that appear in | |
4384 exactly one of @var{list1} and @var{list2}. | |
4385 @end defun | |
4386 | |
4387 @defun nset-exclusive-or list1 list2 @t{&key :test :test-not :key} | |
4388 This is a destructive @code{set-exclusive-or}, which will try | |
4389 to reuse @var{list1} and @var{list2} if possible. | |
4390 @end defun | |
4391 | |
4392 @defun subsetp list1 list2 @t{&key :test :test-not :key} | |
4393 This function checks whether @var{list1} represents a subset | |
4394 of @var{list2}, i.e., whether every element of @var{list1} | |
4395 also appears in @var{list2}. | |
4396 @end defun | |
4397 | |
31572
384603523f41
Remove obsolescences, especially hash tables and refs to Emacs 19.
Dave Love <fx@gnu.org>
parents:
30009
diff
changeset
|
4398 @node Association Lists, , Lists as Sets, Lists |
25829 | 4399 @section Association Lists |
4400 | |
4401 @noindent | |
4402 An @dfn{association list} is a list representing a mapping from | |
4403 one set of values to another; any list whose elements are cons | |
4404 cells is an association list. | |
4405 | |
4406 @defun assoc* item a-list @t{&key :test :test-not :key} | |
4407 This function searches the association list @var{a-list} for an | |
4408 element whose @code{car} matches (in the sense of @code{:test}, | |
4409 @code{:test-not}, and @code{:key}, or by comparison with @code{eql}) | |
4410 a given @var{item}. It returns the matching element, if any, | |
4411 otherwise @code{nil}. It ignores elements of @var{a-list} which | |
4412 are not cons cells. (This corresponds to the behavior of | |
4413 @code{assq} and @code{assoc} in Emacs Lisp; Common Lisp's | |
4414 @code{assoc} ignores @code{nil}s but considers any other non-cons | |
4415 elements of @var{a-list} to be an error.) | |
4416 @end defun | |
4417 | |
4418 @defun rassoc* item a-list @t{&key :test :test-not :key} | |
4419 This function searches for an element whose @code{cdr} matches | |
4420 @var{item}. If @var{a-list} represents a mapping, this applies | |
4421 the inverse of the mapping to @var{item}. | |
4422 @end defun | |
4423 | |
4424 @findex assoc-if | |
4425 @findex assoc-if-not | |
4426 @findex rassoc-if | |
4427 @findex rassoc-if-not | |
4428 The @code{assoc-if}, @code{assoc-if-not}, @code{rassoc-if}, | |
4429 and @code{rassoc-if-not} functions are defined similarly. | |
4430 | |
4431 Two simple functions for constructing association lists are: | |
4432 | |
4433 @defun acons key value alist | |
4434 This is equivalent to @code{(cons (cons @var{key} @var{value}) @var{alist})}. | |
4435 @end defun | |
4436 | |
4437 @defun pairlis keys values &optional alist | |
4438 This is equivalent to @code{(nconc (mapcar* 'cons @var{keys} @var{values}) | |
4439 @var{alist})}. | |
4440 @end defun | |
4441 | |
4442 @iftex | |
4443 @chapno=18 | |
4444 @end iftex | |
4445 | |
31572
384603523f41
Remove obsolescences, especially hash tables and refs to Emacs 19.
Dave Love <fx@gnu.org>
parents:
30009
diff
changeset
|
4446 @node Structures, Assertions, Lists, Top |
25829 | 4447 @chapter Structures |
4448 | |
4449 @noindent | |
4450 The Common Lisp @dfn{structure} mechanism provides a general way | |
4451 to define data types similar to C's @code{struct} types. A | |
4452 structure is a Lisp object containing some number of @dfn{slots}, | |
4453 each of which can hold any Lisp data object. Functions are | |
4454 provided for accessing and setting the slots, creating or copying | |
4455 structure objects, and recognizing objects of a particular structure | |
4456 type. | |
4457 | |
4458 In true Common Lisp, each structure type is a new type distinct | |
4459 from all existing Lisp types. Since the underlying Emacs Lisp | |
4460 system provides no way to create new distinct types, this package | |
4461 implements structures as vectors (or lists upon request) with a | |
4462 special ``tag'' symbol to identify them. | |
4463 | |
4464 @defspec defstruct name slots@dots{} | |
4465 The @code{defstruct} form defines a new structure type called | |
4466 @var{name}, with the specified @var{slots}. (The @var{slots} | |
4467 may begin with a string which documents the structure type.) | |
4468 In the simplest case, @var{name} and each of the @var{slots} | |
4469 are symbols. For example, | |
4470 | |
4471 @example | |
4472 (defstruct person name age sex) | |
4473 @end example | |
4474 | |
4475 @noindent | |
4476 defines a struct type called @code{person} which contains three | |
4477 slots. Given a @code{person} object @var{p}, you can access those | |
4478 slots by calling @code{(person-name @var{p})}, @code{(person-age @var{p})}, | |
4479 and @code{(person-sex @var{p})}. You can also change these slots by | |
4480 using @code{setf} on any of these place forms: | |
4481 | |
4482 @example | |
4483 (incf (person-age birthday-boy)) | |
4484 @end example | |
4485 | |
4486 You can create a new @code{person} by calling @code{make-person}, | |
4487 which takes keyword arguments @code{:name}, @code{:age}, and | |
4488 @code{:sex} to specify the initial values of these slots in the | |
4489 new object. (Omitting any of these arguments leaves the corresponding | |
4490 slot ``undefined,'' according to the Common Lisp standard; in Emacs | |
4491 Lisp, such uninitialized slots are filled with @code{nil}.) | |
4492 | |
4493 Given a @code{person}, @code{(copy-person @var{p})} makes a new | |
4494 object of the same type whose slots are @code{eq} to those of @var{p}. | |
4495 | |
4496 Given any Lisp object @var{x}, @code{(person-p @var{x})} returns | |
4497 true if @var{x} looks like a @code{person}, false otherwise. (Again, | |
4498 in Common Lisp this predicate would be exact; in Emacs Lisp the | |
4499 best it can do is verify that @var{x} is a vector of the correct | |
4500 length which starts with the correct tag symbol.) | |
4501 | |
4502 Accessors like @code{person-name} normally check their arguments | |
4503 (effectively using @code{person-p}) and signal an error if the | |
4504 argument is the wrong type. This check is affected by | |
4505 @code{(optimize (safety @dots{}))} declarations. Safety level 1, | |
4506 the default, uses a somewhat optimized check that will detect all | |
4507 incorrect arguments, but may use an uninformative error message | |
4508 (e.g., ``expected a vector'' instead of ``expected a @code{person}''). | |
4509 Safety level 0 omits all checks except as provided by the underlying | |
4510 @code{aref} call; safety levels 2 and 3 do rigorous checking that will | |
4511 always print a descriptive error message for incorrect inputs. | |
4512 @xref{Declarations}. | |
4513 | |
4514 @example | |
4515 (setq dave (make-person :name "Dave" :sex 'male)) | |
4516 @result{} [cl-struct-person "Dave" nil male] | |
