Mercurial > hg > xemacs-beta
annotate man/cl.texi @ 4917:fce43cb76a1c
xlike cleanup, documentation
-------------------- ChangeLog entries follow: --------------------
man/ChangeLog addition:
2010-02-03 Ben Wing <ben@xemacs.org>
* internals/internals.texi (Top):
* internals/internals.texi (Evaluation; Stack Frames; Bindings):
* internals/internals.texi (Ben's README):
* internals/internals.texi (Consoles; Devices; Frames; Windows):
* internals/internals.texi (Window Hierarchy):
* internals/internals.texi (The Window Object):
* internals/internals.texi (Modules for the Basic Displayable Lisp Objects):
* internals/internals.texi (Window-System Support):
* internals/internals.texi (Creating a Window-System Type):
* internals/internals.texi (Discussion -- Garbage Collection):
Update the part at the top about how to maintain the file with
more tips.
Add a chapter on "window-system support" describing in a general
way how the support for different window systems/device types
works, including the separation between device-independent and
device-dependent parts, device methods, the specific device types
and the "xlike" pseudo-type.
src/ChangeLog addition:
2010-02-03 Ben Wing <ben@xemacs.org>
* Makefile.in.in:
* Makefile.in.in (x_objs):
* Makefile.in.in (gtk_gui_objs):
* console-xlike-inc.h:
* depend:
* device-x.c:
* emacs.c:
* gccache-gtk.h:
* gccache-gtk.h (gc_cache_lookup):
* gccache-x.c:
* gccache-x.c (GCCACHE_HASH):
* gccache-x.h:
* toolbar-gtk.c:
* toolbar-gtk.c (gtk_initialize_frame_toolbars):
* toolbar-x.c:
* toolbar-x.c (x_initialize_frame_toolbars):
* toolbar-xlike.c:
* toolbar-xlike.c (xlike_draw_blank_toolbar_button):
* toolbar-xlike.c (xlike_output_toolbar_button):
* toolbar-xlike.c (xlike_get_button_size):
* toolbar-xlike.c (XLIKE_OUTPUT_BUTTONS_LOOP):
* toolbar-xlike.c (xlike_output_toolbar):
* toolbar-xlike.c (xlike_clear_toolbar):
* toolbar-xlike.c (xlike_output_frame_toolbars):
* toolbar-xlike.c (xlike_clear_frame_toolbars):
* toolbar-xlike.c (xlike_redraw_exposed_toolbar):
* toolbar-xlike.c (xlike_redraw_exposed_toolbars):
* toolbar-xlike.c (xlike_redraw_frame_toolbars):
* toolbar-xlike.h:
* toolbar-xlike.h (xlike_clear_frame_toolbars):
Rename some files to make them consistent with general naming rules:
xgccache.c -> gccache-x.c
xgccache.h -> gccache-x.h
toolbar-common.c -> toolbar-xlike.c
toolbar-common.h -> toolbar-xlike.h
Fix include-file references. Also change the names of functions
in now-named toolbar-xlike.c to be xlike_foo() instead of common_foo().
Add a longish comment in console-xlike-inc.h describing the "xlike"
system, how it works and what the various files are used for.
author | Ben Wing <ben@xemacs.org> |
---|---|
date | Wed, 03 Feb 2010 02:46:50 -0600 |
parents | 755ae5b97edb |
children | 378a34562cbe |
rev | line source |
---|---|
428 | 1 \input texinfo @c -*-texinfo-*- |
2 @setfilename ../info/cl.info | |
3 @settitle Common Lisp Extensions | |
4 | |
5 @iftex | |
6 @finalout | |
7 @end iftex | |
8 | |
9 @ifinfo | |
10 @dircategory XEmacs Editor | |
11 @direntry | |
1353 | 12 * Common Lisp: (cl). XEmacs Common Lisp emulation package. |
428 | 13 @end direntry |
14 | |
1353 | 15 This file documents the XEmacs Common Lisp emulation package. |
428 | 16 |
17 Copyright (C) 1993 Free Software Foundation, Inc. | |
18 | |
19 Permission is granted to make and distribute verbatim copies of this | |
20 manual provided the copyright notice and this permission notice are | |
21 preserved on all copies. | |
22 | |
23 @ignore | |
24 Permission is granted to process this file through TeX and print the | |
25 results, provided the printed document carries copying permission notice | |
26 identical to this one except for the removal of this paragraph (this | |
27 paragraph not being relevant to the printed manual). | |
28 | |
29 @end ignore | |
30 Permission is granted to copy and distribute modified versions of this | |
31 manual under the conditions for verbatim copying, provided also that the | |
32 section entitled ``GNU General Public License'' is included exactly as | |
33 in the original, and provided that the entire resulting derived work is | |
34 distributed under the terms of a permission notice identical to this one. | |
35 | |
36 Permission is granted to copy and distribute translations of this manual | |
37 into another language, under the above conditions for modified versions, | |
38 except that the section entitled ``GNU General Public License'' may be | |
39 included in a translation approved by the author instead of in the | |
40 original English. | |
41 @end ifinfo | |
42 | |
43 @titlepage | |
44 @sp 6 | |
45 @center @titlefont{Common Lisp Extensions} | |
46 @sp 4 | |
47 @center For GNU Emacs Lisp | |
48 @sp 1 | |
49 @center Version 2.02 | |
50 @sp 5 | |
51 @center Dave Gillespie | |
52 @center daveg@@synaptics.com | |
53 @page | |
54 | |
55 @vskip 0pt plus 1filll | |
56 Copyright @copyright{} 1993 Free Software Foundation, Inc. | |
57 | |
58 Permission is granted to make and distribute verbatim copies of | |
59 this manual provided the copyright notice and this permission notice | |
60 are preserved on all copies. | |
61 | |
62 @ignore | |
63 Permission is granted to process this file through TeX and print the | |
64 results, provided the printed document carries copying permission notice | |
65 identical to this one except for the removal of this paragraph (this | |
66 paragraph not being relevant to the printed manual). | |
67 | |
68 @end ignore | |
69 Permission is granted to copy and distribute modified versions of this | |
70 manual under the conditions for verbatim copying, provided also that the | |
71 section entitled ``GNU General Public License'' is included exactly as | |
72 in the original, and provided that the entire resulting derived work is | |
73 distributed under the terms of a permission notice identical to this one. | |
74 | |
75 Permission is granted to copy and distribute translations of this manual | |
76 into another language, under the above conditions for modified versions, | |
77 except that the section entitled ``GNU General Public License'' may be | |
78 included in a translation approved by the author instead of in the | |
79 original English. | |
80 @end titlepage | |
81 | |
82 @node Top, Overview,, (dir) | |
83 @chapter Common Lisp Extensions | |
84 | |
85 @noindent | |
86 This document describes a set of Emacs Lisp facilities borrowed from | |
87 Common Lisp. All the facilities are described here in detail; for | |
88 more discussion and examples, Guy L. Steele's @cite{Common Lisp, the | |
89 Language}, second edition, is the definitive book on Common Lisp. | |
90 @iftex | |
91 Chapter numbers and most section numbers of this document parallel | |
92 those of Steele's book. | |
93 @end iftex | |
94 While this document does not assume any prior knowledge of Common | |
95 Lisp, it does assume a basic familiarity with Emacs Lisp. | |
96 | |
97 @menu | |
98 * Overview:: Installation, usage, etc. | |
99 * Program Structure:: Arglists, `eval-when', `defalias' | |
100 * Predicates:: `typep', `eql', and `equalp' | |
101 * Control Structure:: `setf', `when', `do', `loop', etc. | |
102 * Macros:: Destructuring, `define-compiler-macro' | |
103 * Declarations:: `proclaim', `declare', etc. | |
104 * Symbols:: Property lists, `gensym' | |
105 * Numbers:: Predicates, functions, random numbers | |
106 * Sequences:: Mapping, functions, searching, sorting | |
107 * Lists:: `cadr', `sublis', `member*', `assoc*', etc. | |
108 * Hash Tables:: `make-hash-table', `gethash', etc. | |
109 * Structures:: `defstruct' | |
110 * Assertions:: `check-type', `assert', `ignore-errors'. | |
111 | |
112 * Efficiency Concerns:: Hints and techniques | |
113 * Common Lisp Compatibility:: All known differences with Steele | |
114 * Old CL Compatibility:: All known differences with old cl.el | |
115 * Porting Common Lisp:: Hints for porting Common Lisp code | |
116 | |
117 * Function Index:: | |
118 * Variable Index:: | |
119 @end menu | |
120 | |
121 @node Overview, Program Structure, Top, Top | |
122 @ifinfo | |
123 @chapter Overview | |
124 @end ifinfo | |
125 @iftex | |
126 @section Overview | |
127 @end iftex | |
128 | |
129 @noindent | |
130 Common Lisp is a huge language, and Common Lisp systems tend to be | |
131 massive and extremely complex. Emacs Lisp, by contrast, is rather | |
132 minimalist in the choice of Lisp features it offers the programmer. | |
133 As Emacs Lisp programmers have grown in number, and the applications | |
134 they write have grown more ambitious, it has become clear that Emacs | |
135 Lisp could benefit from many of the conveniences of Common Lisp. | |
136 | |
137 The @dfn{CL} package adds a number of Common Lisp functions and | |
138 control structures to Emacs Lisp. While not a 100% complete | |
139 implementation of Common Lisp, @dfn{CL} adds enough functionality | |
140 to make Emacs Lisp programming significantly more convenient. | |
141 | |
142 Some Common Lisp features have been omitted from this package | |
143 for various reasons: | |
144 | |
145 @itemize @bullet | |
146 @item | |
147 Some features are too complex or bulky relative to their benefit | |
148 to Emacs Lisp programmers. CLOS and Common Lisp streams are fine | |
149 examples of this group. | |
150 | |
151 @item | |
152 Other features cannot be implemented without modification to the | |
153 Emacs Lisp interpreter itself, such as multiple return values, | |
154 lexical scoping, case-insensitive symbols, and complex numbers. | |
155 The @dfn{CL} package generally makes no attempt to emulate these | |
156 features. | |
157 | |
158 @item | |
159 Some features conflict with existing things in Emacs Lisp. For | |
160 example, Emacs' @code{assoc} function is incompatible with the | |
161 Common Lisp @code{assoc}. In such cases, this package usually | |
162 adds the suffix @samp{*} to the function name of the Common | |
163 Lisp version of the function (e.g., @code{assoc*}). | |
164 @end itemize | |
165 | |
166 The package described here was written by Dave Gillespie, | |
167 @file{daveg@@synaptics.com}. It is a total rewrite of the original | |
168 1986 @file{cl.el} package by Cesar Quiroz. Most features of | |
169 the Quiroz package have been retained; any incompatibilities are | |
170 noted in the descriptions below. Care has been taken in this | |
171 version to ensure that each function is defined efficiently, | |
172 concisely, and with minimal impact on the rest of the Emacs | |
173 environment. | |
174 | |
175 @menu | |
176 * Usage:: How to use the CL package | |
177 * Organization:: The package's five component files | |
178 * Installation:: Compiling and installing CL | |
179 * Naming Conventions:: Notes on CL function names | |
180 @end menu | |
181 | |
182 @node Usage, Organization, Overview, Overview | |
183 @section Usage | |
184 | |
185 @noindent | |
186 Lisp code that uses features from the @dfn{CL} package should | |
187 include at the beginning: | |
188 | |
189 @example | |
190 (require 'cl) | |
191 @end example | |
192 | |
193 @noindent | |
194 If you want to ensure that the new (Gillespie) version of @dfn{CL} | |
195 is the one that is present, add an additional @code{(require 'cl-19)} | |
196 call: | |
197 | |
198 @example | |
199 (require 'cl) | |
200 (require 'cl-19) | |
201 @end example | |
202 | |
203 @noindent | |
204 The second call will fail (with ``@file{cl-19.el} not found'') if | |
205 the old @file{cl.el} package was in use. | |
206 | |
207 It is safe to arrange to load @dfn{CL} at all times, e.g., | |
208 in your @file{.emacs} file. But it's a good idea, for portability, | |
209 to @code{(require 'cl)} in your code even if you do this. | |
210 | |
211 @node Organization, Installation, Usage, Overview | |
212 @section Organization | |
213 | |
214 @noindent | |
215 The Common Lisp package is organized into four files: | |
216 | |
217 @table @file | |
218 @item cl.el | |
219 This is the ``main'' file, which contains basic functions | |
220 and information about the package. This file is relatively | |
221 compact---about 700 lines. | |
222 | |
223 @item cl-extra.el | |
224 This file contains the larger, more complex or unusual functions. | |
225 It is kept separate so that packages which only want to use Common | |
226 Lisp fundamentals like the @code{cadr} function won't need to pay | |
227 the overhead of loading the more advanced functions. | |
228 | |
229 @item cl-seq.el | |
230 This file contains most of the advanced functions for operating | |
231 on sequences or lists, such as @code{delete-if} and @code{assoc*}. | |
232 | |
233 @item cl-macs.el | |
234 This file contains the features of the packages which are macros | |
235 instead of functions. Macros expand when the caller is compiled, | |
236 not when it is run, so the macros generally only need to be | |
237 present when the byte-compiler is running (or when the macros are | |
238 used in uncompiled code such as a @file{.emacs} file). Most of | |
239 the macros of this package are isolated in @file{cl-macs.el} so | |
240 that they won't take up memory unless you are compiling. | |
241 @end table | |
242 | |
243 The file @file{cl.el} includes all necessary @code{autoload} | |
244 commands for the functions and macros in the other three files. | |
245 All you have to do is @code{(require 'cl)}, and @file{cl.el} | |
246 will take care of pulling in the other files when they are | |
247 needed. | |
248 | |
249 There is another file, @file{cl-compat.el}, which defines some | |
250 routines from the older @file{cl.el} package that are no longer | |
251 present in the new package. This includes internal routines | |
4677
8f1ee2d15784
Support full Common Lisp multiple values in C.
Aidan Kehoe <kehoea@parhasard.net>
parents:
1353
diff
changeset
|
252 like @code{setelt} and @code{zip-lists}, and deprecated features |
8f1ee2d15784
Support full Common Lisp multiple values in C.