4517 (setq other (copy-person dave)) | |
4518 @result{} [cl-struct-person "Dave" nil male] | |
4519 (eq dave other) | |
4520 @result{} nil | |
4521 (eq (person-name dave) (person-name other)) | |
4522 @result{} t | |
4523 (person-p dave) | |
4524 @result{} t | |
4525 (person-p [1 2 3 4]) | |
4526 @result{} nil | |
4527 (person-p "Bogus") | |
4528 @result{} nil | |
4529 (person-p '[cl-struct-person counterfeit person object]) | |
4530 @result{} t | |
4531 @end example | |
4532 | |
4533 In general, @var{name} is either a name symbol or a list of a name | |
4534 symbol followed by any number of @dfn{struct options}; each @var{slot} | |
4535 is either a slot symbol or a list of the form @samp{(@var{slot-name} | |
4536 @var{default-value} @var{slot-options}@dots{})}. The @var{default-value} | |
4537 is a Lisp form which is evaluated any time an instance of the | |
4538 structure type is created without specifying that slot's value. | |
4539 | |
4540 Common Lisp defines several slot options, but the only one | |
4541 implemented in this package is @code{:read-only}. A non-@code{nil} | |
4542 value for this option means the slot should not be @code{setf}-able; | |
4543 the slot's value is determined when the object is created and does | |
4544 not change afterward. | |
4545 | |
4546 @example | |
4547 (defstruct person | |
4548 (name nil :read-only t) | |
4549 age | |
4550 (sex 'unknown)) | |
4551 @end example | |
4552 | |
4553 Any slot options other than @code{:read-only} are ignored. | |
4554 | |
4555 For obscure historical reasons, structure options take a different | |
4556 form than slot options. A structure option is either a keyword | |
4557 symbol, or a list beginning with a keyword symbol possibly followed | |
4558 by arguments. (By contrast, slot options are key-value pairs not | |
4559 enclosed in lists.) | |
4560 | |
4561 @example | |
4562 (defstruct (person (:constructor create-person) | |
4563 (:type list) | |
4564 :named) | |
4565 name age sex) | |
4566 @end example | |
4567 | |
4568 The following structure options are recognized. | |
4569 | |
4570 @table @code | |
4571 @iftex | |
4572 @itemmax=0 in | |
4573 @advance@leftskip-.5@tableindent | |
4574 @end iftex | |
4575 @item :conc-name | |
4576 The argument is a symbol whose print name is used as the prefix for | |
4577 the names of slot accessor functions. The default is the name of | |
4578 the struct type followed by a hyphen. The option @code{(:conc-name p-)} | |
4579 would change this prefix to @code{p-}. Specifying @code{nil} as an | |
4580 argument means no prefix, so that the slot names themselves are used | |
4581 to name the accessor functions. | |
4582 | |
4583 @item :constructor | |
4584 In the simple case, this option takes one argument which is an | |
4585 alternate name to use for the constructor function. The default | |
4586 is @code{make-@var{name}}, e.g., @code{make-person}. The above | |
4587 example changes this to @code{create-person}. Specifying @code{nil} | |
4588 as an argument means that no standard constructor should be | |
4589 generated at all. | |
4590 | |
4591 In the full form of this option, the constructor name is followed | |
4592 by an arbitrary argument list. @xref{Program Structure}, for a | |
4593 description of the format of Common Lisp argument lists. All | |
4594 options, such as @code{&rest} and @code{&key}, are supported. | |
4595 The argument names should match the slot names; each slot is | |
4596 initialized from the corresponding argument. Slots whose names | |
4597 do not appear in the argument list are initialized based on the | |
4598 @var{default-value} in their slot descriptor. Also, @code{&optional} | |
4599 and @code{&key} arguments which don't specify defaults take their | |
4600 defaults from the slot descriptor. It is legal to include arguments | |
4601 which don't correspond to slot names; these are useful if they are | |
4602 referred to in the defaults for optional, keyword, or @code{&aux} | |
4603 arguments which @emph{do} correspond to slots. | |
4604 | |
4605 You can specify any number of full-format @code{:constructor} | |
4606 options on a structure. The default constructor is still generated | |
4607 as well unless you disable it with a simple-format @code{:constructor} | |
4608 option. | |
4609 | |
4610 @example | |
4611 (defstruct | |
4612 (person | |
4613 (:constructor nil) ; no default constructor | |
4614 (:constructor new-person (name sex &optional (age 0))) | |
4615 (:constructor new-hound (&key (name "Rover") | |
4616 (dog-years 0) | |
4617 &aux (age (* 7 dog-years)) | |
4618 (sex 'canine)))) | |
4619 name age sex) | |
4620 @end example | |
4621 | |
4622 The first constructor here takes its arguments positionally rather | |
4623 than by keyword. (In official Common Lisp terminology, constructors | |
4624 that work By Order of Arguments instead of by keyword are called | |
4625 ``BOA constructors.'' No, I'm not making this up.) For example, | |
4626 @code{(new-person "Jane" 'female)} generates a person whose slots | |
4627 are @code{"Jane"}, 0, and @code{female}, respectively. | |
4628 | |
4629 The second constructor takes two keyword arguments, @code{:name}, | |
4630 which initializes the @code{name} slot and defaults to @code{"Rover"}, | |
4631 and @code{:dog-years}, which does not itself correspond to a slot | |
4632 but which is used to initialize the @code{age} slot. The @code{sex} | |
4633 slot is forced to the symbol @code{canine} with no syntax for | |
4634 overriding it. | |
4635 | |
4636 @item :copier | |
4637 The argument is an alternate name for the copier function for | |
4638 this type. The default is @code{copy-@var{name}}. @code{nil} | |
4639 means not to generate a copier function. (In this implementation, | |
4640 all copier functions are simply synonyms for @code{copy-sequence}.) | |
4641 | |
4642 @item :predicate | |
4643 The argument is an alternate name for the predicate which recognizes | |
4644 objects of this type. The default is @code{@var{name}-p}. @code{nil} | |
4645 means not to generate a predicate function. (If the @code{:type} | |
4646 option is used without the @code{:named} option, no predicate is | |
4647 ever generated.) | |
4648 | |
4649 In true Common Lisp, @code{typep} is always able to recognize a | |
4650 structure object even if @code{:predicate} was used. In this | |
4651 package, @code{typep} simply looks for a function called | |
4652 @code{@var{typename}-p}, so it will work for structure types | |
4653 only if they used the default predicate name. | |
4654 | |
4655 @item :include | |
4656 This option implements a very limited form of C++-style inheritance. | |
4657 The argument is the name of another structure type previously | |