Aidan Kehoe <kehoea@parhasard.net>
parents:
1353
diff
changeset
|
253 like @code{defkeyword}. @xref{Old CL Compatibility}. |
428 | 254 |
255 @node Installation, Naming Conventions, Organization, Overview | |
256 @section Installation | |
257 | |
258 @noindent | |
259 Installation of the @dfn{CL} package is simple: Just put the | |
260 byte-compiled files @file{cl.elc}, @file{cl-extra.elc}, | |
261 @file{cl-seq.elc}, @file{cl-macs.elc}, and @file{cl-compat.elc} | |
262 into a directory on your @code{load-path}. | |
263 | |
264 There are no special requirements to compile this package: | |
265 The files do not have to be loaded before they are compiled, | |
266 nor do they need to be compiled in any particular order. | |
267 | |
268 You may choose to put the files into your main @file{lisp/} | |
269 directory, replacing the original @file{cl.el} file there. Or, | |
270 you could put them into a directory that comes before @file{lisp/} | |
271 on your @code{load-path} so that the old @file{cl.el} is | |
272 effectively hidden. | |
273 | |
274 Also, format the @file{cl.texinfo} file and put the resulting | |
275 Info files in the @file{info/} directory or another suitable place. | |
276 | |
277 You may instead wish to leave this package's components all in | |
278 their own directory, and then add this directory to your | |
279 @code{load-path} and (Emacs 19 only) @code{Info-directory-list}. | |
280 Add the directory to the front of the list so the old @dfn{CL} | |
281 package and its documentation are hidden. | |
282 | |
283 @node Naming Conventions, , Installation, Overview | |
284 @section Naming Conventions | |
285 | |
286 @noindent | |
287 Except where noted, all functions defined by this package have the | |
288 same names and calling conventions as their Common Lisp counterparts. | |
289 | |
290 Following is a complete list of functions whose names were changed | |
291 from Common Lisp, usually to avoid conflicts with Emacs. In each | |
292 case, a @samp{*} has been appended to the Common Lisp name to obtain | |
293 the Emacs name: | |
294 | |
295 @example | |
296 defun* defsubst* defmacro* function* | |
440 | 297 member* assoc* rassoc* remove* |
298 delete* mapcar* sort* floor* | |
299 ceiling* truncate* round* mod* | |
300 rem* random* | |
428 | 301 @end example |
302 | |
303 Internal function and variable names in the package are prefixed | |
304 by @code{cl-}. Here is a complete list of functions @emph{not} | |
305 prefixed by @code{cl-} which were not taken from Common Lisp: | |
306 | |
307 @example | |
308 member delete remove remq | |
309 rassoc floatp-safe lexical-let lexical-let* | |
310 callf callf2 letf letf* | |
311 defsubst* defalias add-hook eval-when-compile | |
312 @end example | |
313 | |
314 @noindent | |
315 (Most of these are Emacs 19 features provided to Emacs 18 users, | |
316 or introduced, like @code{remq}, for reasons of symmetry | |
317 with similar features.) | |
318 | |
319 The following simple functions and macros are defined in @file{cl.el}; | |
320 they do not cause other components like @file{cl-extra} to be loaded. | |
321 | |
322 @example | |
323 eql floatp-safe abs endp | |
324 evenp oddp plusp minusp | |
325 last butlast nbutlast caar .. cddddr | |
326 list* ldiff rest first .. tenth | |
327 member [1] copy-list subst mapcar* [2] | |
328 adjoin [3] acons pairlis when | |
329 unless pop [4] push [4] pushnew [3,4] | |
330 incf [4] decf [4] proclaim declaim | |
331 add-hook | |
332 @end example | |
333 | |
334 @noindent | |
335 [1] This is the Emacs 19-compatible function, not @code{member*}. | |
336 | |
337 @noindent | |
338 [2] Only for one sequence argument or two list arguments. | |
339 | |
340 @noindent | |
341 [3] Only if @code{:test} is @code{eq}, @code{equal}, or unspecified, | |
342 and @code{:key} is not used. | |
343 | |
344 @noindent | |
345 [4] Only when @var{place} is a plain variable name. | |
346 | |
347 @iftex | |
348 @chapno=4 | |
349 @end iftex | |
350 | |
351 @node Program Structure, Predicates, Overview, Top | |
352 @chapter Program Structure | |
353 | |
354 @noindent | |
355 This section describes features of the @dfn{CL} package which have to | |
356 do with programs as a whole: advanced argument lists for functions, | |
357 and the @code{eval-when} construct. | |
358 | |
359 @menu | |
360 * Argument Lists:: `&key', `&aux', `defun*', `defmacro*'. | |
361 * Time of Evaluation:: The `eval-when' construct. | |
362 * Function Aliases:: The `defalias' function. | |
363 @end menu | |
364 | |
365 @iftex | |
366 @secno=1 | |
367 @end iftex | |
368 | |
369 @node Argument Lists, Time of Evaluation, Program Structure, Program Structure | |
370 @section Argument Lists | |
371 | |
372 @noindent | |
373 Emacs Lisp's notation for argument lists of functions is a subset of | |
374 the Common Lisp notation. As well as the familiar @code{&optional} | |
375 and @code{&rest} markers, Common Lisp allows you to specify default | |
376 values for optional arguments, and it provides the additional markers | |
377 @code{&key} and @code{&aux}. | |
378 | |
379 Since argument parsing is built-in to Emacs, there is no way for | |
380 this package to implement Common Lisp argument lists seamlessly. | |
381 Instead, this package defines alternates for several Lisp forms | |
382 which you must use if you need Common Lisp argument lists. | |
383 | |
384 @defspec defun* name arglist body... | |
385 This form is identical to the regular @code{defun} form, except | |
386 that @var{arglist} is allowed to be a full Common Lisp argument | |
387 list. Also, the function body is enclosed in an implicit block | |
388 called @var{name}; @pxref{Blocks and Exits}. | |
389 @end defspec | |
390 | |
391 @defspec defsubst* name arglist body... | |
392 This is just like @code{defun*}, except that the function that | |
393 is defined is automatically proclaimed @code{inline}, i.e., | |
394 calls to it may be expanded into in-line code by the byte compiler. | |
395 This is analogous to the @code{defsubst} form in Emacs 19; | |
396 @code{defsubst*} uses a different method (compiler macros) which | |
397 works in all version of Emacs, and also generates somewhat more | |
398 efficient inline expansions. In particular, @code{defsubst*} | |
399 arranges for the processing of keyword arguments, default values, | |
400 etc., to be done at compile-time whenever possible. | |
401 @end defspec | |
402 | |
403 @defspec defmacro* name arglist body... | |
404 This is identical to the regular @code{defmacro} form, | |
405 except that @var{arglist} is allowed to be a full Common Lisp | |
406 argument list. The @code{&environment} keyword is supported as | |
407 described in Steele. The @code{&whole} keyword is supported only | |
408 within destructured lists (see below); top-level @code{&whole} | |
409 cannot be implemented with the current Emacs Lisp interpreter. | |
410 The macro expander body is enclosed in an implicit block called | |
411 @var{name}. | |
412 @end defspec | |
413 | |
414 @defspec function* symbol-or-lambda | |
415 This is identical to the regular @code{function} form, | |
416 except that if the argument is a @code{lambda} form then that | |
417 form may use a full Common Lisp argument list. | |
418 @end defspec | |
419 | |
420 Also, all forms (such as @code{defsetf} and @code{flet}) defined | |
421 in this package that include @var{arglist}s in their syntax allow | |
422 full Common Lisp argument lists. | |
423 | |
424 Note that it is @emph{not} necessary to use @code{defun*} in | |
425 order to have access to most @dfn{CL} features in your function. | |
426 These features are always present; @code{defun*}'s only | |
427 difference from @code{defun} is its more flexible argument | |
428 lists and its implicit block. | |
429 | |
430 The full form of a Common Lisp argument list is | |
431 | |
432 @example | |
433 (@var{var}... | |
434 &optional (@var{var} @var{initform} @var{svar})... | |
435 &rest @var{var} | |
436 &key ((@var{keyword} @var{var}) @var{initform} @var{svar})... | |
437 &aux (@var{var} @var{initform})...) | |
438 @end example | |
439 | |
440 Each of the five argument list sections is optional. The @var{svar}, | |
441 @var{initform}, and @var{keyword} parts are optional; if they are | |
442 omitted, then @samp{(@var{var})} may be written simply @samp{@var{var}}. | |
443 | |
444 The first section consists of zero or more @dfn{required} arguments. | |
445 These arguments must always be specified in a call to the function; | |
446 there is no difference between Emacs Lisp and Common Lisp as far as | |
447 required arguments are concerned. | |
448 | |
449 The second section consists of @dfn{optional} arguments. These | |
450 arguments may be specified in the function call; if they are not, | |
451 @var{initform} specifies the default value used for the argument. | |
452 (No @var{initform} means to use @code{nil} as the default.) The | |
453 @var{initform} is evaluated with the bindings for the preceding | |
454 arguments already established; @code{(a &optional (b (1+ a)))} | |
455 matches one or two arguments, with the second argument defaulting | |
456 to one plus the first argument. If the @var{svar} is specified, | |
457 it is an auxiliary variable which is bound to @code{t} if the optional | |
458 argument was specified, or to @code{nil} if the argument was omitted. | |
459 If you don't use an @var{svar}, then there will be no way for your | |
460 function to tell whether it was called with no argument, or with | |
461 the default value passed explicitly as an argument. | |
462 | |
463 The third section consists of a single @dfn{rest} argument. If | |
464 more arguments were passed to the function than are accounted for | |
465 by the required and optional arguments, those extra arguments are | |
466 collected into a list and bound to the ``rest'' argument variable. | |
467 Common Lisp's @code{&rest} is equivalent to that of Emacs Lisp. | |
468 Common Lisp accepts @code{&body} as a synonym for @code{&rest} in | |
469 macro contexts; this package accepts it all the time. | |
470 | |
471 The fourth section consists of @dfn{keyword} arguments. These | |
472 are optional arguments which are specified by name rather than | |
473 positionally in the argument list. For example, | |
474 | |
475 @example | |
476 (defun* foo (a &optional b &key c d (e 17))) | |
477 @end example | |
478 | |
479 @noindent | |
480 defines a function which may be called with one, two, or more | |
481 arguments. The first two arguments are bound to @code{a} and | |
482 @code{b} in the usual way. The remaining arguments must be | |
483 pairs of the form @code{:c}, @code{:d}, or @code{:e} followed | |
484 by the value to be bound to the corresponding argument variable. | |
485 (Symbols whose names begin with a colon are called @dfn{keywords}, | |
486 and they are self-quoting in the same way as @code{nil} and | |
487 @code{t}.) | |
488 | |
489 For example, the call @code{(foo 1 2 :d 3 :c 4)} sets the five | |
490 arguments to 1, 2, 4, 3, and 17, respectively. If the same keyword | |
491 appears more than once in the function call, the first occurrence | |
492 takes precedence over the later ones. Note that it is not possible | |
493 to specify keyword arguments without specifying the optional | |
494 argument @code{b} as well, since @code{(foo 1 :c 2)} would bind | |
495 @code{b} to the keyword @code{:c}, then signal an error because | |
496 @code{2} is not a valid keyword. | |
497 | |
498 If a @var{keyword} symbol is explicitly specified in the argument | |
499 list as shown in the above diagram, then that keyword will be | |
500 used instead of just the variable name prefixed with a colon. | |
501 You can specify a @var{keyword} symbol which does not begin with | |
502 a colon at all, but such symbols will not be self-quoting; you | |
503 will have to quote them explicitly with an apostrophe in the | |
504 function call. | |
505 | |
506 Ordinarily it is an error to pass an unrecognized keyword to | |
507 a function, e.g., @code{(foo 1 2 :c 3 :goober 4)}. You can ask | |
508 Lisp to ignore unrecognized keywords, either by adding the | |
509 marker @code{&allow-other-keys} after the keyword section | |
510 of the argument list, or by specifying an @code{:allow-other-keys} | |
511 argument in the call whose value is non-@code{nil}. If the | |
512 function uses both @code{&rest} and @code{&key} at the same time, | |
513 the ``rest'' argument is bound to the keyword list as it appears | |
514 in the call. For example: | |
515 | |
516 @smallexample | |
517 (defun* find-thing (thing &rest rest &key need &allow-other-keys) | |
518 (or (apply 'member* thing thing-list :allow-other-keys t rest) | |
519 (if need (error "Thing not found")))) | |
520 @end smallexample | |
521 | |
522 @noindent | |
523 This function takes a @code{:need} keyword argument, but also | |
524 accepts other keyword arguments which are passed on to the | |
525 @code{member*} function. @code{allow-other-keys} is used to | |
526 keep both @code{find-thing} and @code{member*} from complaining | |
527 about each others' keywords in the arguments. | |
528 | |
529 As a (significant) performance optimization, this package | |
530 implements the scan for keyword arguments by calling @code{memq} | |
531 to search for keywords in a ``rest'' argument. Technically | |
532 speaking, this is incorrect, since @code{memq} looks at the | |
533 odd-numbered values as well as the even-numbered keywords. | |
534 The net effect is that if you happen to pass a keyword symbol | |
535 as the @emph{value} of another keyword argument, where that | |
536 keyword symbol happens to equal the name of a valid keyword | |
537 argument of the same function, then the keyword parser will | |
538 become confused. This minor bug can only affect you if you | |
539 use keyword symbols as general-purpose data in your program; | |
540 this practice is strongly discouraged in Emacs Lisp. | |
541 | |
542 The fifth section of the argument list consists of @dfn{auxiliary | |
543 variables}. These are not really arguments at all, but simply | |
544 variables which are bound to @code{nil} or to the specified | |
545 @var{initforms} during execution of the function. There is no | |
546 difference between the following two functions, except for a | |
547 matter of stylistic taste: | |
548 | |
549 @example | |
550 (defun* foo (a b &aux (c (+ a b)) d) | |
551 @var{body}) | |
552 | |
553 (defun* foo (a b) | |
554 (let ((c (+ a b)) d) | |
555 @var{body})) | |
556 @end example | |
557 | |
558 Argument lists support @dfn{destructuring}. In Common Lisp, | |
559 destructuring is only allowed with @code{defmacro}; this package | |
560 allows it with @code{defun*} and other argument lists as well. | |
561 In destructuring, any argument variable (@var{var} in the above | |
562 diagram) can be replaced by a list of variables, or more generally, | |
563 a recursive argument list. The corresponding argument value must | |
564 be a list whose elements match this recursive argument list. | |
565 For example: | |
566 | |
567 @example | |
568 (defmacro* dolist ((var listform &optional resultform) | |
569 &rest body) | |
570 ...) | |
571 @end example | |
572 | |
573 This says that the first argument of @code{dolist} must be a list | |
574 of two or three items; if there are other arguments as well as this | |
575 list, they are stored in @code{body}. All features allowed in | |
576 regular argument lists are allowed in these recursive argument lists. | |
577 In addition, the clause @samp{&whole @var{var}} is allowed at the | |
578 front of a recursive argument list. It binds @var{var} to the | |
579 whole list being matched; thus @code{(&whole all a b)} matches | |
580 a list of two things, with @code{a} bound to the first thing, | |
581 @code{b} bound to the second thing, and @code{all} bound to the | |
582 list itself. (Common Lisp allows @code{&whole} in top-level | |
583 @code{defmacro} argument lists as well, but Emacs Lisp does not | |
584 support this usage.) | |
585 | |
586 One last feature of destructuring is that the argument list may be | |
587 dotted, so that the argument list @code{(a b . c)} is functionally | |
588 equivalent to @code{(a b &rest c)}. | |
589 | |
590 If the optimization quality @code{safety} is set to 0 | |
591 (@pxref{Declarations}), error checking for wrong number of | |
592 arguments and invalid keyword arguments is disabled. By default, | |
593 argument lists are rigorously checked. | |
594 | |
595 @node Time of Evaluation, Function Aliases, Argument Lists, Program Structure | |
596 @section Time of Evaluation | |
597 | |
598 @noindent | |
599 Normally, the byte-compiler does not actually execute the forms in | |
600 a file it compiles. For example, if a file contains @code{(setq foo t)}, | |
601 the act of compiling it will not actually set @code{foo} to @code{t}. | |
602 This is true even if the @code{setq} was a top-level form (i.e., not | |
603 enclosed in a @code{defun} or other form). Sometimes, though, you | |
604 would like to have certain top-level forms evaluated at compile-time. | |
605 For example, the compiler effectively evaluates @code{defmacro} forms | |
606 at compile-time so that later parts of the file can refer to the | |
607 macros that are defined. | |
608 | |
609 @defspec eval-when (situations...) forms... | |
610 This form controls when the body @var{forms} are evaluated. | |
611 The @var{situations} list may contain any set of the symbols | |
612 @code{compile}, @code{load}, and @code{eval} (or their long-winded | |
613 ANSI equivalents, @code{:compile-toplevel}, @code{:load-toplevel}, | |
614 and @code{:execute}). | |
615 | |
616 The @code{eval-when} form is handled differently depending on | |
617 whether or not it is being compiled as a top-level form. | |
618 Specifically, it gets special treatment if it is being compiled | |
619 by a command such as @code{byte-compile-file} which compiles files | |
620 or buffers of code, and it appears either literally at the | |
621 top level of the file or inside a top-level @code{progn}. | |
622 | |
623 For compiled top-level @code{eval-when}s, the body @var{forms} are | |
624 executed at compile-time if @code{compile} is in the @var{situations} | |
625 list, and the @var{forms} are written out to the file (to be executed | |
626 at load-time) if @code{load} is in the @var{situations} list. | |
627 | |
628 For non-compiled-top-level forms, only the @code{eval} situation is | |
629 relevant. (This includes forms executed by the interpreter, forms | |
630 compiled with @code{byte-compile} rather than @code{byte-compile-file}, | |
631 and non-top-level forms.) The @code{eval-when} acts like a | |
632 @code{progn} if @code{eval} is specified, and like @code{nil} | |
633 (ignoring the body @var{forms}) if not. | |
634 | |
635 The rules become more subtle when @code{eval-when}s are nested; | |
636 consult Steele (second edition) for the gruesome details (and | |
637 some gruesome examples). | |
638 | |
639 Some simple examples: | |
640 | |
641 @example | |
642 ;; Top-level forms in foo.el: | |
643 (eval-when (compile) (setq foo1 'bar)) | |
644 (eval-when (load) (setq foo2 'bar)) | |
645 (eval-when (compile load) (setq foo3 'bar)) | |
646 (eval-when (eval) (setq foo4 'bar)) | |
647 (eval-when (eval compile) (setq foo5 'bar)) | |
648 (eval-when (eval load) (setq foo6 'bar)) | |
649 (eval-when (eval compile load) (setq foo7 'bar)) | |
650 @end example | |
651 | |
652 When @file{foo.el} is compiled, these variables will be set during | |
653 the compilation itself: | |
654 | |
655 @example | |
656 foo1 foo3 foo5 foo7 ; `compile' | |
657 @end example | |
658 | |
659 When @file{foo.elc} is loaded, these variables will be set: | |
660 | |
661 @example | |
662 foo2 foo3 foo6 foo7 ; `load' | |
663 @end example | |
664 | |
665 And if @file{foo.el} is loaded uncompiled, these variables will | |
666 be set: | |
667 | |
668 @example | |
669 foo4 foo5 foo6 foo7 ; `eval' | |
670 @end example | |
671 | |
672 If these seven @code{eval-when}s had been, say, inside a @code{defun}, | |
673 then the first three would have been equivalent to @code{nil} and the | |
674 last four would have been equivalent to the corresponding @code{setq}s. | |
675 | |
676 Note that @code{(eval-when (load eval) @dots{})} is equivalent | |
677 to @code{(progn @dots{})} in all contexts. The compiler treats | |
678 certain top-level forms, like @code{defmacro} (sort-of) and | |
679 @code{require}, as if they were wrapped in @code{(eval-when | |
680 (compile load eval) @dots{})}. | |
681 @end defspec | |
682 | |
4905
755ae5b97edb
Change "special form" to "special operator" in our sources.