4658 created with @code{defstruct}. The effect is to cause the new | |
4659 structure type to inherit all of the included structure's slots | |
4660 (plus, of course, any new slots described by this struct's slot | |
4661 descriptors). The new structure is considered a ``specialization'' | |
4662 of the included one. In fact, the predicate and slot accessors | |
4663 for the included type will also accept objects of the new type. | |
4664 | |
4665 If there are extra arguments to the @code{:include} option after | |
4666 the included-structure name, these options are treated as replacement | |
4667 slot descriptors for slots in the included structure, possibly with | |
4668 modified default values. Borrowing an example from Steele: | |
4669 | |
4670 @example | |
4671 (defstruct person name (age 0) sex) | |
4672 @result{} person | |
4673 (defstruct (astronaut (:include person (age 45))) | |
4674 helmet-size | |
4675 (favorite-beverage 'tang)) | |
4676 @result{} astronaut | |
4677 | |
4678 (setq joe (make-person :name "Joe")) | |
4679 @result{} [cl-struct-person "Joe" 0 nil] | |
4680 (setq buzz (make-astronaut :name "Buzz")) | |
4681 @result{} [cl-struct-astronaut "Buzz" 45 nil nil tang] | |
4682 | |
4683 (list (person-p joe) (person-p buzz)) | |
4684 @result{} (t t) | |
4685 (list (astronaut-p joe) (astronaut-p buzz)) | |
4686 @result{} (nil t) | |
4687 | |
4688 (person-name buzz) | |
4689 @result{} "Buzz" | |
4690 (astronaut-name joe) | |
4691 @result{} error: "astronaut-name accessing a non-astronaut" | |
4692 @end example | |
4693 | |
4694 Thus, if @code{astronaut} is a specialization of @code{person}, | |
4695 then every @code{astronaut} is also a @code{person} (but not the | |
4696 other way around). Every @code{astronaut} includes all the slots | |
4697 of a @code{person}, plus extra slots that are specific to | |
4698 astronauts. Operations that work on people (like @code{person-name}) | |
4699 work on astronauts just like other people. | |
4700 | |
4701 @item :print-function | |
4702 In full Common Lisp, this option allows you to specify a function | |
4703 which is called to print an instance of the structure type. The | |
4704 Emacs Lisp system offers no hooks into the Lisp printer which would | |
4705 allow for such a feature, so this package simply ignores | |
4706 @code{:print-function}. | |
4707 | |
4708 @item :type | |
4709 The argument should be one of the symbols @code{vector} or @code{list}. | |
4710 This tells which underlying Lisp data type should be used to implement | |
4711 the new structure type. Vectors are used by default, but | |
4712 @code{(:type list)} will cause structure objects to be stored as | |
4713 lists instead. | |
4714 | |
4715 The vector representation for structure objects has the advantage | |
4716 that all structure slots can be accessed quickly, although creating | |
4717 vectors is a bit slower in Emacs Lisp. Lists are easier to create, | |
4718 but take a relatively long time accessing the later slots. | |
4719 | |
4720 @item :named | |
4721 This option, which takes no arguments, causes a characteristic ``tag'' | |
4722 symbol to be stored at the front of the structure object. Using | |
4723 @code{:type} without also using @code{:named} will result in a | |
4724 structure type stored as plain vectors or lists with no identifying | |
4725 features. | |
4726 | |
4727 The default, if you don't specify @code{:type} explicitly, is to | |
4728 use named vectors. Therefore, @code{:named} is only useful in | |
4729 conjunction with @code{:type}. | |
4730 | |
4731 @example | |
4732 (defstruct (person1) name age sex) | |
4733 (defstruct (person2 (:type list) :named) name age sex) | |
4734 (defstruct (person3 (:type list)) name age sex) | |
4735 | |
4736 (setq p1 (make-person1)) | |
4737 @result{} [cl-struct-person1 nil nil nil] | |
4738 (setq p2 (make-person2)) | |
4739 @result{} (person2 nil nil nil) | |
4740 (setq p3 (make-person3)) | |
4741 @result{} (nil nil nil) | |
4742 | |
4743 (person1-p p1) | |
4744 @result{} t | |
4745 (person2-p p2) | |
4746 @result{} t | |
4747 (person3-p p3) | |
4748 @result{} error: function person3-p undefined | |
4749 @end example | |
4750 | |
4751 Since unnamed structures don't have tags, @code{defstruct} is not | |
4752 able to make a useful predicate for recognizing them. Also, | |
4753 accessors like @code{person3-name} will be generated but they | |
4754 will not be able to do any type checking. The @code{person3-name} | |
4755 function, for example, will simply be a synonym for @code{car} in | |
4756 this case. By contrast, @code{person2-name} is able to verify | |
4757 that its argument is indeed a @code{person2} object before | |
4758 proceeding. | |
4759 | |
4760 @item :initial-offset | |
4761 The argument must be a nonnegative integer. It specifies a | |
4762 number of slots to be left ``empty'' at the front of the | |
4763 structure. If the structure is named, the tag appears at the | |
4764 specified position in the list or vector; otherwise, the first | |
4765 slot appears at that position. Earlier positions are filled | |
4766 with @code{nil} by the constructors and ignored otherwise. If | |
4767 the type @code{:include}s another type, then @code{:initial-offset} | |
4768 specifies a number of slots to be skipped between the last slot | |
4769 of the included type and the first new slot. | |
4770 @end table | |
4771 @end defspec | |
4772 | |
4773 Except as noted, the @code{defstruct} facility of this package is | |
4774 entirely compatible with that of Common Lisp. | |
4775 | |
4776 @iftex | |
4777 @chapno=23 | |
4778 @end iftex | |
4779 | |
4780 @node Assertions, Efficiency Concerns, Structures, Top | |
4781 @chapter Assertions and Errors | |
4782 | |
4783 @noindent | |
4784 This section describes two macros that test @dfn{assertions}, i.e., | |
4785 conditions which must be true if the program is operating correctly. | |
4786 Assertions never add to the behavior of a Lisp program; they simply | |
4787 make ``sanity checks'' to make sure everything is as it should be. | |
4788 | |
4789 If the optimization property @code{speed} has been set to 3, and | |
4790 @code{safety} is less than 3, then the byte-compiler will optimize | |
4791 away the following assertions. Because assertions might be optimized | |
4792 away, it is a bad idea for them to include side-effects. | |
4793 | |
4794 @defspec assert test-form [show-args string args@dots{}] | |
4795 This form verifies that @var{test-form} is true (i.e., evaluates to | |
4796 a non-@code{nil} value). If so, it returns @code{nil}. If the test | |
4797 is not satisfied, @code{assert} signals an error. | |
4798 | |
4799 A default error message will be supplied which includes @var{test-form}. | |
4800 You can specify a different error message by including a @var{string} | |
4801 argument plus optional extra arguments. Those arguments are simply | |
4802 passed to @code{error} to signal the error. | |
4803 | |
4804 If the optional second argument @var{show-args} is @code{t} instead | |
4805 of @code{nil}, then the error message (with or without @var{string}) | |
4806 will also include all non-constant arguments of the top-level | |
4807 @var{form}. For example: | |
4808 | |
4809 @example | |
4810 (assert (> x 10) t "x is too small: %d") | |
4811 @end example | |
4812 | |
4813 This usage of @var{show-args} is an extension to Common Lisp. In | |
4814 true Common Lisp, the second argument gives a list of @var{places} | |
4815 which can be @code{setf}'d by the user before continuing from the | |
4816 error. Since Emacs Lisp does not support continuable errors, it | |
4817 makes no sense to specify @var{places}. | |
4818 @end defspec | |
4819 | |
4820 @defspec check-type form type [string] | |
4821 This form verifies that @var{form} evaluates to a value of type | |
4822 @var{type}. If so, it returns @code{nil}. If not, @code{check-type} | |
4823 signals a @code{wrong-type-argument} error. The default error message | |
4824 lists the erroneous value along with @var{type} and @var{form} | |
4825 themselves. If @var{string} is specified, it is included in the | |
4826 error message in place of @var{type}. For example: | |
4827 | |
4828 @example | |
4829 (check-type x (integer 1 *) "a positive integer") | |
4830 @end example | |
4831 | |
4832 @xref{Type Predicates}, for a description of the type specifiers | |
4833 that may be used for @var{type}. | |
4834 | |
4835 Note that in Common Lisp, the first argument to @code{check-type} | |
4836 must be a @var{place} suitable for use by @code{setf}, because | |
4837 @code{check-type} signals a continuable error that allows the | |
4838 user to modify @var{place}. | |
4839 @end defspec | |
4840 | |
4841 The following error-related macro is also defined: | |
4842 | |
4843 @defspec ignore-errors forms@dots{} | |
4844 This executes @var{forms} exactly like a @code{progn}, except that | |
4845 errors are ignored during the @var{forms}. More precisely, if | |
4846 an error is signaled then @code{ignore-errors} immediately | |
4847 aborts execution of the @var{forms} and returns @code{nil}. | |
4848 If the @var{forms} complete successfully, @code{ignore-errors} | |
4849 returns the result of the last @var{form}. | |
4850 @end defspec | |
4851 | |
4852 @node Efficiency Concerns, Common Lisp Compatibility, Assertions, Top | |
4853 @appendix Efficiency Concerns | |
4854 | |
4855 @appendixsec Macros | |
4856 | |
4857 @noindent | |
4858 Many of the advanced features of this package, such as @code{defun*}, | |
4859 @code{loop}, and @code{setf}, are implemented as Lisp macros. In | |
4860 byte-compiled code, these complex notations will be expanded into | |
4861 equivalent Lisp code which is simple and efficient. For example, | |
4862 the forms | |
4863 | |
4864 @example | |
4865 (incf i n) | |
4866 (push x (car p)) | |
4867 @end example | |
4868 | |
4869 @noindent | |
4870 are expanded at compile-time to the Lisp forms | |
4871 | |
4872 @example | |
4873 (setq i (+ i n)) | |
4874 (setcar p (cons x (car p))) | |
4875 @end example | |
4876 | |
4877 @noindent | |
4878 which are the most efficient ways of doing these respective operations | |
4879 in Lisp. Thus, there is no performance penalty for using the more | |
4880 readable @code{incf} and @code{push} forms in your compiled code. | |
4881 | |
4882 @emph{Interpreted} code, on the other hand, must expand these macros | |
4883 every time they are executed. For this reason it is strongly | |
4884 recommended that code making heavy use of macros be compiled. | |
4885 (The features labeled ``Special Form'' instead of ``Function'' in | |
4886 this manual are macros.) A loop using @code{incf} a hundred times | |
4887 will execute considerably faster if compiled, and will also | |
4888 garbage-collect less because the macro expansion will not have | |
4889 to be generated, used, and thrown away a hundred times. | |
4890 | |
4891 You can find out how a macro expands by using the | |
4892 @code{cl-prettyexpand} function. | |
4893 | |
4894 @defun cl-prettyexpand form &optional full | |
4895 This function takes a single Lisp form as an argument and inserts | |
4896 a nicely formatted copy of it in the current buffer (which must be | |
4897 in Lisp mode so that indentation works properly). It also expands | |
4898 all Lisp macros which appear in the form. The easiest way to use | |
4899 this function is to go to the @code{*scratch*} buffer and type, say, | |
4900 | |
4901 @example | |
4902 (cl-prettyexpand '(loop for x below 10 collect x)) | |
4903 @end example | |
4904 | |
4905 @noindent | |
4906 and type @kbd{C-x C-e} immediately after the closing parenthesis; | |
4907 the expansion | |
4908 | |
4909 @example | |
4910 (block nil | |
4911 (let* ((x 0) | |
4912 (G1004 nil)) | |
4913 (while (< x 10) | |
4914 (setq G1004 (cons x G1004)) | |
4915 (setq x (+ x 1))) | |
4916 (nreverse G1004))) | |
4917 @end example | |
4918 | |
4919 @noindent | |
4920 will be inserted into the buffer. (The @code{block} macro is | |
4921 expanded differently in the interpreter and compiler, so | |
4922 @code{cl-prettyexpand} just leaves it alone. The temporary | |
4923 variable @code{G1004} was created by @code{gensym}.) | |
4924 | |
4925 If the optional argument @var{full} is true, then @emph{all} | |
4926 macros are expanded, including @code{block}, @code{eval-when}, | |
4927 and compiler macros. Expansion is done as if @var{form} were | |
4928 a top-level form in a file being compiled. For example, | |
4929 | |
4930 @example | |
4931 (cl-prettyexpand '(pushnew 'x list)) | |
4932 @print{} (setq list (adjoin 'x list)) | |
4933 (cl-prettyexpand '(pushnew 'x list) t) | |
4934 @print{} (setq list (if (memq 'x list) list (cons 'x list))) | |
4935 (cl-prettyexpand '(caddr (member* 'a list)) t) | |
4936 @print{} (car (cdr (cdr (memq 'a list)))) | |
4937 @end example | |
4938 | |
4939 Note that @code{adjoin}, @code{caddr}, and @code{member*} all | |
4940 have built-in compiler macros to optimize them in common cases. | |
4941 @end defun | |
4942 | |
4943 @ifinfo | |
4944 @example | |
4945 | |
4946 @end example | |
4947 @end ifinfo | |
4948 @appendixsec Error Checking | |
4949 | |
4950 @noindent | |
4951 Common Lisp compliance has in general not been sacrificed for the | |
4952 sake of efficiency. A few exceptions have been made for cases | |
4953 where substantial gains were possible at the expense of marginal | |
33295
d66d9f108b83
Remove obsolete warnings about use of memq for keyword args.