Aidan Kehoe <kehoea@parhasard.net>
parents:
4677
diff
changeset
|
683 Emacs 19 includes two special operators related to @code{eval-when}. |
428 | 684 One of these, @code{eval-when-compile}, is not quite equivalent to |
685 any @code{eval-when} construct and is described below. This package | |
686 defines a version of @code{eval-when-compile} for the benefit of | |
687 Emacs 18 users. | |
688 | |
689 The other form, @code{(eval-and-compile @dots{})}, is exactly | |
690 equivalent to @samp{(eval-when (compile load eval) @dots{})} and | |
691 so is not itself defined by this package. | |
692 | |
693 @defspec eval-when-compile forms... | |
694 The @var{forms} are evaluated at compile-time; at execution time, | |
695 this form acts like a quoted constant of the resulting value. Used | |
696 at top-level, @code{eval-when-compile} is just like @samp{eval-when | |
697 (compile eval)}. In other contexts, @code{eval-when-compile} | |
698 allows code to be evaluated once at compile-time for efficiency | |
699 or other reasons. | |
700 | |
701 This form is similar to the @samp{#.} syntax of true Common Lisp. | |
702 @end defspec | |
703 | |
704 @defspec load-time-value form | |
705 The @var{form} is evaluated at load-time; at execution time, | |
706 this form acts like a quoted constant of the resulting value. | |
707 | |
708 Early Common Lisp had a @samp{#,} syntax that was similar to | |
709 this, but ANSI Common Lisp replaced it with @code{load-time-value} | |
710 and gave it more well-defined semantics. | |
711 | |
712 In a compiled file, @code{load-time-value} arranges for @var{form} | |
713 to be evaluated when the @file{.elc} file is loaded and then used | |
714 as if it were a quoted constant. In code compiled by | |
715 @code{byte-compile} rather than @code{byte-compile-file}, the | |
716 effect is identical to @code{eval-when-compile}. In uncompiled | |
717 code, both @code{eval-when-compile} and @code{load-time-value} | |
718 act exactly like @code{progn}. | |
719 | |
720 @example | |
721 (defun report () | |
722 (insert "This function was executed on: " | |
723 (current-time-string) | |
724 ", compiled on: " | |
725 (eval-when-compile (current-time-string)) | |
726 ;; or '#.(current-time-string) in real Common Lisp | |
727 ", and loaded on: " | |
728 (load-time-value (current-time-string)))) | |
729 @end example | |
730 | |
731 @noindent | |
732 Byte-compiled, the above defun will result in the following code | |
733 (or its compiled equivalent, of course) in the @file{.elc} file: | |
734 | |
735 @example | |
736 (setq --temp-- (current-time-string)) | |
737 (defun report () | |
738 (insert "This function was executed on: " | |
739 (current-time-string) | |
740 ", compiled on: " | |
741 '"Wed Jun 23 18:33:43 1993" | |
742 ", and loaded on: " | |
743 --temp--)) | |
744 @end example | |
745 @end defspec | |
746 | |
747 @node Function Aliases, , Time of Evaluation, Program Structure | |
748 @section Function Aliases | |
749 | |
750 @noindent | |
751 This section describes a feature from GNU Emacs 19 which this | |
752 package makes available in other versions of Emacs. | |
753 | |
754 @defun defalias symbol function | |
755 This function sets @var{symbol}'s function cell to @var{function}. | |
756 It is equivalent to @code{fset}, except that in GNU Emacs 19 it also | |
757 records the setting in @code{load-history} so that it can be undone | |
758 by a later @code{unload-feature}. | |
759 | |
760 In other versions of Emacs, @code{defalias} is a synonym for | |
761 @code{fset}. | |
762 @end defun | |
763 | |
764 @node Predicates, Control Structure, Program Structure, Top | |
765 @chapter Predicates | |
766 | |
767 @noindent | |
768 This section describes functions for testing whether various | |
769 facts are true or false. | |
770 | |
771 @menu | |
772 * Type Predicates:: `typep', `deftype', and `coerce' | |
773 * Equality Predicates:: `eql' and `equalp' | |
774 @end menu | |
775 | |
776 @node Type Predicates, Equality Predicates, Predicates, Predicates | |
777 @section Type Predicates | |
778 | |
779 @noindent | |
780 The @dfn{CL} package defines a version of the Common Lisp @code{typep} | |
781 predicate. | |
782 | |
783 @defun typep object type | |
784 Check if @var{object} is of type @var{type}, where @var{type} is a | |
785 (quoted) type name of the sort used by Common Lisp. For example, | |
786 @code{(typep foo 'integer)} is equivalent to @code{(integerp foo)}. | |
787 @end defun | |
788 | |
789 The @var{type} argument to the above function is either a symbol | |
790 or a list beginning with a symbol. | |
791 | |
792 @itemize @bullet | |
793 @item | |
794 If the type name is a symbol, Emacs appends @samp{-p} to the | |
795 symbol name to form the name of a predicate function for testing | |
796 the type. (Built-in predicates whose names end in @samp{p} rather | |
797 than @samp{-p} are used when appropriate.) | |
798 | |
799 @item | |
800 The type symbol @code{t} stands for the union of all types. | |
801 @code{(typep @var{object} t)} is always true. Likewise, the | |
802 type symbol @code{nil} stands for nothing at all, and | |
803 @code{(typep @var{object} nil)} is always false. | |
804 | |
805 @item | |
806 The type symbol @code{null} represents the symbol @code{nil}. | |
807 Thus @code{(typep @var{object} 'null)} is equivalent to | |
808 @code{(null @var{object})}. | |
809 | |
810 @item | |
811 The type symbol @code{real} is a synonym for @code{number}, and | |
812 @code{fixnum} is a synonym for @code{integer}. | |
813 | |
814 @item | |
815 The type symbols @code{character} and @code{string-char} match | |
816 characters. In Emacs-19 and XEmacs-19, characters are the same thing as | |
817 integers in the range 0-255. In XEmacs-20, where characters are a | |
818 first-class data type, this checks for actual characters, and | |
819 @code{(typep @var{8bit-integer} 'character)} will return @code{nil}. | |
820 | |
821 @item | |
822 The type symbol @code{float} uses the @code{floatp-safe} predicate | |
823 defined by this package rather than @code{floatp}, so it will work | |
824 correctly even in Emacs versions without floating-point support. | |
825 | |
826 @item | |
827 The type list @code{(integer @var{low} @var{high})} represents all | |
828 integers between @var{low} and @var{high}, inclusive. Either bound | |
829 may be a list of a single integer to specify an exclusive limit, | |
830 or a @code{*} to specify no limit. The type @code{(integer * *)} | |
831 is thus equivalent to @code{integer}. | |
832 | |
833 @item | |
834 Likewise, lists beginning with @code{float}, @code{real}, or | |
835 @code{number} represent numbers of that type falling in a particular | |
836 range. | |
837 | |
838 @item | |
839 Lists beginning with @code{and}, @code{or}, and @code{not} form | |
840 combinations of types. For example, @code{(or integer (float 0 *))} | |
841 represents all objects that are integers or non-negative floats. | |
842 | |
843 @item | |
844 Lists beginning with @code{member} or @code{member*} represent | |
845 objects @code{eql} to any of the following values. For example, | |
846 @code{(member 1 2 3 4)} is equivalent to @code{(integer 1 4)}, | |
847 and @code{(member nil)} is equivalent to @code{null}. | |
848 | |
849 @item | |
850 Lists of the form @code{(satisfies @var{predicate})} represent | |
851 all objects for which @var{predicate} returns true when called | |
852 with that object as an argument. | |
853 @end itemize | |
854 | |
855 The following function and macro (not technically predicates) are | |
856 related to @code{typep}. | |
857 | |
858 @defun coerce object type | |
859 This function attempts to convert @var{object} to the specified | |
860 @var{type}. If @var{object} is already of that type as determined by | |
861 @code{typep}, it is simply returned. Otherwise, certain types of | |
862 conversions will be made: If @var{type} is any sequence type | |
863 (@code{string}, @code{list}, etc.) then @var{object} will be | |
864 converted to that type if possible. If @var{type} is | |
865 @code{character}, then strings of length one and symbols with | |
866 one-character names can be coerced. If @var{type} is @code{float}, | |
867 then integers can be coerced in versions of Emacs that support | |
868 floats. In all other circumstances, @code{coerce} signals an | |
869 error. | |
870 @end defun | |
871 | |
872 @defspec deftype name arglist forms... | |
873 This macro defines a new type called @var{name}. It is similar | |
874 to @code{defmacro} in many ways; when @var{name} is encountered | |
875 as a type name, the body @var{forms} are evaluated and should | |
876 return a type specifier that is equivalent to the type. The | |
877 @var{arglist} is a Common Lisp argument list of the sort accepted | |
878 by @code{defmacro*}. The type specifier @samp{(@var{name} @var{args}...)} | |
879 is expanded by calling the expander with those arguments; the type | |
880 symbol @samp{@var{name}} is expanded by calling the expander with | |
881 no arguments. The @var{arglist} is processed the same as for | |
882 @code{defmacro*} except that optional arguments without explicit | |
883 defaults use @code{*} instead of @code{nil} as the ``default'' | |
884 default. Some examples: | |
885 | |
886 @example | |
887 (deftype null () '(satisfies null)) ; predefined | |
888 (deftype list () '(or null cons)) ; predefined | |
889 (deftype unsigned-byte (&optional bits) | |
890 (list 'integer 0 (if (eq bits '*) bits (1- (lsh 1 bits))))) | |
891 (unsigned-byte 8) @equiv{} (integer 0 255) | |
892 (unsigned-byte) @equiv{} (integer 0 *) | |
893 unsigned-byte @equiv{} (integer 0 *) | |
894 @end example | |
895 | |
896 @noindent | |
897 The last example shows how the Common Lisp @code{unsigned-byte} | |
898 type specifier could be implemented if desired; this package does | |
899 not implement @code{unsigned-byte} by default. | |
900 @end defspec | |
901 | |
902 The @code{typecase} and @code{check-type} macros also use type | |
903 names. @xref{Conditionals}. @xref{Assertions}. The @code{map}, | |
904 @code{concatenate}, and @code{merge} functions take type-name | |
905 arguments to specify the type of sequence to return. @xref{Sequences}. | |
906 | |
907 @node Equality Predicates, , Type Predicates, Predicates | |
908 @section Equality Predicates | |
909 | |
910 @noindent | |
911 This package defines two Common Lisp predicates, @code{eql} and | |
912 @code{equalp}. | |
913 | |
914 @defun eql a b | |
915 This function is almost the same as @code{eq}, except that if @var{a} | |
916 and @var{b} are numbers of the same type, it compares them for numeric | |
917 equality (as if by @code{equal} instead of @code{eq}). This makes a | |
918 difference only for versions of Emacs that are compiled with | |
919 floating-point support, such as Emacs 19. Emacs floats are allocated | |
920 objects just like cons cells, which means that @code{(eq 3.0 3.0)} | |
921 will not necessarily be true---if the two @code{3.0}s were allocated | |
922 separately, the pointers will be different even though the numbers are | |
923 the same. But @code{(eql 3.0 3.0)} will always be true. | |
924 | |
925 The types of the arguments must match, so @code{(eql 3 3.0)} is | |
926 still false. | |
927 | |
928 Note that Emacs integers are ``direct'' rather than allocated, which | |
929 basically means @code{(eq 3 3)} will always be true. Thus @code{eq} | |
930 and @code{eql} behave differently only if floating-point numbers are | |
931 involved, and are indistinguishable on Emacs versions that don't | |
932 support floats. | |
933 | |
934 There is a slight inconsistency with Common Lisp in the treatment of | |
935 positive and negative zeros. Some machines, notably those with IEEE | |
936 standard arithmetic, represent @code{+0} and @code{-0} as distinct | |
937 values. Normally this doesn't matter because the standard specifies | |
938 that @code{(= 0.0 -0.0)} should always be true, and this is indeed | |
939 what Emacs Lisp and Common Lisp do. But the Common Lisp standard | |
940 states that @code{(eql 0.0 -0.0)} and @code{(equal 0.0 -0.0)} should | |
941 be false on IEEE-like machines; Emacs Lisp does not do this, and in | |
942 fact the only known way to distinguish between the two zeros in Emacs | |
943 Lisp is to @code{format} them and check for a minus sign. | |
944 @end defun | |
945 | |
946 @defun equalp a b | |
947 This function is a more flexible version of @code{equal}. In | |
948 particular, it compares strings and characters case-insensitively, and | |
949 it compares numbers without regard to type (so that @code{(equalp 3 | |
950 3.0)} is true). Vectors and conses are compared recursively. All other | |
951 objects are compared as if by @code{equal}. | |
952 | |
953 This function differs from Common Lisp @code{equalp} in several | |
954 respects. In keeping with the idea that strings are less | |
955 vector-like in Emacs Lisp, this package's @code{equalp} also will not | |
956 compare strings against vectors of integers. | |
957 @end defun | |
958 | |
959 Also note that the Common Lisp functions @code{member} and @code{assoc} | |
960 use @code{eql} to compare elements, whereas Emacs Lisp follows the | |
961 MacLisp tradition and uses @code{equal} for these two functions. | |
962 In Emacs, use @code{member*} and @code{assoc*} to get functions | |
963 which use @code{eql} for comparisons. | |
964 | |
965 @node Control Structure, Macros, Predicates, Top | |
966 @chapter Control Structure | |
967 | |
968 @noindent | |
969 The features described in the following sections implement | |
970 various advanced control structures, including the powerful | |
971 @code{setf} facility and a number of looping and conditional | |
972 constructs. | |
973 | |
974 @menu | |
975 * Assignment:: The `psetq' form | |
976 * Generalized Variables:: `setf', `incf', `push', etc. | |
977 * Variable Bindings:: `progv', `lexical-let', `flet', `macrolet' | |
978 * Conditionals:: `when', `unless', `case', `typecase' | |
979 * Blocks and Exits:: `block', `return', `return-from' | |
980 * Iteration:: `do', `dotimes', `dolist', `do-symbols' | |
981 * Loop Facility:: The Common Lisp `loop' macro | |
982 * Multiple Values:: `values', `multiple-value-bind', etc. | |
983 @end menu | |
984 | |
985 @node Assignment, Generalized Variables, Control Structure, Control Structure | |
986 @section Assignment | |
987 | |
988 @noindent | |
989 The @code{psetq} form is just like @code{setq}, except that multiple | |
990 assignments are done in parallel rather than sequentially. | |
991 | |
992 @defspec psetq [symbol form]@dots{} | |
4905
755ae5b97edb
Change "special form" to "special operator" in our sources.