Stefan Monnier <monnier@iro.umontreal.ca>
parents:
32315
diff
changeset
|
4954 incompatibility. |
25829 | 4955 |
4956 The Common Lisp standard (as embodied in Steele's book) uses the | |
4957 phrase ``it is an error if'' to indicate a situation which is not | |
4958 supposed to arise in complying programs; implementations are strongly | |
4959 encouraged but not required to signal an error in these situations. | |
4960 This package sometimes omits such error checking in the interest of | |
4961 compactness and efficiency. For example, @code{do} variable | |
4962 specifiers are supposed to be lists of one, two, or three forms; | |
4963 extra forms are ignored by this package rather than signaling a | |
4964 syntax error. The @code{endp} function is simply a synonym for | |
4965 @code{null} in this package. Functions taking keyword arguments | |
4966 will accept an odd number of arguments, treating the trailing | |
4967 keyword as if it were followed by the value @code{nil}. | |
4968 | |
4969 Argument lists (as processed by @code{defun*} and friends) | |
4970 @emph{are} checked rigorously except for the minor point just | |
4971 mentioned; in particular, keyword arguments are checked for | |
4972 validity, and @code{&allow-other-keys} and @code{:allow-other-keys} | |
4973 are fully implemented. Keyword validity checking is slightly | |
4974 time consuming (though not too bad in byte-compiled code); | |
4975 you can use @code{&allow-other-keys} to omit this check. Functions | |
4976 defined in this package such as @code{find} and @code{member*} | |
4977 do check their keyword arguments for validity. | |
4978 | |
4979 @ifinfo | |
4980 @example | |
4981 | |
4982 @end example | |
4983 @end ifinfo | |
4984 @appendixsec Optimizing Compiler | |
4985 | |
4986 @noindent | |
31572
384603523f41
Remove obsolescences, especially hash tables and refs to Emacs 19.
Dave Love <fx@gnu.org>
parents:
30009
diff
changeset
|
4987 Use of the optimizing Emacs compiler is highly recommended; many of the Common |
27225
ee242fddb1ab
Remove various old stuff and references thereto.
Dave Love <fx@gnu.org>
parents:
25829
diff
changeset
|
4988 Lisp macros emit |
25829 | 4989 code which can be improved by optimization. In particular, |
4990 @code{block}s (whether explicit or implicit in constructs like | |
4991 @code{defun*} and @code{loop}) carry a fair run-time penalty; the | |
4992 optimizing compiler removes @code{block}s which are not actually | |
4993 referenced by @code{return} or @code{return-from} inside the block. | |
4994 | |
4995 @node Common Lisp Compatibility, Old CL Compatibility, Efficiency Concerns, Top | |
4996 @appendix Common Lisp Compatibility | |
4997 | |
4998 @noindent | |
4999 Following is a list of all known incompatibilities between this | |
5000 package and Common Lisp as documented in Steele (2nd edition). | |
5001 | |
5002 Certain function names, such as @code{member}, @code{assoc}, and | |
5003 @code{floor}, were already taken by (incompatible) Emacs Lisp | |
5004 functions; this package appends @samp{*} to the names of its | |
5005 Common Lisp versions of these functions. | |
5006 | |
5007 The word @code{defun*} is required instead of @code{defun} in order | |
5008 to use extended Common Lisp argument lists in a function. Likewise, | |
5009 @code{defmacro*} and @code{function*} are versions of those forms | |
5010 which understand full-featured argument lists. The @code{&whole} | |
5011 keyword does not work in @code{defmacro} argument lists (except | |
5012 inside recursive argument lists). | |
5013 | |
5014 The @code{eql} and @code{equal} predicates do not distinguish | |
5015 between IEEE floating-point plus and minus zero. The @code{equalp} | |
5016 predicate has several differences with Common Lisp; @pxref{Predicates}. | |
5017 | |
5018 The @code{setf} mechanism is entirely compatible, except that | |
5019 setf-methods return a list of five values rather than five | |
5020 values directly. Also, the new ``@code{setf} function'' concept | |
5021 (typified by @code{(defun (setf foo) @dots{})}) is not implemented. | |
5022 | |
5023 The @code{do-all-symbols} form is the same as @code{do-symbols} | |
5024 with no @var{obarray} argument. In Common Lisp, this form would | |
5025 iterate over all symbols in all packages. Since Emacs obarrays | |
5026 are not a first-class package mechanism, there is no way for | |
5027 @code{do-all-symbols} to locate any but the default obarray. | |
5028 | |
5029 The @code{loop} macro is complete except that @code{loop-finish} | |
5030 and type specifiers are unimplemented. | |
5031 | |
5032 The multiple-value return facility treats lists as multiple | |
5033 values, since Emacs Lisp cannot support multiple return values | |
5034 directly. The macros will be compatible with Common Lisp if | |
5035 @code{values} or @code{values-list} is always used to return to | |
5036 a @code{multiple-value-bind} or other multiple-value receiver; | |
5037 if @code{values} is used without @code{multiple-value-@dots{}} | |
5038 or vice-versa the effect will be different from Common Lisp. | |
5039 | |
5040 Many Common Lisp declarations are ignored, and others match | |
5041 the Common Lisp standard in concept but not in detail. For | |
5042 example, local @code{special} declarations, which are purely | |
5043 advisory in Emacs Lisp, do not rigorously obey the scoping rules | |
5044 set down in Steele's book. | |
5045 | |
5046 The variable @code{*gensym-counter*} starts out with a pseudo-random | |
5047 value rather than with zero. This is to cope with the fact that | |
5048 generated symbols become interned when they are written to and | |
5049 loaded back from a file. | |
5050 | |
5051 The @code{defstruct} facility is compatible, except that structures | |
5052 are of type @code{:type vector :named} by default rather than some | |
5053 special, distinct type. Also, the @code{:type} slot option is ignored. | |