Aidan Kehoe <kehoea@parhasard.net>
parents:
4677
diff
changeset
|
993 This macro is used to assign to several |
428 | 994 variables simultaneously. Given only one @var{symbol} and @var{form}, |
995 it has the same effect as @code{setq}. Given several @var{symbol} | |
996 and @var{form} pairs, it evaluates all the @var{form}s in advance | |
997 and then stores the corresponding variables afterwards. | |
998 | |
999 @example | |
1000 (setq x 2 y 3) | |
1001 (setq x (+ x y) y (* x y)) | |
1002 x | |
1003 @result{} 5 | |
1004 y ; @r{@code{y} was computed after @code{x} was set.} | |
1005 @result{} 15 | |
1006 (setq x 2 y 3) | |
1007 (psetq x (+ x y) y (* x y)) | |
1008 x | |
1009 @result{} 5 | |
1010 y ; @r{@code{y} was computed before @code{x} was set.} | |
1011 @result{} 6 | |
1012 @end example | |
1013 | |
1014 The simplest use of @code{psetq} is @code{(psetq x y y x)}, which | |
1015 exchanges the values of two variables. (The @code{rotatef} form | |
1016 provides an even more convenient way to swap two variables; | |
1017 @pxref{Modify Macros}.) | |
1018 | |
1019 @code{psetq} always returns @code{nil}. | |
1020 @end defspec | |
1021 | |
1022 @node Generalized Variables, Variable Bindings, Assignment, Control Structure | |
1023 @section Generalized Variables | |
1024 | |
1025 @noindent | |
1026 A ``generalized variable'' or ``place form'' is one of the many places | |
1027 in Lisp memory where values can be stored. The simplest place form is | |
1028 a regular Lisp variable. But the cars and cdrs of lists, elements | |
1029 of arrays, properties of symbols, and many other locations are also | |
1030 places where Lisp values are stored. | |
1031 | |
1032 The @code{setf} form is like @code{setq}, except that it accepts | |
1033 arbitrary place forms on the left side rather than just | |
1034 symbols. For example, @code{(setf (car a) b)} sets the car of | |
1035 @code{a} to @code{b}, doing the same operation as @code{(setcar a b)} | |
1036 but without having to remember two separate functions for setting | |
1037 and accessing every type of place. | |
1038 | |
1039 Generalized variables are analogous to ``lvalues'' in the C | |
1040 language, where @samp{x = a[i]} gets an element from an array | |
1041 and @samp{a[i] = x} stores an element using the same notation. | |
1042 Just as certain forms like @code{a[i]} can be lvalues in C, there | |
1043 is a set of forms that can be generalized variables in Lisp. | |
1044 | |
1045 @menu | |
1046 * Basic Setf:: `setf' and place forms | |
1047 * Modify Macros:: `incf', `push', `rotatef', `letf', `callf', etc. | |
1048 * Customizing Setf:: `define-modify-macro', `defsetf', `define-setf-method' | |
1049 @end menu | |
1050 | |
1051 @node Basic Setf, Modify Macros, Generalized Variables, Generalized Variables | |
1052 @subsection Basic Setf | |
1053 | |
1054 @noindent | |
1055 The @code{setf} macro is the most basic way to operate on generalized | |
1056 variables. | |
1057 | |
1058 @defspec setf [place form]@dots{} | |
1059 This macro evaluates @var{form} and stores it in @var{place}, which | |
1060 must be a valid generalized variable form. If there are several | |
1061 @var{place} and @var{form} pairs, the assignments are done sequentially | |
1062 just as with @code{setq}. @code{setf} returns the value of the last | |
1063 @var{form}. | |
1064 | |
1065 The following Lisp forms will work as generalized variables, and | |
1066 so may legally appear in the @var{place} argument of @code{setf}: | |
1067 | |
1068 @itemize @bullet | |
1069 @item | |
1070 A symbol naming a variable. In other words, @code{(setf x y)} is | |
1071 exactly equivalent to @code{(setq x y)}, and @code{setq} itself is | |
1072 strictly speaking redundant now that @code{setf} exists. Many | |
1073 programmers continue to prefer @code{setq} for setting simple | |
1074 variables, though, purely for stylistic or historical reasons. | |
446 | 1075 The form @code{(setf x y)} actually expands to @code{(setq x y)}, |
428 | 1076 so there is no performance penalty for using it in compiled code. |
1077 | |
1078 @item | |
1079 A call to any of the following Lisp functions: | |
1080 | |
1081 @smallexample | |
1082 car cdr caar .. cddddr | |
1083 nth rest first .. tenth | |
1084 aref elt nthcdr | |
1085 symbol-function symbol-value symbol-plist | |
440 | 1086 get getf gethash |
1087 subseq | |
428 | 1088 @end smallexample |
1089 | |
1090 @noindent | |
1091 Note that for @code{nthcdr} and @code{getf}, the list argument | |
1092 of the function must itself be a valid @var{place} form. For | |
1093 example, @code{(setf (nthcdr 0 foo) 7)} will set @code{foo} itself | |
1094 to 7. Note that @code{push} and @code{pop} on an @code{nthcdr} | |
1095 place can be used to insert or delete at any position in a list. | |
1096 The use of @code{nthcdr} as a @var{place} form is an extension | |
1097 to standard Common Lisp. | |
1098 | |
1099 @item | |
1100 The following Emacs-specific functions are also @code{setf}-able. | |
1101 (Some of these are defined only in Emacs 19 or only in XEmacs.) | |
1102 | |
1103 @smallexample | |
1104 buffer-file-name marker-position | |
1105 buffer-modified-p match-data | |
1106 buffer-name mouse-position | |
1107 buffer-string overlay-end | |
1108 buffer-substring overlay-get | |
1109 current-buffer overlay-start | |
1110 current-case-table point | |
1111 current-column point-marker | |
1112 current-global-map point-max | |
1113 current-input-mode point-min | |
1114 current-local-map process-buffer | |
1115 current-window-configuration process-filter | |
1116 default-file-modes process-sentinel | |
1117 default-value read-mouse-position | |
1118 documentation-property screen-height | |
1119 extent-data screen-menubar | |
1120 extent-end-position screen-width | |
1121 extent-start-position selected-window | |
1122 face-background selected-screen | |
1123 face-background-pixmap selected-frame | |
1124 face-font standard-case-table | |
1125 face-foreground syntax-table | |
1126 face-underline-p window-buffer | |
1127 file-modes window-dedicated-p | |
1128 frame-height window-display-table | |
1129 frame-parameters window-height | |
1130 frame-visible-p window-hscroll | |
1131 frame-width window-point | |
1132 get-register window-start | |
1133 getenv window-width | |
1134 global-key-binding x-get-cut-buffer | |
1135 keymap-parent x-get-cutbuffer | |
1136 local-key-binding x-get-secondary-selection | |
1137 mark x-get-selection | |
1138 mark-marker | |
1139 @end smallexample | |
1140 | |
1141 Most of these have directly corresponding ``set'' functions, like | |
1142 @code{use-local-map} for @code{current-local-map}, or @code{goto-char} | |
1143 for @code{point}. A few, like @code{point-min}, expand to longer | |
1144 sequences of code when they are @code{setf}'d (@code{(narrow-to-region | |
1145 x (point-max))} in this case). | |
1146 | |
1147 @item | |
1148 A call of the form @code{(substring @var{subplace} @var{n} [@var{m}])}, | |
1149 where @var{subplace} is itself a legal generalized variable whose | |
1150 current value is a string, and where the value stored is also a | |
1151 string. The new string is spliced into the specified part of the | |
1152 destination string. For example: | |
1153 | |
1154 @example | |
1155 (setq a (list "hello" "world")) | |
1156 @result{} ("hello" "world") | |
1157 (cadr a) | |
1158 @result{} "world" | |
1159 (substring (cadr a) 2 4) | |
1160 @result{} "rl" | |
1161 (setf (substring (cadr a) 2 4) "o") | |
1162 @result{} "o" | |
1163 (cadr a) | |
1164 @result{} "wood" | |
1165 a | |
1166 @result{} ("hello" "wood") | |
1167 @end example | |
1168 | |
1169 The generalized variable @code{buffer-substring}, listed above, | |
1170 also works in this way by replacing a portion of the current buffer. | |
1171 | |
1172 @item | |
1173 A call of the form @code{(apply '@var{func} @dots{})} or | |
1174 @code{(apply (function @var{func}) @dots{})}, where @var{func} | |
1175 is a @code{setf}-able function whose store function is ``suitable'' | |
1176 in the sense described in Steele's book; since none of the standard | |
1177 Emacs place functions are suitable in this sense, this feature is | |
1178 only interesting when used with places you define yourself with | |
1179 @code{define-setf-method} or the long form of @code{defsetf}. | |
1180 | |
1181 @item | |
1182 A macro call, in which case the macro is expanded and @code{setf} | |
1183 is applied to the resulting form. | |
1184 | |
1185 @item | |
1186 Any form for which a @code{defsetf} or @code{define-setf-method} | |
1187 has been made. | |
1188 @end itemize | |
1189 | |
1190 Using any forms other than these in the @var{place} argument to | |
1191 @code{setf} will signal an error. | |
1192 | |
1193 The @code{setf} macro takes care to evaluate all subforms in | |
1194 the proper left-to-right order; for example, | |
1195 | |
1196 @example | |
1197 (setf (aref vec (incf i)) i) | |
1198 @end example | |
1199 | |
1200 @noindent | |
1201 looks like it will evaluate @code{(incf i)} exactly once, before the | |
1202 following access to @code{i}; the @code{setf} expander will insert | |
1203 temporary variables as necessary to ensure that it does in fact work | |
1204 this way no matter what setf-method is defined for @code{aref}. | |
1205 (In this case, @code{aset} would be used and no such steps would | |
1206 be necessary since @code{aset} takes its arguments in a convenient | |
1207 order.) | |
1208 | |
1209 However, if the @var{place} form is a macro which explicitly | |
1210 evaluates its arguments in an unusual order, this unusual order | |
1211 will be preserved. Adapting an example from Steele, given | |
1212 | |
1213 @example | |
1214 (defmacro wrong-order (x y) (list 'aref y x)) | |
1215 @end example | |
1216 | |
1217 @noindent | |
1218 the form @code{(setf (wrong-order @var{a} @var{b}) 17)} will | |
1219 evaluate @var{b} first, then @var{a}, just as in an actual call | |
1220 to @code{wrong-order}. | |
1221 @end defspec | |
1222 | |
1223 @node Modify Macros, Customizing Setf, Basic Setf, Generalized Variables | |
1224 @subsection Modify Macros | |
1225 | |
1226 @noindent | |
1227 This package defines a number of other macros besides @code{setf} | |
1228 that operate on generalized variables. Many are interesting and | |
1229 useful even when the @var{place} is just a variable name. | |
1230 | |
1231 @defspec psetf [place form]@dots{} | |
1232 This macro is to @code{setf} what @code{psetq} is to @code{setq}: | |
1233 When several @var{place}s and @var{form}s are involved, the | |
1234 assignments take place in parallel rather than sequentially. | |
1235 Specifically, all subforms are evaluated from left to right, then | |
1236 all the assignments are done (in an undefined order). | |
1237 @end defspec | |
1238 | |
1239 @defspec incf place &optional x | |
1240 This macro increments the number stored in @var{place} by one, or | |
1241 by @var{x} if specified. The incremented value is returned. For | |
1242 example, @code{(incf i)} is equivalent to @code{(setq i (1+ i))}, and | |
1243 @code{(incf (car x) 2)} is equivalent to @code{(setcar x (+ (car x) 2))}. | |
1244 | |
1245 Once again, care is taken to preserve the ``apparent'' order of | |
1246 evaluation. For example, | |
1247 | |
1248 @example | |
1249 (incf (aref vec (incf i))) | |
1250 @end example | |
1251 | |
1252 @noindent | |
1253 appears to increment @code{i} once, then increment the element of | |
1254 @code{vec} addressed by @code{i}; this is indeed exactly what it | |
1255 does, which means the above form is @emph{not} equivalent to the | |
1256 ``obvious'' expansion, | |
1257 | |
1258 @example | |
1259 (setf (aref vec (incf i)) (1+ (aref vec (incf i)))) ; Wrong! | |
1260 @end example | |
1261 | |
1262 @noindent | |
1263 but rather to something more like | |
1264 | |
1265 @example | |
1266 (let ((temp (incf i))) | |
1267 (setf (aref vec temp) (1+ (aref vec temp)))) | |
1268 @end example | |
1269 | |
1270 @noindent | |
1271 Again, all of this is taken care of automatically by @code{incf} and | |
1272 the other generalized-variable macros. | |
1273 | |
1274 As a more Emacs-specific example of @code{incf}, the expression | |
1275 @code{(incf (point) @var{n})} is essentially equivalent to | |
1276 @code{(forward-char @var{n})}. | |
1277 @end defspec | |
1278 | |
1279 @defspec decf place &optional x | |
1280 This macro decrements the number stored in @var{place} by one, or | |
1281 by @var{x} if specified. | |
1282 @end defspec | |
1283 | |
1284 @defspec pop place | |
1285 This macro removes and returns the first element of the list stored | |
1286 in @var{place}. It is analogous to @code{(prog1 (car @var{place}) | |
1287 (setf @var{place} (cdr @var{place})))}, except that it takes care | |
1288 to evaluate all subforms only once. | |
1289 @end defspec | |
1290 | |
1291 @defspec push x place | |
1292 This macro inserts @var{x} at the front of the list stored in | |
1293 @var{place}. It is analogous to @code{(setf @var{place} (cons | |
1294 @var{x} @var{place}))}, except for evaluation of the subforms. | |
1295 @end defspec | |
1296 | |
1297 @defspec pushnew x place @t{&key :test :test-not :key} | |
1298 This macro inserts @var{x} at the front of the list stored in | |
1299 @var{place}, but only if @var{x} was not @code{eql} to any | |
1300 existing element of the list. The optional keyword arguments | |
1301 are interpreted in the same way as for @code{adjoin}. | |
1302 @xref{Lists as Sets}. | |
1303 @end defspec | |
1304 | |
1305 @defspec shiftf place@dots{} newvalue | |
1306 This macro shifts the @var{place}s left by one, shifting in the | |
1307 value of @var{newvalue} (which may be any Lisp expression, not just | |
1308 a generalized variable), and returning the value shifted out of | |
1309 the first @var{place}. Thus, @code{(shiftf @var{a} @var{b} @var{c} | |
1310 @var{d})} is equivalent to | |
1311 | |
1312 @example | |
1313 (prog1 | |
1314 @var{a} | |
1315 (psetf @var{a} @var{b} | |
1316 @var{b} @var{c} | |
1317 @var{c} @var{d})) | |
1318 @end example | |
1319 | |
1320 @noindent | |
1321 except that the subforms of @var{a}, @var{b}, and @var{c} are actually | |
1322 evaluated only once each and in the apparent order. | |
1323 @end defspec | |
1324 | |
1325 @defspec rotatef place@dots{} | |
1326 This macro rotates the @var{place}s left by one in circular fashion. | |
1327 Thus, @code{(rotatef @var{a} @var{b} @var{c} @var{d})} is equivalent to | |
1328 | |
1329 @example | |
1330 (psetf @var{a} @var{b} | |
1331 @var{b} @var{c} | |
1332 @var{c} @var{d} | |
1333 @var{d} @var{a}) | |
1334 @end example | |
1335 | |
1336 @noindent | |
1337 except for the evaluation of subforms. @code{rotatef} always | |
1338 returns @code{nil}. Note that @code{(rotatef @var{a} @var{b})} | |
1339 conveniently exchanges @var{a} and @var{b}. | |
1340 @end defspec | |
1341 | |
1342 The following macros were invented for this package; they have no | |
1343 analogues in Common Lisp. | |
1344 | |
1345 @defspec letf (bindings@dots{}) forms@dots{} | |
1346 This macro is analogous to @code{let}, but for generalized variables | |
1347 rather than just symbols. Each @var{binding} should be of the form | |
1348 @code{(@var{place} @var{value})}; the original contents of the | |
1349 @var{place}s are saved, the @var{value}s are stored in them, and | |
1350 then the body @var{form}s are executed. Afterwards, the @var{places} | |
1351 are set back to their original saved contents. This cleanup happens | |
1352 even if the @var{form}s exit irregularly due to a @code{throw} or an | |
1353 error. | |
1354 | |
1355 For example, | |
1356 | |
1357 @example | |
1358 (letf (((point) (point-min)) | |
1359 (a 17)) | |
1360 ...) | |
1361 @end example | |
1362 | |
1363 @noindent | |
1364 moves ``point'' in the current buffer to the beginning of the buffer, | |
1365 and also binds @code{a} to 17 (as if by a normal @code{let}, since | |
1366 @code{a} is just a regular variable). After the body exits, @code{a} | |
1367 is set back to its original value and point is moved back to its | |
1368 original position. | |
1369 | |
1370 Note that @code{letf} on @code{(point)} is not quite like a | |
1371 @code{save-excursion}, as the latter effectively saves a marker | |
1372 which tracks insertions and deletions in the buffer. Actually, | |
1373 a @code{letf} of @code{(point-marker)} is much closer to this | |
1374 behavior. (@code{point} and @code{point-marker} are equivalent | |
1375 as @code{setf} places; each will accept either an integer or a | |
1376 marker as the stored value.) | |
1377 | |
1378 Since generalized variables look like lists, @code{let}'s shorthand | |
1379 of using @samp{foo} for @samp{(foo nil)} as a @var{binding} would | |
1380 be ambiguous in @code{letf} and is not allowed. | |
1381 | |
1382 However, a @var{binding} specifier may be a one-element list | |
1383 @samp{(@var{place})}, which is similar to @samp{(@var{place} | |
1384 @var{place})}. In other words, the @var{place} is not disturbed | |
1385 on entry to the body, and the only effect of the @code{letf} is | |
1386 to restore the original value of @var{place} afterwards. (The | |
1387 redundant access-and-store suggested by the @code{(@var{place} | |
1388 @var{place})} example does not actually occur.) | |
1389 | |
1390 In most cases, the @var{place} must have a well-defined value on | |
1391 entry to the @code{letf} form. The only exceptions are plain | |
1392 variables and calls to @code{symbol-value} and @code{symbol-function}. | |
1393 If the symbol is not bound on entry, it is simply made unbound by | |
1394 @code{makunbound} or @code{fmakunbound} on exit. | |
1395 @end defspec | |
1396 | |
1397 @defspec letf* (bindings@dots{}) forms@dots{} | |
1398 This macro is to @code{letf} what @code{let*} is to @code{let}: | |
1399 It does the bindings in sequential rather than parallel order. | |
1400 @end defspec | |
1401 | |
1402 @defspec callf @var{function} @var{place} @var{args}@dots{} | |
1403 This is the ``generic'' modify macro. It calls @var{function}, | |
1404 which should be an unquoted function name, macro name, or lambda. | |