5054 | |
5055 The second argument of @code{check-type} is treated differently. | |
5056 | |
5057 @node Old CL Compatibility, Porting Common Lisp, Common Lisp Compatibility, Top | |
5058 @appendix Old CL Compatibility | |
5059 | |
5060 @noindent | |
5061 Following is a list of all known incompatibilities between this package | |
5062 and the older Quiroz @file{cl.el} package. | |
5063 | |
5064 This package's emulation of multiple return values in functions is | |
5065 incompatible with that of the older package. That package attempted | |
5066 to come as close as possible to true Common Lisp multiple return | |
5067 values; unfortunately, it could not be 100% reliable and so was prone | |
5068 to occasional surprises if used freely. This package uses a simpler | |
5069 method, namely replacing multiple values with lists of values, which | |
5070 is more predictable though more noticeably different from Common Lisp. | |
5071 | |
5072 The @code{defkeyword} form and @code{keywordp} function are not | |
5073 implemented in this package. | |
5074 | |
5075 The @code{member}, @code{floor}, @code{ceiling}, @code{truncate}, | |
5076 @code{round}, @code{mod}, and @code{rem} functions are suffixed | |
5077 by @samp{*} in this package to avoid collision with existing | |
27225
ee242fddb1ab
Remove various old stuff and references thereto.
Dave Love <fx@gnu.org>
parents:
25829
diff
changeset
|
5078 functions in Emacs. The older package simply |
25829 | 5079 redefined these functions, overwriting the built-in meanings and |
31572
384603523f41
Remove obsolescences, especially hash tables and refs to Emacs 19.
Dave Love <fx@gnu.org>
parents:
30009
diff
changeset
|
5080 causing serious portability problems. (Some more |
25829 | 5081 recent versions of the Quiroz package changed the names to |
5082 @code{cl-member}, etc.; this package defines the latter names as | |
5083 aliases for @code{member*}, etc.) | |
5084 | |
5085 Certain functions in the old package which were buggy or inconsistent | |
5086 with the Common Lisp standard are incompatible with the conforming | |
5087 versions in this package. For example, @code{eql} and @code{member} | |
5088 were synonyms for @code{eq} and @code{memq} in that package, @code{setf} | |
5089 failed to preserve correct order of evaluation of its arguments, etc. | |
5090 | |
5091 Finally, unlike the older package, this package is careful to | |
5092 prefix all of its internal names with @code{cl-}. Except for a | |
5093 few functions which are explicitly defined as additional features | |
5094 (such as @code{floatp-safe} and @code{letf}), this package does not | |
5095 export any non-@samp{cl-} symbols which are not also part of Common | |
5096 Lisp. | |
5097 | |
5098 @ifinfo | |
5099 @example | |
5100 | |
5101 @end example | |
5102 @end ifinfo | |
5103 @appendixsec The @code{cl-compat} package | |
5104 | |
5105 @noindent | |
5106 The @dfn{CL} package includes emulations of some features of the | |
5107 old @file{cl.el}, in the form of a compatibility package | |
5108 @code{cl-compat}. To use it, put @code{(require 'cl-compat)} in | |
5109 your program. | |
5110 | |
5111 The old package defined a number of internal routines without | |
5112 @code{cl-} prefixes or other annotations. Call to these routines | |
5113 may have crept into existing Lisp code. @code{cl-compat} | |
5114 provides emulations of the following internal routines: | |
5115 @code{pair-with-newsyms}, @code{zip-lists}, @code{unzip-lists}, | |
5116 @code{reassemble-arglists}, @code{duplicate-symbols-p}, | |
5117 @code{safe-idiv}. | |
5118 | |
5119 Some @code{setf} forms translated into calls to internal | |
5120 functions that user code might call directly. The functions | |
5121 @code{setnth}, @code{setnthcdr}, and @code{setelt} fall in | |
5122 this category; they are defined by @code{cl-compat}, but the | |
5123 best fix is to change to use @code{setf} properly. | |
5124 | |
5125 The @code{cl-compat} file defines the keyword functions | |
5126 @code{keywordp}, @code{keyword-of}, and @code{defkeyword}, | |
5127 which are not defined by the new @dfn{CL} package because the | |
5128 use of keywords as data is discouraged. | |
5129 | |
5130 The @code{build-klist} mechanism for parsing keyword arguments | |
5131 is emulated by @code{cl-compat}; the @code{with-keyword-args} | |
5132 macro is not, however, and in any case it's best to change to | |
5133 use the more natural keyword argument processing offered by | |
5134 @code{defun*}. | |
5135 | |
5136 Multiple return values are treated differently by the two | |
5137 Common Lisp packages. The old package's method was more | |
5138 compatible with true Common Lisp, though it used heuristics | |
5139 that caused it to report spurious multiple return values in | |
5140 certain cases. The @code{cl-compat} package defines a set | |
5141 of multiple-value macros that are compatible with the old | |
5142 CL package; again, they are heuristic in nature, but they | |
5143 are guaranteed to work in any case where the old package's | |
5144 macros worked. To avoid name collision with the ``official'' | |
5145 multiple-value facilities, the ones in @code{cl-compat} have | |
5146 capitalized names: @code{Values}, @code{Values-list}, | |
5147 @code{Multiple-value-bind}, etc. | |
5148 | |
5149 The functions @code{cl-floor}, @code{cl-ceiling}, @code{cl-truncate}, | |
5150 and @code{cl-round} are defined by @code{cl-compat} to use the | |
5151 old-style multiple-value mechanism, just as they did in the old | |
5152 package. The newer @code{floor*} and friends return their two | |
5153 results in a list rather than as multiple values. Note that | |
5154 older versions of the old package used the unadorned names | |
5155 @code{floor}, @code{ceiling}, etc.; @code{cl-compat} cannot use | |
31572
384603523f41
Remove obsolescences, especially hash tables and refs to Emacs 19.