1405 It passes @var{place} and @var{args} as arguments, and assigns the | |
1406 result back to @var{place}. For example, @code{(incf @var{place} | |
1407 @var{n})} is the same as @code{(callf + @var{place} @var{n})}. | |
1408 Some more examples: | |
1409 | |
1410 @example | |
1411 (callf abs my-number) | |
1412 (callf concat (buffer-name) "<" (int-to-string n) ">") | |
1413 (callf union happy-people (list joe bob) :test 'same-person) | |
1414 @end example | |
1415 | |
1416 @xref{Customizing Setf}, for @code{define-modify-macro}, a way | |
1417 to create even more concise notations for modify macros. Note | |
1418 again that @code{callf} is an extension to standard Common Lisp. | |
1419 @end defspec | |
1420 | |
1421 @defspec callf2 @var{function} @var{arg1} @var{place} @var{args}@dots{} | |
1422 This macro is like @code{callf}, except that @var{place} is | |
1423 the @emph{second} argument of @var{function} rather than the | |
1424 first. For example, @code{(push @var{x} @var{place})} is | |
1425 equivalent to @code{(callf2 cons @var{x} @var{place})}. | |
1426 @end defspec | |
1427 | |
1428 The @code{callf} and @code{callf2} macros serve as building | |
1429 blocks for other macros like @code{incf}, @code{pushnew}, and | |
1430 @code{define-modify-macro}. The @code{letf} and @code{letf*} | |
1431 macros are used in the processing of symbol macros; | |
1432 @pxref{Macro Bindings}. | |
1433 | |
1434 @node Customizing Setf, , Modify Macros, Generalized Variables | |
1435 @subsection Customizing Setf | |
1436 | |
1437 @noindent | |
1438 Common Lisp defines three macros, @code{define-modify-macro}, | |
1439 @code{defsetf}, and @code{define-setf-method}, that allow the | |
1440 user to extend generalized variables in various ways. | |
1441 | |
1442 @defspec define-modify-macro name arglist function [doc-string] | |
1443 This macro defines a ``read-modify-write'' macro similar to | |
1444 @code{incf} and @code{decf}. The macro @var{name} is defined | |
1445 to take a @var{place} argument followed by additional arguments | |
1446 described by @var{arglist}. The call | |
1447 | |
1448 @example | |
1449 (@var{name} @var{place} @var{args}...) | |
1450 @end example | |
1451 | |
1452 @noindent | |
1453 will be expanded to | |
1454 | |
1455 @example | |
1456 (callf @var{func} @var{place} @var{args}...) | |
1457 @end example | |
1458 | |
1459 @noindent | |
1460 which in turn is roughly equivalent to | |
1461 | |
1462 @example | |
1463 (setf @var{place} (@var{func} @var{place} @var{args}...)) | |
1464 @end example | |
1465 | |
1466 For example: | |
1467 | |
1468 @example | |
1469 (define-modify-macro incf (&optional (n 1)) +) | |
1470 (define-modify-macro concatf (&rest args) concat) | |
1471 @end example | |
1472 | |
1473 Note that @code{&key} is not allowed in @var{arglist}, but | |
1474 @code{&rest} is sufficient to pass keywords on to the function. | |
1475 | |
1476 Most of the modify macros defined by Common Lisp do not exactly | |
1477 follow the pattern of @code{define-modify-macro}. For example, | |
1478 @code{push} takes its arguments in the wrong order, and @code{pop} | |
1479 is completely irregular. You can define these macros ``by hand'' | |
1480 using @code{get-setf-method}, or consult the source file | |
1481 @file{cl-macs.el} to see how to use the internal @code{setf} | |
1482 building blocks. | |
1483 @end defspec | |
1484 | |
1485 @defspec defsetf access-fn update-fn | |
1486 This is the simpler of two @code{defsetf} forms. Where | |
1487 @var{access-fn} is the name of a function which accesses a place, | |
1488 this declares @var{update-fn} to be the corresponding store | |
1489 function. From now on, | |
1490 | |
1491 @example | |
1492 (setf (@var{access-fn} @var{arg1} @var{arg2} @var{arg3}) @var{value}) | |
1493 @end example | |
1494 | |
1495 @noindent | |
1496 will be expanded to | |
1497 | |
1498 @example | |
1499 (@var{update-fn} @var{arg1} @var{arg2} @var{arg3} @var{value}) | |
1500 @end example | |
1501 | |
1502 @noindent | |
1503 The @var{update-fn} is required to be either a true function, or | |
1504 a macro which evaluates its arguments in a function-like way. Also, | |
1505 the @var{update-fn} is expected to return @var{value} as its result. | |
1506 Otherwise, the above expansion would not obey the rules for the way | |
1507 @code{setf} is supposed to behave. | |
1508 | |
1509 As a special (non-Common-Lisp) extension, a third argument of @code{t} | |
1510 to @code{defsetf} says that the @code{update-fn}'s return value is | |
1511 not suitable, so that the above @code{setf} should be expanded to | |
1512 something more like | |
1513 | |
1514 @example | |
1515 (let ((temp @var{value})) | |
1516 (@var{update-fn} @var{arg1} @var{arg2} @var{arg3} temp) | |
1517 temp) | |
1518 @end example | |
1519 | |
1520 Some examples of the use of @code{defsetf}, drawn from the standard | |
1521 suite of setf methods, are: | |
1522 | |
1523 @example | |
1524 (defsetf car setcar) | |
1525 (defsetf symbol-value set) | |
1526 (defsetf buffer-name rename-buffer t) | |
1527 @end example | |
1528 @end defspec | |
1529 | |
1530 @defspec defsetf access-fn arglist (store-var) forms@dots{} | |
1531 This is the second, more complex, form of @code{defsetf}. It is | |
1532 rather like @code{defmacro} except for the additional @var{store-var} | |
1533 argument. The @var{forms} should return a Lisp form which stores | |
1534 the value of @var{store-var} into the generalized variable formed | |
1535 by a call to @var{access-fn} with arguments described by @var{arglist}. | |
1536 The @var{forms} may begin with a string which documents the @code{setf} | |
1537 method (analogous to the doc string that appears at the front of a | |
1538 function). | |
1539 | |
1540 For example, the simple form of @code{defsetf} is shorthand for | |
1541 | |
1542 @example | |
1543 (defsetf @var{access-fn} (&rest args) (store) | |
1544 (append '(@var{update-fn}) args (list store))) | |
1545 @end example | |
1546 | |
1547 The Lisp form that is returned can access the arguments from | |
1548 @var{arglist} and @var{store-var} in an unrestricted fashion; | |
1549 macros like @code{setf} and @code{incf} which invoke this | |
1550 setf-method will insert temporary variables as needed to make | |
1551 sure the apparent order of evaluation is preserved. | |
1552 | |
1553 Another example drawn from the standard package: | |
1554 | |
1555 @example | |
1556 (defsetf nth (n x) (store) | |
1557 (list 'setcar (list 'nthcdr n x) store)) | |
1558 @end example | |
1559 @end defspec | |
1560 | |
1561 @defspec define-setf-method access-fn arglist forms@dots{} | |
1562 This is the most general way to create new place forms. When | |
1563 a @code{setf} to @var{access-fn} with arguments described by | |
1564 @var{arglist} is expanded, the @var{forms} are evaluated and | |
1565 must return a list of five items: | |
1566 | |
1567 @enumerate | |
1568 @item | |
1569 A list of @dfn{temporary variables}. | |
1570 | |
1571 @item | |
1572 A list of @dfn{value forms} corresponding to the temporary variables | |
1573 above. The temporary variables will be bound to these value forms | |
1574 as the first step of any operation on the generalized variable. | |
1575 | |
1576 @item | |
1577 A list of exactly one @dfn{store variable} (generally obtained | |
1578 from a call to @code{gensym}). | |
1579 | |
1580 @item | |
1581 A Lisp form which stores the contents of the store variable into | |
1582 the generalized variable, assuming the temporaries have been | |
1583 bound as described above. | |
1584 | |
1585 @item | |
1586 A Lisp form which accesses the contents of the generalized variable, | |
1587 assuming the temporaries have been bound. | |
1588 @end enumerate | |
1589 | |
1590 This is exactly like the Common Lisp macro of the same name, | |
1591 except that the method returns a list of five values rather | |
1592 than the five values themselves, since Emacs Lisp does not | |
1593 support Common Lisp's notion of multiple return values. | |
1594 | |
1595 Once again, the @var{forms} may begin with a documentation string. | |
1596 | |
1597 A setf-method should be maximally conservative with regard to | |
1598 temporary variables. In the setf-methods generated by | |
1599 @code{defsetf}, the second return value is simply the list of | |
1600 arguments in the place form, and the first return value is a | |
1601 list of a corresponding number of temporary variables generated | |
1602 by @code{gensym}. Macros like @code{setf} and @code{incf} which | |
1603 use this setf-method will optimize away most temporaries that | |
1604 turn out to be unnecessary, so there is little reason for the | |
1605 setf-method itself to optimize. | |
1606 @end defspec | |
1607 | |
1608 @defun get-setf-method place &optional env | |
1609 This function returns the setf-method for @var{place}, by | |
1610 invoking the definition previously recorded by @code{defsetf} | |
1611 or @code{define-setf-method}. The result is a list of five | |
1612 values as described above. You can use this function to build | |
1613 your own @code{incf}-like modify macros. (Actually, it is | |
1614 better to use the internal functions @code{cl-setf-do-modify} | |
1615 and @code{cl-setf-do-store}, which are a bit easier to use and | |
1616 which also do a number of optimizations; consult the source | |
1617 code for the @code{incf} function for a simple example.) | |
1618 | |
1619 The argument @var{env} specifies the ``environment'' to be | |
1620 passed on to @code{macroexpand} if @code{get-setf-method} should | |
1621 need to expand a macro in @var{place}. It should come from | |
1622 an @code{&environment} argument to the macro or setf-method | |
1623 that called @code{get-setf-method}. | |
1624 | |
1625 See also the source code for the setf-methods for @code{apply} | |
1626 and @code{substring}, each of which works by calling | |
1627 @code{get-setf-method} on a simpler case, then massaging | |
1628 the result in various ways. | |
1629 @end defun | |
1630 | |
1631 Modern Common Lisp defines a second, independent way to specify | |
1632 the @code{setf} behavior of a function, namely ``@code{setf} | |
1633 functions'' whose names are lists @code{(setf @var{name})} | |
1634 rather than symbols. For example, @code{(defun (setf foo) @dots{})} | |
1635 defines the function that is used when @code{setf} is applied to | |
1636 @code{foo}. This package does not currently support @code{setf} | |
1637 functions. In particular, it is a compile-time error to use | |
1638 @code{setf} on a form which has not already been @code{defsetf}'d | |
1639 or otherwise declared; in newer Common Lisps, this would not be | |
1640 an error since the function @code{(setf @var{func})} might be | |
1641 defined later. | |
1642 | |
1643 @iftex | |
1644 @secno=4 | |
1645 @end iftex | |
1646 | |
1647 @node Variable Bindings, Conditionals, Generalized Variables, Control Structure | |
1648 @section Variable Bindings | |
1649 | |
1650 @noindent | |
1651 These Lisp forms make bindings to variables and function names, | |
1652 analogous to Lisp's built-in @code{let} form. | |
1653 | |
1654 @xref{Modify Macros}, for the @code{letf} and @code{letf*} forms which | |
1655 are also related to variable bindings. | |
1656 | |
1657 @menu | |
1658 * Dynamic Bindings:: The `progv' form | |
1659 * Lexical Bindings:: `lexical-let' and lexical closures | |
1660 * Function Bindings:: `flet' and `labels' | |
1661 * Macro Bindings:: `macrolet' and `symbol-macrolet' | |
1662 @end menu | |
1663 | |
1664 @node Dynamic Bindings, Lexical Bindings, Variable Bindings, Variable Bindings | |
1665 @subsection Dynamic Bindings | |
1666 | |
1667 @noindent | |
1668 The standard @code{let} form binds variables whose names are known | |
1669 at compile-time. The @code{progv} form provides an easy way to | |
1670 bind variables whose names are computed at run-time. | |
1671 | |
1672 @defspec progv symbols values forms@dots{} | |
1673 This form establishes @code{let}-style variable bindings on a | |
1674 set of variables computed at run-time. The expressions | |
1675 @var{symbols} and @var{values} are evaluated, and must return lists | |
1676 of symbols and values, respectively. The symbols are bound to the | |
1677 corresponding values for the duration of the body @var{form}s. | |
1678 If @var{values} is shorter than @var{symbols}, the last few symbols | |
1679 are made unbound (as if by @code{makunbound}) inside the body. | |
1680 If @var{symbols} is shorter than @var{values}, the excess values | |
1681 are ignored. | |
1682 @end defspec | |
1683 | |
1684 @node Lexical Bindings, Function Bindings, Dynamic Bindings, Variable Bindings | |
1685 @subsection Lexical Bindings | |
1686 | |
1687 @noindent | |
1688 The @dfn{CL} package defines the following macro which | |
1689 more closely follows the Common Lisp @code{let} form: | |
1690 | |
1691 @defspec lexical-let (bindings@dots{}) forms@dots{} | |
1692 This form is exactly like @code{let} except that the bindings it | |
1693 establishes are purely lexical. Lexical bindings are similar to | |
1694 local variables in a language like C: Only the code physically | |
1695 within the body of the @code{lexical-let} (after macro expansion) | |
1696 may refer to the bound variables. | |
1697 | |
1698 @example | |
1699 (setq a 5) | |
1700 (defun foo (b) (+ a b)) | |
1701 (let ((a 2)) (foo a)) | |
1702 @result{} 4 | |
1703 (lexical-let ((a 2)) (foo a)) | |
1704 @result{} 7 | |
1705 @end example | |
1706 | |
1707 @noindent | |
1708 In this example, a regular @code{let} binding of @code{a} actually | |
1709 makes a temporary change to the global variable @code{a}, so @code{foo} | |
1710 is able to see the binding of @code{a} to 2. But @code{lexical-let} | |
1711 actually creates a distinct local variable @code{a} for use within its | |
1712 body, without any effect on the global variable of the same name. | |
1713 | |
1714 The most important use of lexical bindings is to create @dfn{closures}. | |
1715 A closure is a function object that refers to an outside lexical | |
1716 variable. For example: | |
1717 | |
1718 @example | |
1719 (defun make-adder (n) | |
1720 (lexical-let ((n n)) | |
1721 (function (lambda (m) (+ n m))))) | |
1722 (setq add17 (make-adder 17)) | |
1723 (funcall add17 4) | |
1724 @result{} 21 | |
1725 @end example | |
1726 | |
1727 @noindent | |
1728 The call @code{(make-adder 17)} returns a function object which adds | |
1729 17 to its argument. If @code{let} had been used instead of | |
1730 @code{lexical-let}, the function object would have referred to the | |
1731 global @code{n}, which would have been bound to 17 only during the | |
1732 call to @code{make-adder} itself. | |
1733 | |
1734 @example | |
1735 (defun make-counter () | |
1736 (lexical-let ((n 0)) | |
1737 (function* (lambda (&optional (m 1)) (incf n m))))) | |
1738 (setq count-1 (make-counter)) | |
1739 (funcall count-1 3) | |
1740 @result{} 3 | |
1741 (funcall count-1 14) | |
1742 @result{} 17 | |
1743 (setq count-2 (make-counter)) | |
1744 (funcall count-2 5) | |
1745 @result{} 5 | |
1746 (funcall count-1 2) | |
1747 @result{} 19 | |
1748 (funcall count-2) | |
1749 @result{} 6 | |
1750 @end example | |
1751 | |
1752 @noindent | |
1753 Here we see that each call to @code{make-counter} creates a distinct | |
1754 local variable @code{n}, which serves as a private counter for the | |
1755 function object that is returned. | |
1756 | |
1757 Closed-over lexical variables persist until the last reference to | |
1758 them goes away, just like all other Lisp objects. For example, | |
1759 @code{count-2} refers to a function object which refers to an | |
1760 instance of the variable @code{n}; this is the only reference | |
1761 to that variable, so after @code{(setq count-2 nil)} the garbage | |
1762 collector would be able to delete this instance of @code{n}. | |
1763 Of course, if a @code{lexical-let} does not actually create any | |
1764 closures, then the lexical variables are free as soon as the | |
1765 @code{lexical-let} returns. | |
1766 | |
1767 Many closures are used only during the extent of the bindings they | |
1768 refer to; these are known as ``downward funargs'' in Lisp parlance. | |
1769 When a closure is used in this way, regular Emacs Lisp dynamic | |
1770 bindings suffice and will be more efficient than @code{lexical-let} | |
1771 closures: | |
1772 | |
1773 @example | |
1774 (defun add-to-list (x list) | |
1775 (mapcar (function (lambda (y) (+ x y))) list)) | |
1776 (add-to-list 7 '(1 2 5)) | |
1777 @result{} (8 9 12) | |
1778 @end example | |
1779 | |
1780 @noindent | |
1781 Since this lambda is only used while @code{x} is still bound, | |
1782 it is not necessary to make a true closure out of it. | |
1783 | |
1784 You can use @code{defun} or @code{flet} inside a @code{lexical-let} | |
1785 to create a named closure. If several closures are created in the | |
1786 body of a single @code{lexical-let}, they all close over the same | |
1787 instance of the lexical variable. | |
1788 | |
1789 The @code{lexical-let} form is an extension to Common Lisp. In | |
1790 true Common Lisp, all bindings are lexical unless declared otherwise. | |
1791 @end defspec | |
1792 | |
1793 @defspec lexical-let* (bindings@dots{}) forms@dots{} | |
1794 This form is just like @code{lexical-let}, except that the bindings | |
1795 are made sequentially in the manner of @code{let*}. | |
1796 @end defspec | |
1797 | |
1798 @node Function Bindings, Macro Bindings, Lexical Bindings, Variable Bindings | |
1799 @subsection Function Bindings | |
1800 | |
1801 @noindent | |
1802 These forms make @code{let}-like bindings to functions instead | |
1803 of variables. | |
1804 | |
1805 @defspec flet (bindings@dots{}) forms@dots{} | |
1806 This form establishes @code{let}-style bindings on the function | |
1807 cells of symbols rather than on the value cells. Each @var{binding} | |
1808 must be a list of the form @samp{(@var{name} @var{arglist} | |
1809 @var{forms}@dots{})}, which defines a function exactly as if | |
1810 it were a @code{defun*} form. The function @var{name} is defined | |
1811 accordingly for the duration of the body of the @code{flet}; then | |
1812 the old function definition, or lack thereof, is restored. | |
1813 | |
1814 While @code{flet} in Common Lisp establishes a lexical binding of | |
1815 @var{name}, Emacs Lisp @code{flet} makes a dynamic binding. The | |
1816 result is that @code{flet} affects indirect calls to a function as | |
1817 well as calls directly inside the @code{flet} form itself. | |
1818 | |
1819 You can use @code{flet} to disable or modify the behavior of a | |
1820 function in a temporary fashion. This will even work on Emacs | |
1821 primitives, although note that some calls to primitive functions | |
1822 internal to Emacs are made without going through the symbol's | |
1823 function cell, and so will not be affected by @code{flet}. For | |
1824 example, | |
1825 | |
1826 @example | |
1827 (flet ((message (&rest args) (push args saved-msgs))) | |
1828 (do-something)) | |
1829 @end example | |
1830 | |
1831 This code attempts to replace the built-in function @code{message} | |
1832 with a function that simply saves the messages in a list rather | |
1833 than displaying them. The original definition of @code{message} | |
1834 will be restored after @code{do-something} exits. This code will | |
1835 work fine on messages generated by other Lisp code, but messages | |
1836 generated directly inside Emacs will not be caught since they make | |
1837 direct C-language calls to the message routines rather than going | |
1838 through the Lisp @code{message} function. | |
1839 | |
1840 Functions defined by @code{flet} may use the full Common Lisp | |
1841 argument notation supported by @code{defun*}; also, the function | |
1842 body is enclosed in an implicit block as if by @code{defun*}. | |
1843 @xref{Program Structure}. | |
1844 @end defspec | |
1845 | |
1846 @defspec labels (bindings@dots{}) forms@dots{} | |
1847 The @code{labels} form is a synonym for @code{flet}. (In Common | |
1848 Lisp, @code{labels} and @code{flet} differ in ways that depend on | |
1849 their lexical scoping; these distinctions vanish in dynamically | |
1850 scoped Emacs Lisp.) | |
1851 @end defspec | |
1852 | |
1853 @node Macro Bindings, , Function Bindings, Variable Bindings | |
1854 @subsection Macro Bindings | |
1855 | |
1856 @noindent | |
1857 These forms create local macros and ``symbol macros.'' | |
1858 | |
1859 @defspec macrolet (bindings@dots{}) forms@dots{} | |
1860 This form is analogous to @code{flet}, but for macros instead of | |
1861 functions. Each @var{binding} is a list of the same form as the | |
1862 arguments to @code{defmacro*} (i.e., a macro name, argument list, | |
1863 and macro-expander forms). The macro is defined accordingly for | |
1864 use within the body of the @code{macrolet}. | |
1865 | |
1866 Because of the nature of macros, @code{macrolet} is lexically | |
1867 scoped even in Emacs Lisp: The @code{macrolet} binding will | |
1868 affect only calls that appear physically within the body | |
1869 @var{forms}, possibly after expansion of other macros in the | |
1870 body. | |
1871 @end defspec | |
1872 | |
1873 @defspec symbol-macrolet (bindings@dots{}) forms@dots{} | |
1874 This form creates @dfn{symbol macros}, which are macros that look | |
1875 like variable references rather than function calls. Each | |
1876 @var{binding} is a list @samp{(@var{var} @var{expansion})}; | |
1877 any reference to @var{var} within the body @var{forms} is | |
1878 replaced by @var{expansion}. | |
1879 | |
1880 @example | |
1881 (setq bar '(5 . 9)) | |
1882 (symbol-macrolet ((foo (car bar))) | |
1883 (incf foo)) | |
1884 bar | |
1885 @result{} (6 . 9) | |
1886 @end example | |
1887 | |
1888 A @code{setq} of a symbol macro is treated the same as a @code{setf}. | |
1889 I.e., @code{(setq foo 4)} in the above would be equivalent to | |
1890 @code{(setf foo 4)}, which in turn expands to @code{(setf (car bar) 4)}. | |
1891 | |
1892 Likewise, a @code{let} or @code{let*} binding a symbol macro is | |
1893 treated like a @code{letf} or @code{letf*}. This differs from true | |
1894 Common Lisp, where the rules of lexical scoping cause a @code{let} | |
1895 binding to shadow a @code{symbol-macrolet} binding. In this package, | |
1896 only @code{lexical-let} and @code{lexical-let*} will shadow a symbol | |
1897 macro. | |
1898 | |
1899 There is no analogue of @code{defmacro} for symbol macros; all symbol | |
1900 macros are local. A typical use of @code{symbol-macrolet} is in the | |
1901 expansion of another macro: | |
1902 | |
1903 @example | |
1904 (defmacro* my-dolist ((x list) &rest body) | |
1905 (let ((var (gensym))) | |
1906 (list 'loop 'for var 'on list 'do | |
1907 (list* 'symbol-macrolet (list (list x (list 'car var))) | |
1908 body)))) | |
1909 | |
1910 (setq mylist '(1 2 3 4)) | |
1911 (my-dolist (x mylist) (incf x)) | |
1912 mylist | |
1913 @result{} (2 3 4 5) | |
1914 @end example | |
1915 | |
1916 @noindent | |
1917 In this example, the @code{my-dolist} macro is similar to @code{dolist} | |
1918 (@pxref{Iteration}) except that the variable @code{x} becomes a true | |
1919 reference onto the elements of the list. The @code{my-dolist} call | |
1920 shown here expands to | |
1921 | |
1922 @example | |
1923 (loop for G1234 on mylist do | |
1924 (symbol-macrolet ((x (car G1234))) | |
1925 (incf x))) | |
1926 @end example | |
1927 | |
1928 @noindent | |
1929 which in turn expands to | |
1930 | |
1931 @example | |
1932 (loop for G1234 on mylist do (incf (car G1234))) | |
1933 @end example | |
1934 | |
1935 @xref{Loop Facility}, for a description of the @code{loop} macro. | |
1936 This package defines a nonstandard @code{in-ref} loop clause that | |
1937 works much like @code{my-dolist}. | |
1938 @end defspec | |
1939 | |
1940 @node Conditionals, Blocks and Exits, Variable Bindings, Control Structure | |
1941 @section Conditionals | |
1942 | |
1943 @noindent | |
1944 These conditional forms augment Emacs Lisp's simple @code{if}, | |
1945 @code{and}, @code{or}, and @code{cond} forms. | |
1946 | |
1947 @defspec when test forms@dots{} | |
1948 This is a variant of @code{if} where there are no ``else'' forms, | |
1949 and possibly several ``then'' forms. In particular, | |
1950 | |
1951 @example | |
1952 (when @var{test} @var{a} @var{b} @var{c}) | |
1953 @end example | |
1954 | |
1955 @noindent | |
1956 is entirely equivalent to | |
1957 | |
1958 @example | |
1959 (if @var{test} (progn @var{a} @var{b} @var{c}) nil) | |
1960 @end example | |
1961 @end defspec | |
1962 | |
1963 @defspec unless test forms@dots{} | |
1964 This is a variant of @code{if} where there are no ``then'' forms, | |
1965 and possibly several ``else'' forms: | |
1966 | |
1967 @example | |
1968 (unless @var{test} @var{a} @var{b} @var{c}) | |
1969 @end example | |
1970 | |
1971 @noindent | |
1972 is entirely equivalent to | |
1973 | |
1974 @example | |
1975 (when (not @var{test}) @var{a} @var{b} @var{c}) | |
1976 @end example | |
1977 @end defspec | |
1978 | |
1979 @defspec case keyform clause@dots{} | |
1980 This macro evaluates @var{keyform}, then compares it with the key | |
1981 values listed in the various @var{clause}s. Whichever clause matches | |
1982 the key is executed; comparison is done by @code{eql}. If no clause | |
1983 matches, the @code{case} form returns @code{nil}. The clauses are | |
1984 of the form | |
1985 | |
1986 @example | |
1987 (@var{keylist} @var{body-forms}@dots{}) | |
1988 @end example | |
1989 | |
1990 @noindent | |
1991 where @var{keylist} is a list of key values. If there is exactly | |
1992 one value, and it is not a cons cell or the symbol @code{nil} or | |
1993 @code{t}, then it can be used by itself as a @var{keylist} without | |
1994 being enclosed in a list. All key values in the @code{case} form | |
1995 must be distinct. The final clauses may use @code{t} in place of | |
1996 a @var{keylist} to indicate a default clause that should be taken | |
1997 if none of the other clauses match. (The symbol @code{otherwise} | |
1998 is also recognized in place of @code{t}. To make a clause that | |
1999 matches the actual symbol @code{t}, @code{nil}, or @code{otherwise}, | |
2000 enclose the symbol in a list.) | |
2001 | |
2002 For example, this expression reads a keystroke, then does one of | |
2003 four things depending on whether it is an @samp{a}, a @samp{b}, | |
2004 a @key{RET} or @key{LFD}, or anything else. | |
2005 | |
2006 @example | |
2007 (case (read-char) | |
2008 (?a (do-a-thing)) | |
2009 (?b (do-b-thing)) | |
2010 ((?\r ?\n) (do-ret-thing)) | |
2011 (t (do-other-thing))) | |
2012 @end example | |
2013 @end defspec | |
2014 | |
2015 @defspec ecase keyform clause@dots{} | |
2016 This macro is just like @code{case}, except that if the key does | |
2017 not match any of the clauses, an error is signalled rather than | |
2018 simply returning @code{nil}. | |
2019 @end defspec | |
2020 | |
2021 @defspec typecase keyform clause@dots{} | |
2022 This macro is a version of @code{case} that checks for types | |
2023 rather than values. Each @var{clause} is of the form | |
2024 @samp{(@var{type} @var{body}...)}. @xref{Type Predicates}, | |
2025 for a description of type specifiers. For example, | |
2026 | |
2027 @example | |
2028 (typecase x | |
2029 (integer (munch-integer x)) | |
2030 (float (munch-float x)) | |
2031 (string (munch-integer (string-to-int x))) | |
2032 (t (munch-anything x))) | |
2033 @end example | |
2034 | |
2035 The type specifier @code{t} matches any type of object; the word | |
2036 @code{otherwise} is also allowed. To make one clause match any of | |
2037 several types, use an @code{(or ...)} type specifier. | |
2038 @end defspec | |
2039 | |
2040 @defspec etypecase keyform clause@dots{} | |
2041 This macro is just like @code{typecase}, except that if the key does | |
2042 not match any of the clauses, an error is signalled rather than | |
2043 simply returning @code{nil}. | |
2044 @end defspec | |
2045 | |
2046 @node Blocks and Exits, Iteration, Conditionals, Control Structure | |
2047 @section Blocks and Exits | |
2048 | |
2049 @noindent | |
2050 Common Lisp @dfn{blocks} provide a non-local exit mechanism very | |
2051 similar to @code{catch} and @code{throw}, but lexically rather than | |
2052 dynamically scoped. This package actually implements @code{block} | |
2053 in terms of @code{catch}; however, the lexical scoping allows the | |
2054 optimizing byte-compiler to omit the costly @code{catch} step if the | |
2055 body of the block does not actually @code{return-from} the block. | |
2056 | |
2057 @defspec block name forms@dots{} | |
2058 The @var{forms} are evaluated as if by a @code{progn}. However, | |
2059 if any of the @var{forms} execute @code{(return-from @var{name})}, | |
2060 they will jump out and return directly from the @code{block} form. | |
2061 The @code{block} returns the result of the last @var{form} unless | |
2062 a @code{return-from} occurs. | |
2063 | |
2064 The @code{block}/@code{return-from} mechanism is quite similar to | |
2065 the @code{catch}/@code{throw} mechanism. The main differences are | |
2066 that block @var{name}s are unevaluated symbols, rather than forms | |
2067 (such as quoted symbols) which evaluate to a tag at run-time; and | |
2068 also that blocks are lexically scoped whereas @code{catch}/@code{throw} | |
2069 are dynamically scoped. This means that functions called from the | |
2070 body of a @code{catch} can also @code{throw} to the @code{catch}, | |
2071 but the @code{return-from} referring to a block name must appear | |
2072 physically within the @var{forms} that make up the body of the block. | |
2073 They may not appear within other called functions, although they may | |
2074 appear within macro expansions or @code{lambda}s in the body. Block | |
2075 names and @code{catch} names form independent name-spaces. | |
2076 | |
2077 In true Common Lisp, @code{defun} and @code{defmacro} surround | |
2078 the function or expander bodies with implicit blocks with the | |
2079 same name as the function or macro. This does not occur in Emacs | |
2080 Lisp, but this package provides @code{defun*} and @code{defmacro*} | |
2081 forms which do create the implicit block. | |
2082 | |
2083 The Common Lisp looping constructs defined by this package, | |
2084 such as @code{loop} and @code{dolist}, also create implicit blocks | |
2085 just as in Common Lisp. | |
2086 | |
2087 Because they are implemented in terms of Emacs Lisp @code{catch} | |
2088 and @code{throw}, blocks have the same overhead as actual | |
2089 @code{catch} constructs (roughly two function calls). However, | |
2090 Zawinski and Furuseth's optimizing byte compiler (standard in | |
2091 Emacs 19) will optimize away the @code{catch} if the block does | |
2092 not in fact contain any @code{return} or @code{return-from} calls | |
2093 that jump to it. This means that @code{do} loops and @code{defun*} | |
2094 functions which don't use @code{return} don't pay the overhead to | |
2095 support it. | |
2096 @end defspec | |
2097 | |
2098 @defspec return-from name [result] | |
2099 This macro returns from the block named @var{name}, which must be | |
2100 an (unevaluated) symbol. If a @var{result} form is specified, it | |
2101 is evaluated to produce the result returned from the @code{block}. | |
2102 Otherwise, @code{nil} is returned. | |
2103 @end defspec | |
2104 | |
2105 @defspec return [result] | |
2106 This macro is exactly like @code{(return-from nil @var{result})}. | |
2107 Common Lisp loops like @code{do} and @code{dolist} implicitly enclose | |
2108 themselves in @code{nil} blocks. | |
2109 @end defspec | |
2110 | |
2111 @node Iteration, Loop Facility, Blocks and Exits, Control Structure | |
2112 @section Iteration | |
2113 | |
2114 @noindent | |
2115 The macros described here provide more sophisticated, high-level | |
2116 looping constructs to complement Emacs Lisp's basic @code{while} | |
2117 loop. | |
2118 | |
2119 @defspec loop forms@dots{} | |
2120 The @dfn{CL} package supports both the simple, old-style meaning of | |
2121 @code{loop} and the extremely powerful and flexible feature known as | |
2122 the @dfn{Loop Facility} or @dfn{Loop Macro}. This more advanced | |
2123 facility is discussed in the following section; @pxref{Loop Facility}. | |
2124 The simple form of @code{loop} is described here. | |
2125 | |
2126 If @code{loop} is followed by zero or more Lisp expressions, | |
2127 then @code{(loop @var{exprs}@dots{})} simply creates an infinite | |
2128 loop executing the expressions over and over. The loop is | |
2129 enclosed in an implicit @code{nil} block. Thus, | |
2130 | |
2131 @example | |
2132 (loop (foo) (if (no-more) (return 72)) (bar)) | |
2133 @end example | |
2134 | |
2135 @noindent | |
2136 is exactly equivalent to | |
2137 | |
2138 @example | |
2139 (block nil (while t (foo) (if (no-more) (return 72)) (bar))) | |
2140 @end example | |
2141 | |
2142 If any of the expressions are plain symbols, the loop is instead | |
2143 interpreted as a Loop Macro specification as described later. | |
2144 (This is not a restriction in practice, since a plain symbol | |
2145 in the above notation would simply access and throw away the | |
2146 value of a variable.) | |
2147 @end defspec | |
2148 | |
2149 @defspec do (spec@dots{}) (end-test [result@dots{}]) forms@dots{} | |
2150 This macro creates a general iterative loop. Each @var{spec} is | |
2151 of the form | |
2152 | |
2153 @example | |
2154 (@var{var} [@var{init} [@var{step}]]) | |
2155 @end example | |
2156 | |
2157 The loop works as follows: First, each @var{var} is bound to the | |
2158 associated @var{init} value as if by a @code{let} form. Then, in | |
2159 each iteration of the loop, the @var{end-test} is evaluated; if | |
2160 true, the loop is finished. Otherwise, the body @var{forms} are | |
2161 evaluated, then each @var{var} is set to the associated @var{step} | |
2162 expression (as if by a @code{psetq} form) and the next iteration | |
2163 begins. Once the @var{end-test} becomes true, the @var{result} | |
2164 forms are evaluated (with the @var{var}s still bound to their | |
2165 values) to produce the result returned by @code{do}. | |
2166 | |
2167 The entire @code{do} loop is enclosed in an implicit @code{nil} | |
2168 block, so that you can use @code{(return)} to break out of the | |
2169 loop at any time. | |
2170 | |
2171 If there are no @var{result} forms, the loop returns @code{nil}. | |
2172 If a given @var{var} has no @var{step} form, it is bound to its | |
2173 @var{init} value but not otherwise modified during the @code{do} | |
2174 loop (unless the code explicitly modifies it); this case is just | |
2175 a shorthand for putting a @code{(let ((@var{var} @var{init})) @dots{})} | |
2176 around the loop. If @var{init} is also omitted it defaults to | |
2177 @code{nil}, and in this case a plain @samp{@var{var}} can be used | |
2178 in place of @samp{(@var{var})}, again following the analogy with | |
2179 @code{let}. | |
2180 | |
2181 This example (from Steele) illustrates a loop which applies the | |
2182 function @code{f} to successive pairs of values from the lists | |
2183 @code{foo} and @code{bar}; it is equivalent to the call | |
2184 @code{(mapcar* 'f foo bar)}. Note that this loop has no body | |
2185 @var{forms} at all, performing all its work as side effects of | |
2186 the rest of the loop. | |
2187 | |
2188 @example | |
2189 (do ((x foo (cdr x)) | |
2190 (y bar (cdr y)) | |
2191 (z nil (cons (f (car x) (car y)) z))) | |
2192 ((or (null x) (null y)) | |
2193 (nreverse z))) | |
2194 @end example | |
2195 @end defspec | |
2196 | |
2197 @defspec do* (spec@dots{}) (end-test [result@dots{}]) forms@dots{} | |
2198 This is to @code{do} what @code{let*} is to @code{let}. In | |
2199 particular, the initial values are bound as if by @code{let*} | |
2200 rather than @code{let}, and the steps are assigned as if by | |
2201 @code{setq} rather than @code{psetq}. | |
2202 | |
2203 Here is another way to write the above loop: | |
2204 | |
2205 @example | |
2206 (do* ((xp foo (cdr xp)) | |
2207 (yp bar (cdr yp)) | |