Dave Love <fx@gnu.org>
parents:
30009
diff
changeset
|
5156 these names because they conflict with Emacs built-ins. |
25829 | 5157 |
5158 @node Porting Common Lisp, Function Index, Old CL Compatibility, Top | |
5159 @appendix Porting Common Lisp | |
5160 | |
5161 @noindent | |
5162 This package is meant to be used as an extension to Emacs Lisp, | |
5163 not as an Emacs implementation of true Common Lisp. Some of the | |
5164 remaining differences between Emacs Lisp and Common Lisp make it | |
5165 difficult to port large Common Lisp applications to Emacs. For | |
5166 one, some of the features in this package are not fully compliant | |
5167 with ANSI or Steele; @pxref{Common Lisp Compatibility}. But there | |
5168 are also quite a few features that this package does not provide | |
5169 at all. Here are some major omissions that you will want watch out | |
5170 for when bringing Common Lisp code into Emacs. | |
5171 | |
5172 @itemize @bullet | |
5173 @item | |
5174 Case-insensitivity. Symbols in Common Lisp are case-insensitive | |
5175 by default. Some programs refer to a function or variable as | |
5176 @code{foo} in one place and @code{Foo} or @code{FOO} in another. | |
5177 Emacs Lisp will treat these as three distinct symbols. | |
5178 | |
5179 Some Common Lisp code is written entirely in upper case. While Emacs | |
5180 is happy to let the program's own functions and variables use | |
5181 this convention, calls to Lisp builtins like @code{if} and | |
5182 @code{defun} will have to be changed to lower case. | |
5183 | |
5184 @item | |
5185 Lexical scoping. In Common Lisp, function arguments and @code{let} | |
5186 bindings apply only to references physically within their bodies | |
5187 (or within macro expansions in their bodies). Emacs Lisp, by | |
5188 contrast, uses @dfn{dynamic scoping} wherein a binding to a | |
5189 variable is visible even inside functions called from the body. | |
5190 | |
5191 Variables in Common Lisp can be made dynamically scoped by | |
5192 declaring them @code{special} or using @code{defvar}. In Emacs | |
5193 Lisp it is as if all variables were declared @code{special}. | |
5194 | |
5195 Often you can use code that was written for lexical scoping | |
5196 even in a dynamically scoped Lisp, but not always. Here is | |
5197 an example of a Common Lisp code fragment that would fail in | |
5198 Emacs Lisp: | |
5199 | |
5200 @example | |
5201 (defun map-odd-elements (func list) | |
5202 (loop for x in list | |
5203 for flag = t then (not flag) | |
5204 collect (if flag x (funcall func x)))) | |
5205 | |
5206 (defun add-odd-elements (list x) | |
28344
54fda0e8528a
Weed out redundant uses of `function'
Stefan Monnier <monnier@iro.umontreal.ca>
parents:
28039
diff
changeset
|
5207 (map-odd-elements (lambda (a) (+ a x))) list) |
25829 | 5208 @end example |
5209 | |
5210 @noindent | |
5211 In Common Lisp, the two functions' usages of @code{x} are completely | |
5212 independent. In Emacs Lisp, the binding to @code{x} made by | |
5213 @code{add-odd-elements} will have been hidden by the binding | |
5214 in @code{map-odd-elements} by the time the @code{(+ a x)} function | |
5215 is called. | |
5216 | |
5217 (This package avoids such problems in its own mapping functions | |
5218 by using names like @code{cl-x} instead of @code{x} internally; | |
5219 as long as you don't use the @code{cl-} prefix for your own | |
5220 variables no collision can occur.) | |
5221 | |
5222 @xref{Lexical Bindings}, for a description of the @code{lexical-let} | |
5223 form which establishes a Common Lisp-style lexical binding, and some | |
5224 examples of how it differs from Emacs' regular @code{let}. | |
5225 | |
5226 @item | |
5227 Reader macros. Common Lisp includes a second type of macro that | |
5228 works at the level of individual characters. For example, Common | |
5229 Lisp implements the quote notation by a reader macro called @code{'}, | |
5230 whereas Emacs Lisp's parser just treats quote as a special case. | |
5231 Some Lisp packages use reader macros to create special syntaxes | |
5232 for themselves, which the Emacs parser is incapable of reading. | |
5233 | |
5234 The lack of reader macros, incidentally, is the reason behind | |
5235 Emacs Lisp's unusual backquote syntax. Since backquotes are | |
5236 implemented as a Lisp package and not built-in to the Emacs | |
5237 parser, they are forced to use a regular macro named @code{`} | |
5238 which is used with the standard function/macro call notation. | |
5239 | |
5240 @item | |
5241 Other syntactic features. Common Lisp provides a number of | |
5242 notations beginning with @code{#} that the Emacs Lisp parser | |
5243 won't understand. For example, @samp{#| ... |#} is an | |
5244 alternate comment notation, and @samp{#+lucid (foo)} tells | |
5245 the parser to ignore the @code{(foo)} except in Lucid Common | |
5246 Lisp. | |
5247 | |
5248 @item | |
5249 Packages. In Common Lisp, symbols are divided into @dfn{packages}. | |
5250 Symbols that are Lisp built-ins are typically stored in one package; | |
5251 symbols that are vendor extensions are put in another, and each | |
5252 application program would have a package for its own symbols. | |
5253 Certain symbols are ``exported'' by a package and others are | |
5254 internal; certain packages ``use'' or import the exported symbols | |
5255 of other packages. To access symbols that would not normally be | |
5256 visible due to this importing and exporting, Common Lisp provides | |
5257 a syntax like @code{package:symbol} or @code{package::symbol}. | |
5258 | |
5259 Emacs Lisp has a single namespace for all interned symbols, and | |
5260 then uses a naming convention of putting a prefix like @code{cl-} | |
5261 in front of the name. Some Emacs packages adopt the Common Lisp-like | |
5262 convention of using @code{cl:} or @code{cl::} as the prefix. | |
5263 However, the Emacs parser does not understand colons and just | |