2208 (x (car xp) (car xp)) | |
2209 (y (car yp) (car yp)) | |
2210 z) | |
2211 ((or (null xp) (null yp)) | |
2212 (nreverse z)) | |
2213 (push (f x y) z)) | |
2214 @end example | |
2215 @end defspec | |
2216 | |
2217 @defspec dolist (var list [result]) forms@dots{} | |
2218 This is a more specialized loop which iterates across the elements | |
2219 of a list. @var{list} should evaluate to a list; the body @var{forms} | |
2220 are executed with @var{var} bound to each element of the list in | |
2221 turn. Finally, the @var{result} form (or @code{nil}) is evaluated | |
2222 with @var{var} bound to @code{nil} to produce the result returned by | |
2223 the loop. The loop is surrounded by an implicit @code{nil} block. | |
2224 @end defspec | |
2225 | |
2226 @defspec dotimes (var count [result]) forms@dots{} | |
2227 This is a more specialized loop which iterates a specified number | |
2228 of times. The body is executed with @var{var} bound to the integers | |
2229 from zero (inclusive) to @var{count} (exclusive), in turn. Then | |
2230 the @code{result} form is evaluated with @var{var} bound to the total | |
2231 number of iterations that were done (i.e., @code{(max 0 @var{count})}) | |
2232 to get the return value for the loop form. The loop is surrounded | |
2233 by an implicit @code{nil} block. | |
2234 @end defspec | |
2235 | |
2236 @defspec do-symbols (var [obarray [result]]) forms@dots{} | |
2237 This loop iterates over all interned symbols. If @var{obarray} | |
2238 is specified and is not @code{nil}, it loops over all symbols in | |
2239 that obarray. For each symbol, the body @var{forms} are evaluated | |
2240 with @var{var} bound to that symbol. The symbols are visited in | |
2241 an unspecified order. Afterward the @var{result} form, if any, | |
2242 is evaluated (with @var{var} bound to @code{nil}) to get the return | |
2243 value. The loop is surrounded by an implicit @code{nil} block. | |
2244 @end defspec | |
2245 | |
2246 @defspec do-all-symbols (var [result]) forms@dots{} | |
2247 This is identical to @code{do-symbols} except that the @var{obarray} | |
2248 argument is omitted; it always iterates over the default obarray. | |
2249 @end defspec | |
2250 | |
2251 @xref{Mapping over Sequences}, for some more functions for | |
2252 iterating over vectors or lists. | |
2253 | |
2254 @node Loop Facility, Multiple Values, Iteration, Control Structure | |
2255 @section Loop Facility | |
2256 | |
2257 @noindent | |
2258 A common complaint with Lisp's traditional looping constructs is | |
2259 that they are either too simple and limited, such as Common Lisp's | |
2260 @code{dotimes} or Emacs Lisp's @code{while}, or too unreadable and | |
2261 obscure, like Common Lisp's @code{do} loop. | |
2262 | |
2263 To remedy this, recent versions of Common Lisp have added a new | |
2264 construct called the ``Loop Facility'' or ``@code{loop} macro,'' | |
2265 with an easy-to-use but very powerful and expressive syntax. | |
2266 | |
2267 @menu | |
2268 * Loop Basics:: `loop' macro, basic clause structure | |
2269 * Loop Examples:: Working examples of `loop' macro | |
2270 * For Clauses:: Clauses introduced by `for' or `as' | |
2271 * Iteration Clauses:: `repeat', `while', `thereis', etc. | |
2272 * Accumulation Clauses:: `collect', `sum', `maximize', etc. | |
2273 * Other Clauses:: `with', `if', `initially', `finally' | |
2274 @end menu | |
2275 | |
2276 @node Loop Basics, Loop Examples, Loop Facility, Loop Facility | |
2277 @subsection Loop Basics | |
2278 | |
2279 @noindent | |
2280 The @code{loop} macro essentially creates a mini-language within | |
2281 Lisp that is specially tailored for describing loops. While this | |
2282 language is a little strange-looking by the standards of regular Lisp, | |
2283 it turns out to be very easy to learn and well-suited to its purpose. | |
2284 | |
2285 Since @code{loop} is a macro, all parsing of the loop language | |
2286 takes place at byte-compile time; compiled @code{loop}s are just | |
2287 as efficient as the equivalent @code{while} loops written longhand. | |
2288 | |
2289 @defspec loop clauses@dots{} | |
2290 A loop construct consists of a series of @var{clause}s, each | |
2291 introduced by a symbol like @code{for} or @code{do}. Clauses | |
2292 are simply strung together in the argument list of @code{loop}, | |
2293 with minimal extra parentheses. The various types of clauses | |
2294 specify initializations, such as the binding of temporary | |
2295 variables, actions to be taken in the loop, stepping actions, | |
2296 and final cleanup. | |
2297 | |
2298 Common Lisp specifies a certain general order of clauses in a | |
2299 loop: | |
2300 | |
2301 @example | |
2302 (loop @var{name-clause} | |
2303 @var{var-clauses}@dots{} | |
2304 @var{action-clauses}@dots{}) | |
2305 @end example | |
2306 | |
2307 The @var{name-clause} optionally gives a name to the implicit | |
2308 block that surrounds the loop. By default, the implicit block | |
2309 is named @code{nil}. The @var{var-clauses} specify what | |
2310 variables should be bound during the loop, and how they should | |
2311 be modified or iterated throughout the course of the loop. The | |
2312 @var{action-clauses} are things to be done during the loop, such | |
2313 as computing, collecting, and returning values. | |
2314 | |
2315 The Emacs version of the @code{loop} macro is less restrictive about | |
2316 the order of clauses, but things will behave most predictably if | |
2317 you put the variable-binding clauses @code{with}, @code{for}, and | |
2318 @code{repeat} before the action clauses. As in Common Lisp, | |
2319 @code{initially} and @code{finally} clauses can go anywhere. | |
2320 | |
2321 Loops generally return @code{nil} by default, but you can cause | |
2322 them to return a value by using an accumulation clause like | |
2323 @code{collect}, an end-test clause like @code{always}, or an | |
2324 explicit @code{return} clause to jump out of the implicit block. | |
2325 (Because the loop body is enclosed in an implicit block, you can | |
2326 also use regular Lisp @code{return} or @code{return-from} to | |
2327 break out of the loop.) | |
2328 @end defspec | |
2329 | |
2330 The following sections give some examples of the Loop Macro in | |
2331 action, and describe the particular loop clauses in great detail. | |
2332 Consult the second edition of Steele's @dfn{Common Lisp, the Language}, | |
2333 for additional discussion and examples of the @code{loop} macro. | |
2334 | |
2335 @node Loop Examples, For Clauses, Loop Basics, Loop Facility | |
2336 @subsection Loop Examples | |
2337 | |
2338 @noindent | |
2339 Before listing the full set of clauses that are allowed, let's | |
2340 look at a few example loops just to get a feel for the @code{loop} | |
2341 language. | |
2342 | |
2343 @example | |
2344 (loop for buf in (buffer-list) | |
2345 collect (buffer-file-name buf)) | |
2346 @end example | |
2347 | |
2348 @noindent | |
2349 This loop iterates over all Emacs buffers, using the list | |
2350 returned by @code{buffer-list}. For each buffer @code{buf}, | |
2351 it calls @code{buffer-file-name} and collects the results into | |
2352 a list, which is then returned from the @code{loop} construct. | |
2353 The result is a list of the file names of all the buffers in | |
2354 Emacs' memory. The words @code{for}, @code{in}, and @code{collect} | |
2355 are reserved words in the @code{loop} language. | |
2356 | |
2357 @example | |
2358 (loop repeat 20 do (insert "Yowsa\n")) | |
2359 @end example | |
2360 | |
2361 @noindent | |
2362 This loop inserts the phrase ``Yowsa'' twenty times in the | |
2363 current buffer. | |
2364 | |
2365 @example | |
2366 (loop until (eobp) do (munch-line) (forward-line 1)) | |
2367 @end example | |
2368 | |
2369 @noindent | |
2370 This loop calls @code{munch-line} on every line until the end | |
2371 of the buffer. If point is already at the end of the buffer, | |
2372 the loop exits immediately. | |
2373 | |
2374 @example | |
2375 (loop do (munch-line) until (eobp) do (forward-line 1)) | |
2376 @end example | |
2377 | |
2378 @noindent | |
2379 This loop is similar to the above one, except that @code{munch-line} | |
2380 is always called at least once. | |
2381 | |
2382 @example | |
2383 (loop for x from 1 to 100 | |
2384 for y = (* x x) | |
2385 until (>= y 729) | |
2386 finally return (list x (= y 729))) | |
2387 @end example | |
2388 | |
2389 @noindent | |
2390 This more complicated loop searches for a number @code{x} whose | |
2391 square is 729. For safety's sake it only examines @code{x} | |
2392 values up to 100; dropping the phrase @samp{to 100} would | |
2393 cause the loop to count upwards with no limit. The second | |
2394 @code{for} clause defines @code{y} to be the square of @code{x} | |
2395 within the loop; the expression after the @code{=} sign is | |
2396 reevaluated each time through the loop. The @code{until} | |
2397 clause gives a condition for terminating the loop, and the | |
2398 @code{finally} clause says what to do when the loop finishes. | |
2399 (This particular example was written less concisely than it | |
2400 could have been, just for the sake of illustration.) | |
2401 | |
2402 Note that even though this loop contains three clauses (two | |
2403 @code{for}s and an @code{until}) that would have been enough to | |
2404 define loops all by themselves, it still creates a single loop | |
2405 rather than some sort of triple-nested loop. You must explicitly | |
2406 nest your @code{loop} constructs if you want nested loops. | |
2407 | |
2408 @node For Clauses, Iteration Clauses, Loop Examples, Loop Facility | |
2409 @subsection For Clauses | |
2410 | |
2411 @noindent | |
2412 Most loops are governed by one or more @code{for} clauses. | |
2413 A @code{for} clause simultaneously describes variables to be | |
2414 bound, how those variables are to be stepped during the loop, | |
2415 and usually an end condition based on those variables. | |
2416 | |
2417 The word @code{as} is a synonym for the word @code{for}. This | |
2418 word is followed by a variable name, then a word like @code{from} | |
2419 or @code{across} that describes the kind of iteration desired. | |
2420 In Common Lisp, the phrase @code{being the} sometimes precedes | |
2421 the type of iteration; in this package both @code{being} and | |
2422 @code{the} are optional. The word @code{each} is a synonym | |
2423 for @code{the}, and the word that follows it may be singular | |
2424 or plural: @samp{for x being the elements of y} or | |
2425 @samp{for x being each element of y}. Which form you use | |
2426 is purely a matter of style. | |
2427 | |
2428 The variable is bound around the loop as if by @code{let}: | |
2429 | |
2430 @example | |
2431 (setq i 'happy) | |
2432 (loop for i from 1 to 10 do (do-something-with i)) | |
2433 i | |
2434 @result{} happy | |
2435 @end example | |
2436 | |
2437 @table @code | |
2438 @item for @var{var} from @var{expr1} to @var{expr2} by @var{expr3} | |
2439 This type of @code{for} clause creates a counting loop. Each of | |
2440 the three sub-terms is optional, though there must be at least one | |
2441 term so that the clause is marked as a counting clause. | |
2442 | |
2443 The three expressions are the starting value, the ending value, and | |
2444 the step value, respectively, of the variable. The loop counts | |
2445 upwards by default (@var{expr3} must be positive), from @var{expr1} | |
2446 to @var{expr2} inclusively. If you omit the @code{from} term, the | |
2447 loop counts from zero; if you omit the @code{to} term, the loop | |
2448 counts forever without stopping (unless stopped by some other | |
2449 loop clause, of course); if you omit the @code{by} term, the loop | |
2450 counts in steps of one. | |
2451 | |
2452 You can replace the word @code{from} with @code{upfrom} or | |
2453 @code{downfrom} to indicate the direction of the loop. Likewise, | |
2454 you can replace @code{to} with @code{upto} or @code{downto}. | |
2455 For example, @samp{for x from 5 downto 1} executes five times | |
2456 with @code{x} taking on the integers from 5 down to 1 in turn. | |
2457 Also, you can replace @code{to} with @code{below} or @code{above}, | |
2458 which are like @code{upto} and @code{downto} respectively except | |
2459 that they are exclusive rather than inclusive limits: | |
2460 | |
2461 @example | |
2462 (loop for x to 10 collect x) | |
2463 @result{} (0 1 2 3 4 5 6 7 8 9 10) | |
2464 (loop for x below 10 collect x) | |
2465 @result{} (0 1 2 3 4 5 6 7 8 9) | |
2466 @end example | |
2467 | |
2468 The @code{by} value is always positive, even for downward-counting | |
2469 loops. Some sort of @code{from} value is required for downward | |
2470 loops; @samp{for x downto 5} is not a legal loop clause all by | |
2471 itself. | |
2472 | |
2473 @item for @var{var} in @var{list} by @var{function} | |
2474 This clause iterates @var{var} over all the elements of @var{list}, | |
2475 in turn. If you specify the @code{by} term, then @var{function} | |
2476 is used to traverse the list instead of @code{cdr}; it must be a | |
2477 function taking one argument. For example: | |
2478 | |
2479 @example | |
2480 (loop for x in '(1 2 3 4 5 6) collect (* x x)) | |
2481 @result{} (1 4 9 16 25 36) | |
2482 (loop for x in '(1 2 3 4 5 6) by 'cddr collect (* x x)) | |
2483 @result{} (1 9 25) | |
2484 @end example | |
2485 | |
2486 @item for @var{var} on @var{list} by @var{function} | |
2487 This clause iterates @var{var} over all the cons cells of @var{list}. | |
2488 | |
2489 @example | |
2490 (loop for x on '(1 2 3 4) collect x) | |
2491 @result{} ((1 2 3 4) (2 3 4) (3 4) (4)) | |
2492 @end example | |
2493 | |
2494 With @code{by}, there is no real reason that the @code{on} expression | |
2495 must be a list. For example: | |
2496 | |
2497 @example | |
2498 (loop for x on first-animal by 'next-animal collect x) | |
2499 @end example | |
2500 | |
2501 @noindent | |
2502 where @code{(next-animal x)} takes an ``animal'' @var{x} and returns | |
2503 the next in the (assumed) sequence of animals, or @code{nil} if | |
2504 @var{x} was the last animal in the sequence. | |
2505 | |
2506 @item for @var{var} in-ref @var{list} by @var{function} | |
2507 This is like a regular @code{in} clause, but @var{var} becomes | |
2508 a @code{setf}-able ``reference'' onto the elements of the list | |
2509 rather than just a temporary variable. For example, | |
2510 | |
2511 @example | |
2512 (loop for x in-ref my-list do (incf x)) | |
2513 @end example | |
2514 | |
2515 @noindent | |
2516 increments every element of @code{my-list} in place. This clause | |
2517 is an extension to standard Common Lisp. | |
2518 | |
2519 @item for @var{var} across @var{array} | |
2520 This clause iterates @var{var} over all the elements of @var{array}, | |
2521 which may be a vector or a string. | |
2522 | |
2523 @example | |
2524 (loop for x across "aeiou" | |
2525 do (use-vowel (char-to-string x))) | |
2526 @end example | |
2527 | |
2528 @item for @var{var} across-ref @var{array} | |
2529 This clause iterates over an array, with @var{var} a @code{setf}-able | |
2530 reference onto the elements; see @code{in-ref} above. | |
2531 | |
2532 @item for @var{var} being the elements of @var{sequence} | |
2533 This clause iterates over the elements of @var{sequence}, which may | |
2534 be a list, vector, or string. Since the type must be determined | |
2535 at run-time, this is somewhat less efficient than @code{in} or | |
2536 @code{across}. The clause may be followed by the additional term | |
2537 @samp{using (index @var{var2})} to cause @var{var2} to be bound to | |
2538 the successive indices (starting at 0) of the elements. | |
2539 | |
2540 This clause type is taken from older versions of the @code{loop} macro, | |
2541 and is not present in modern Common Lisp. The @samp{using (sequence ...)} | |
2542 term of the older macros is not supported. | |
2543 | |
2544 @item for @var{var} being the elements of-ref @var{sequence} | |
2545 This clause iterates over a sequence, with @var{var} a @code{setf}-able | |
2546 reference onto the elements; see @code{in-ref} above. | |
2547 | |
2548 @item for @var{var} being the symbols [of @var{obarray}] | |
2549 This clause iterates over symbols, either over all interned symbols | |
2550 or over all symbols in @var{obarray}. The loop is executed with | |
2551 @var{var} bound to each symbol in turn. The symbols are visited in | |
2552 an unspecified order. | |
2553 | |
2554 As an example, | |
2555 | |
2556 @example | |
2557 (loop for sym being the symbols | |
2558 when (fboundp sym) | |
2559 when (string-match "^map" (symbol-name sym)) | |
2560 collect sym) | |
2561 @end example | |
2562 | |
2563 @noindent | |
2564 returns a list of all the functions whose names begin with @samp{map}. | |
2565 | |
2566 The Common Lisp words @code{external-symbols} and @code{present-symbols} | |
2567 are also recognized but are equivalent to @code{symbols} in Emacs Lisp. | |
2568 | |
2569 Due to a minor implementation restriction, it will not work to have | |
2570 more than one @code{for} clause iterating over symbols, hash tables, | |
2571 keymaps, overlays, or intervals in a given @code{loop}. Fortunately, | |
2572 it would rarely if ever be useful to do so. It @emph{is} legal to mix | |
2573 one of these types of clauses with other clauses like @code{for ... to} | |
2574 or @code{while}. | |
2575 | |
2576 @item for @var{var} being the hash-keys of @var{hash-table} | |
2577 This clause iterates over the entries in @var{hash-table}. For each | |
2578 hash table entry, @var{var} is bound to the entry's key. If you write | |
2579 @samp{the hash-values} instead, @var{var} is bound to the values | |
2580 of the entries. The clause may be followed by the additional | |
2581 term @samp{using (hash-values @var{var2})} (where @code{hash-values} | |
2582 is the opposite word of the word following @code{the}) to cause | |
2583 @var{var} and @var{var2} to be bound to the two parts of each | |
2584 hash table entry. | |
2585 | |
2586 @item for @var{var} being the key-codes of @var{keymap} | |