5264 treats them as part of the symbol name. Thus, while @code{mapcar} | |
5265 and @code{lisp:mapcar} may refer to the same symbol in Common | |
5266 Lisp, they are totally distinct in Emacs Lisp. Common Lisp | |
5267 programs which refer to a symbol by the full name sometimes | |
5268 and the short name other times will not port cleanly to Emacs. | |
5269 | |
5270 Emacs Lisp does have a concept of ``obarrays,'' which are | |
5271 package-like collections of symbols, but this feature is not | |
5272 strong enough to be used as a true package mechanism. | |
5273 | |
5274 @item | |
5275 The @code{format} function is quite different between Common | |
5276 Lisp and Emacs Lisp. It takes an additional ``destination'' | |
5277 argument before the format string. A destination of @code{nil} | |
5278 means to format to a string as in Emacs Lisp; a destination | |
5279 of @code{t} means to write to the terminal (similar to | |
5280 @code{message} in Emacs). Also, format control strings are | |
5281 utterly different; @code{~} is used instead of @code{%} to | |
5282 introduce format codes, and the set of available codes is | |
5283 much richer. There are no notations like @code{\n} for | |
5284 string literals; instead, @code{format} is used with the | |
5285 ``newline'' format code, @code{~%}. More advanced formatting | |
5286 codes provide such features as paragraph filling, case | |
5287 conversion, and even loops and conditionals. | |
5288 | |
5289 While it would have been possible to implement most of Common | |
5290 Lisp @code{format} in this package (under the name @code{format*}, | |
5291 of course), it was not deemed worthwhile. It would have required | |
5292 a huge amount of code to implement even a decent subset of | |
5293 @code{format*}, yet the functionality it would provide over | |
5294 Emacs Lisp's @code{format} would rarely be useful. | |
5295 | |
5296 @item | |
5297 Vector constants use square brackets in Emacs Lisp, but | |
5298 @code{#(a b c)} notation in Common Lisp. To further complicate | |
31572
384603523f41
Remove obsolescences, especially hash tables and refs to Emacs 19.
Dave Love <fx@gnu.org>
parents:
30009
diff
changeset
|
5299 matters, Emacs has its own @code{#(} notation for |
25829 | 5300 something entirely different---strings with properties. |
5301 | |
5302 @item | |
5303 Characters are distinct from integers in Common Lisp. The | |
5304 notation for character constants is also different: @code{#\A} | |
5305 instead of @code{?A}. Also, @code{string=} and @code{string-equal} | |
5306 are synonyms in Emacs Lisp whereas the latter is case-insensitive | |
5307 in Common Lisp. | |
5308 | |
5309 @item | |
5310 Data types. Some Common Lisp data types do not exist in Emacs | |
5311 Lisp. Rational numbers and complex numbers are not present, | |
5312 nor are large integers (all integers are ``fixnums''). All | |
5313 arrays are one-dimensional. There are no readtables or pathnames; | |
5314 streams are a set of existing data types rather than a new data | |
5315 type of their own. Hash tables, random-states, structures, and | |
5316 packages (obarrays) are built from Lisp vectors or lists rather | |
5317 than being distinct types. | |
5318 | |
5319 @item | |
5320 The Common Lisp Object System (CLOS) is not implemented, | |
5321 nor is the Common Lisp Condition System. However, the EIEIO package | |
5322 from @uref{ftp://ftp.ultranet.com/pub/zappo} does implement some | |
5323 CLOS functionality. | |
5324 | |
5325 @item | |
5326 Common Lisp features that are completely redundant with Emacs | |
5327 Lisp features of a different name generally have not been | |
5328 implemented. For example, Common Lisp writes @code{defconstant} | |
5329 where Emacs Lisp uses @code{defconst}. Similarly, @code{make-list} | |
5330 takes its arguments in different ways in the two Lisps but does | |
5331 exactly the same thing, so this package has not bothered to | |
5332 implement a Common Lisp-style @code{make-list}. | |
5333 | |
5334 @item | |
5335 A few more notable Common Lisp features not included in this | |
5336 package: @code{compiler-let}, @code{tagbody}, @code{prog}, | |
5337 @code{ldb/dpb}, @code{parse-integer}, @code{cerror}. | |
5338 | |
5339 @item | |
5340 Recursion. While recursion works in Emacs Lisp just like it | |
5341 does in Common Lisp, various details of the Emacs Lisp system | |
5342 and compiler make recursion much less efficient than it is in | |
5343 most Lisps. Some schools of thought prefer to use recursion | |
5344 in Lisp over other techniques; they would sum a list of | |
5345 numbers using something like | |
5346 | |
5347 @example | |
5348 (defun sum-list (list) | |
5349 (if list | |
5350 (+ (car list) (sum-list (cdr list))) | |
5351 0)) | |
5352 @end example | |
5353 | |
5354 @noindent | |
5355 where a more iteratively-minded programmer might write one of | |
5356 these forms: | |
5357 | |
5358 @example | |
5359 (let ((total 0)) (dolist (x my-list) (incf total x)) total) | |
5360 (loop for x in my-list sum x) | |
5361 @end example | |
5362 | |
5363 While this would be mainly a stylistic choice in most Common Lisps, | |
5364 in Emacs Lisp you should be aware that the iterative forms are | |
5365 much faster than recursion. Also, Lisp programmers will want to | |
5366 note that the current Emacs Lisp compiler does not optimize tail | |
5367 recursion. | |
5368 @end itemize | |
5369 | |
5370 @node Function Index, Variable Index, Porting Common Lisp, Top | |
5371 @unnumbered Function Index | |
5372 | |
5373 @printindex fn | |
5374 | |
31572
384603523f41
Remove obsolescences, especially hash tables and refs to Emacs 19.
Dave Love <fx@gnu.org>
parents:
30009
diff
changeset
|
5375 @node Variable Index, , Function Index, Top |
25829 | 5376 @unnumbered Variable Index |
5377 | |
5378 @printindex vr | |
5379 | |
29713 | 5380 @setchapternewpage odd |
25829 | 5381 @contents |
5382 @bye |