2587 This clause iterates over the entries in @var{keymap}. In GNU Emacs 18 | |
2588 and 19, keymaps are either alists or vectors, and key-codes are integers | |
2589 or symbols. In XEmacs, keymaps are a special new data type, and | |
2590 key-codes are symbols or lists of symbols. The iteration does not enter | |
2591 nested keymaps or inherited (parent) keymaps. You can use @samp{the | |
2592 key-bindings} to access the commands bound to the keys rather than the | |
2593 key codes, and you can add a @code{using} clause to access both the | |
2594 codes and the bindings together. | |
2595 | |
2596 @item for @var{var} being the key-seqs of @var{keymap} | |
2597 This clause iterates over all key sequences defined by @var{keymap} | |
2598 and its nested keymaps, where @var{var} takes on values which are | |
2599 strings in Emacs 18 or vectors in Emacs 19. The strings or vectors | |
2600 are reused for each iteration, so you must copy them if you wish to keep | |
2601 them permanently. You can add a @samp{using (key-bindings ...)} | |
2602 clause to get the command bindings as well. | |
2603 | |
2604 @item for @var{var} being the overlays [of @var{buffer}] @dots{} | |
2605 This clause iterates over the Emacs 19 ``overlays'' or XEmacs | |
2606 ``extents'' of a buffer (the clause @code{extents} is synonymous with | |
2607 @code{overlays}). Under Emacs 18, this clause iterates zero times. If | |
2608 the @code{of} term is omitted, the current buffer is used. This clause | |
2609 also accepts optional @samp{from @var{pos}} and @samp{to @var{pos}} | |
2610 terms, limiting the clause to overlays which overlap the specified | |
2611 region. | |
2612 | |
2613 @item for @var{var} being the intervals [of @var{buffer}] @dots{} | |
2614 This clause iterates over all intervals of a buffer with constant | |
2615 text properties. The variable @var{var} will be bound to conses | |
2616 of start and end positions, where one start position is always equal | |
2617 to the previous end position. The clause allows @code{of}, | |
2618 @code{from}, @code{to}, and @code{property} terms, where the latter | |
2619 term restricts the search to just the specified property. The | |
2620 @code{of} term may specify either a buffer or a string. This | |
2621 clause is useful only in GNU Emacs 19; in other versions, all | |
2622 buffers and strings consist of a single interval. | |
2623 | |
2624 @item for @var{var} being the frames | |
2625 This clause iterates over all frames, i.e., X window system windows | |
2626 open on Emacs files. This clause works only under Emacs 19. The | |
2627 clause @code{screens} is a synonym for @code{frames}. The frames | |
2628 are visited in @code{next-frame} order starting from | |
2629 @code{selected-frame}. | |
2630 | |
2631 @item for @var{var} being the windows [of @var{frame}] | |
2632 This clause iterates over the windows (in the Emacs sense) of | |
2633 the current frame, or of the specified @var{frame}. (In Emacs 18 | |
2634 there is only ever one frame, and the @code{of} term is not | |
2635 allowed there.) | |
2636 | |
2637 @item for @var{var} being the buffers | |
2638 This clause iterates over all buffers in Emacs. It is equivalent | |
2639 to @samp{for @var{var} in (buffer-list)}. | |
2640 | |
2641 @item for @var{var} = @var{expr1} then @var{expr2} | |
2642 This clause does a general iteration. The first time through | |
2643 the loop, @var{var} will be bound to @var{expr1}. On the second | |
2644 and successive iterations it will be set by evaluating @var{expr2} | |
2645 (which may refer to the old value of @var{var}). For example, | |
2646 these two loops are effectively the same: | |
2647 | |
2648 @example | |
2649 (loop for x on my-list by 'cddr do ...) | |
2650 (loop for x = my-list then (cddr x) while x do ...) | |
2651 @end example | |
2652 | |
2653 Note that this type of @code{for} clause does not imply any sort | |
2654 of terminating condition; the above example combines it with a | |
2655 @code{while} clause to tell when to end the loop. | |
2656 | |
2657 If you omit the @code{then} term, @var{expr1} is used both for | |
2658 the initial setting and for successive settings: | |
2659 | |
2660 @example | |
2661 (loop for x = (random) when (> x 0) return x) | |
2662 @end example | |
2663 | |
2664 @noindent | |
2665 This loop keeps taking random numbers from the @code{(random)} | |
2666 function until it gets a positive one, which it then returns. | |
2667 @end table | |
2668 | |
2669 If you include several @code{for} clauses in a row, they are | |
2670 treated sequentially (as if by @code{let*} and @code{setq}). | |
2671 You can instead use the word @code{and} to link the clauses, | |
2672 in which case they are processed in parallel (as if by @code{let} | |
2673 and @code{psetq}). | |
2674 | |
2675 @example | |
2676 (loop for x below 5 for y = nil then x collect (list x y)) | |
2677 @result{} ((0 nil) (1 1) (2 2) (3 3) (4 4)) | |
2678 (loop for x below 5 and y = nil then x collect (list x y)) | |
2679 @result{} ((0 nil) (1 0) (2 1) (3 2) (4 3)) | |
2680 @end example | |
2681 | |
2682 @noindent | |
2683 In the first loop, @code{y} is set based on the value of @code{x} | |
2684 that was just set by the previous clause; in the second loop, | |
2685 @code{x} and @code{y} are set simultaneously so @code{y} is set | |
2686 based on the value of @code{x} left over from the previous time | |
2687 through the loop. | |
2688 | |
2689 Another feature of the @code{loop} macro is @dfn{destructuring}, | |
2690 similar in concept to the destructuring provided by @code{defmacro}. | |
2691 The @var{var} part of any @code{for} clause can be given as a list | |
2692 of variables instead of a single variable. The values produced | |
2693 during loop execution must be lists; the values in the lists are | |
2694 stored in the corresponding variables. | |
2695 | |
2696 @example | |
2697 (loop for (x y) in '((2 3) (4 5) (6 7)) collect (+ x y)) | |
2698 @result{} (5 9 13) | |
2699 @end example | |
2700 | |
2701 In loop destructuring, if there are more values than variables | |
2702 the trailing values are ignored, and if there are more variables | |
2703 than values the trailing variables get the value @code{nil}. | |
2704 If @code{nil} is used as a variable name, the corresponding | |
2705 values are ignored. Destructuring may be nested, and dotted | |
2706 lists of variables like @code{(x . y)} are allowed. | |
2707 | |
2708 @node Iteration Clauses, Accumulation Clauses, For Clauses, Loop Facility | |
2709 @subsection Iteration Clauses | |
2710 | |
2711 @noindent | |
2712 Aside from @code{for} clauses, there are several other loop clauses | |
2713 that control the way the loop operates. They might be used by | |
2714 themselves, or in conjunction with one or more @code{for} clauses. | |
2715 | |
2716 @table @code | |
2717 @item repeat @var{integer} | |
2718 This clause simply counts up to the specified number using an | |
2719 internal temporary variable. The loops | |
2720 | |
2721 @example | |
2722 (loop repeat n do ...) | |
2723 (loop for temp to n do ...) | |
2724 @end example | |
2725 | |
2726 @noindent | |
2727 are identical except that the second one forces you to choose | |
2728 a name for a variable you aren't actually going to use. | |
2729 | |
2730 @item while @var{condition} | |
2731 This clause stops the loop when the specified condition (any Lisp | |
2732 expression) becomes @code{nil}. For example, the following two | |
2733 loops are equivalent, except for the implicit @code{nil} block | |
2734 that surrounds the second one: | |
2735 | |
2736 @example | |
2737 (while @var{cond} @var{forms}@dots{}) | |
2738 (loop while @var{cond} do @var{forms}@dots{}) | |
2739 @end example | |
2740 | |
2741 @item until @var{condition} | |
2742 This clause stops the loop when the specified condition is true, | |
2743 i.e., non-@code{nil}. | |
2744 | |
2745 @item always @var{condition} | |
2746 This clause stops the loop when the specified condition is @code{nil}. | |
2747 Unlike @code{while}, it stops the loop using @code{return nil} so that | |
2748 the @code{finally} clauses are not executed. If all the conditions | |
2749 were non-@code{nil}, the loop returns @code{t}: | |
2750 | |
2751 @example | |
2752 (if (loop for size in size-list always (> size 10)) | |
2753 (some-big-sizes) | |
2754 (no-big-sizes)) | |
2755 @end example | |
2756 | |
2757 @item never @var{condition} | |
2758 This clause is like @code{always}, except that the loop returns | |
2759 @code{t} if any conditions were false, or @code{nil} otherwise. | |
2760 | |
2761 @item thereis @var{condition} | |
2762 This clause stops the loop when the specified form is non-@code{nil}; | |
2763 in this case, it returns that non-@code{nil} value. If all the | |
2764 values were @code{nil}, the loop returns @code{nil}. | |
2765 @end table | |
2766 | |
2767 @node Accumulation Clauses, Other Clauses, Iteration Clauses, Loop Facility | |
2768 @subsection Accumulation Clauses | |
2769 | |
2770 @noindent | |
2771 These clauses cause the loop to accumulate information about the | |
2772 specified Lisp @var{form}. The accumulated result is returned | |
2773 from the loop unless overridden, say, by a @code{return} clause. | |
2774 | |
2775 @table @code | |
2776 @item collect @var{form} | |
2777 This clause collects the values of @var{form} into a list. Several | |
2778 examples of @code{collect} appear elsewhere in this manual. | |
2779 | |
2780 The word @code{collecting} is a synonym for @code{collect}, and | |
2781 likewise for the other accumulation clauses. | |
2782 | |
2783 @item append @var{form} | |
2784 This clause collects lists of values into a result list using | |
2785 @code{append}. | |
2786 | |
2787 @item nconc @var{form} | |
2788 This clause collects lists of values into a result list by | |
2789 destructively modifying the lists rather than copying them. | |
2790 | |
2791 @item concat @var{form} | |
2792 This clause concatenates the values of the specified @var{form} | |
2793 into a string. (It and the following clause are extensions to | |
2794 standard Common Lisp.) | |
2795 | |
2796 @item vconcat @var{form} | |
2797 This clause concatenates the values of the specified @var{form} | |
2798 into a vector. | |
2799 | |
2800 @item count @var{form} | |
2801 This clause counts the number of times the specified @var{form} | |
2802 evaluates to a non-@code{nil} value. | |
2803 | |
2804 @item sum @var{form} | |
2805 This clause accumulates the sum of the values of the specified | |
2806 @var{form}, which must evaluate to a number. | |
2807 | |
2808 @item maximize @var{form} | |
2809 This clause accumulates the maximum value of the specified @var{form}, | |
2810 which must evaluate to a number. The return value is undefined if | |
2811 @code{maximize} is executed zero times. | |
2812 | |
2813 @item minimize @var{form} | |
2814 This clause accumulates the minimum value of the specified @var{form}. | |
2815 @end table | |
2816 | |
2817 Accumulation clauses can be followed by @samp{into @var{var}} to | |
2818 cause the data to be collected into variable @var{var} (which is | |
2819 automatically @code{let}-bound during the loop) rather than an | |
2820 unnamed temporary variable. Also, @code{into} accumulations do | |
2821 not automatically imply a return value. The loop must use some | |
2822 explicit mechanism, such as @code{finally return}, to return | |
2823 the accumulated result. | |
2824 | |
2825 It is legal for several accumulation clauses of the same type to | |
2826 accumulate into the same place. From Steele: | |
2827 | |
2828 @example | |
2829 (loop for name in '(fred sue alice joe june) | |
2830 for kids in '((bob ken) () () (kris sunshine) ()) | |
2831 collect name | |
2832 append kids) | |
2833 @result{} (fred bob ken sue alice joe kris sunshine june) | |
2834 @end example | |
2835 | |
2836 @node Other Clauses, , Accumulation Clauses, Loop Facility | |
2837 @subsection Other Clauses | |
2838 | |
2839 @noindent | |
2840 This section describes the remaining loop clauses. | |
2841 | |
2842 @table @code | |
2843 @item with @var{var} = @var{value} | |
2844 This clause binds a variable to a value around the loop, but | |
2845 otherwise leaves the variable alone during the loop. The following | |
2846 loops are basically equivalent: | |
2847 | |
2848 @example | |
2849 (loop with x = 17 do ...) | |
2850 (let ((x 17)) (loop do ...)) | |
2851 (loop for x = 17 then x do ...) | |
2852 @end example | |
2853 | |
2854 Naturally, the variable @var{var} might be used for some purpose | |
2855 in the rest of the loop. For example: | |
2856 | |
2857 @example | |
2858 (loop for x in my-list with res = nil do (push x res) | |
2859 finally return res) | |
2860 @end example | |
2861 | |
2862 This loop inserts the elements of @code{my-list} at the front of | |
2863 a new list being accumulated in @code{res}, then returns the | |
2864 list @code{res} at the end of the loop. The effect is similar | |
2865 to that of a @code{collect} clause, but the list gets reversed | |
2866 by virtue of the fact that elements are being pushed onto the | |
2867 front of @code{res} rather than the end. | |
2868 | |
2869 If you omit the @code{=} term, the variable is initialized to | |
2870 @code{nil}. (Thus the @samp{= nil} in the above example is | |
2871 unnecessary.) | |
2872 | |
2873 Bindings made by @code{with} are sequential by default, as if | |
2874 by @code{let*}. Just like @code{for} clauses, @code{with} clauses | |
2875 can be linked with @code{and} to cause the bindings to be made by | |
2876 @code{let} instead. | |
2877 | |
2878 @item if @var{condition} @var{clause} | |
2879 This clause executes the following loop clause only if the specified | |
2880 condition is true. The following @var{clause} should be an accumulation, | |
2881 @code{do}, @code{return}, @code{if}, or @code{unless} clause. | |
2882 Several clauses may be linked by separating them with @code{and}. | |
2883 These clauses may be followed by @code{else} and a clause or clauses | |
2884 to execute if the condition was false. The whole construct may | |
2885 optionally be followed by the word @code{end} (which may be used to | |
2886 disambiguate an @code{else} or @code{and} in a nested @code{if}). | |
2887 | |
2888 The actual non-@code{nil} value of the condition form is available | |
2889 by the name @code{it} in the ``then'' part. For example: | |
2890 | |
2891 @example | |
2892 (setq funny-numbers '(6 13 -1)) | |
2893 @result{} (6 13 -1) | |
2894 (loop for x below 10 | |
2895 if (oddp x) | |
2896 collect x into odds | |
2897 and if (memq x funny-numbers) return (cdr it) end | |
2898 else | |
2899 collect x into evens | |
2900 finally return (vector odds evens)) | |
2901 @result{} [(1 3 5 7 9) (0 2 4 6 8)] | |
2902 (setq funny-numbers '(6 7 13 -1)) | |
2903 @result{} (6 7 13 -1) | |
2904 (loop <@r{same thing again}>) | |
2905 @result{} (13 -1) | |
2906 @end example | |
2907 | |
2908 Note the use of @code{and} to put two clauses into the ``then'' | |
2909 part, one of which is itself an @code{if} clause. Note also that | |
2910 @code{end}, while normally optional, was necessary here to make | |
2911 it clear that the @code{else} refers to the outermost @code{if} | |
2912 clause. In the first case, the loop returns a vector of lists | |
2913 of the odd and even values of @var{x}. In the second case, the | |
2914 odd number 7 is one of the @code{funny-numbers} so the loop | |
2915 returns early; the actual returned value is based on the result | |
2916 of the @code{memq} call. | |
2917 | |
2918 @item when @var{condition} @var{clause} | |
2919 This clause is just a synonym for @code{if}. | |
2920 | |
2921 @item unless @var{condition} @var{clause} | |
2922 The @code{unless} clause is just like @code{if} except that the | |
2923 sense of the condition is reversed. | |
2924 | |
2925 @item named @var{name} | |
2926 This clause gives a name other than @code{nil} to the implicit | |
2927 block surrounding the loop. The @var{name} is the symbol to be | |
2928 used as the block name. | |
2929 | |
2930 @item initially [do] @var{forms}... | |
2931 This keyword introduces one or more Lisp forms which will be | |
2932 executed before the loop itself begins (but after any variables | |
2933 requested by @code{for} or @code{with} have been bound to their | |
2934 initial values). @code{initially} clauses can appear anywhere; | |
2935 if there are several, they are executed in the order they appear | |
2936 in the loop. The keyword @code{do} is optional. | |
2937 | |
2938 @item finally [do] @var{forms}... | |
2939 This introduces Lisp forms which will be executed after the loop | |
2940 finishes (say, on request of a @code{for} or @code{while}). | |
2941 @code{initially} and @code{finally} clauses may appear anywhere | |
2942 in the loop construct, but they are executed (in the specified | |
2943 order) at the beginning or end, respectively, of the loop. | |
2944 | |
2945 @item finally return @var{form} | |
2946 This says that @var{form} should be executed after the loop | |
2947 is done to obtain a return value. (Without this, or some other | |
2948 clause like @code{collect} or @code{return}, the loop will simply | |
2949 return @code{nil}.) Variables bound by @code{for}, @code{with}, | |
2950 or @code{into} will still contain their final values when @var{form} | |
2951 is executed. | |
2952 | |
2953 @item do @var{forms}... | |
2954 The word @code{do} may be followed by any number of Lisp expressions | |
2955 which are executed as an implicit @code{progn} in the body of the | |
2956 loop. Many of the examples in this section illustrate the use of | |
2957 @code{do}. | |
2958 | |
2959 @item return @var{form} | |
2960 This clause causes the loop to return immediately. The following | |
2961 Lisp form is evaluated to give the return value of the @code{loop} | |
2962 form. The @code{finally} clauses, if any, are not executed. | |
2963 Of course, @code{return} is generally used inside an @code{if} or | |
2964 @code{unless}, as its use in a top-level loop clause would mean | |
2965 the loop would never get to ``loop'' more than once. | |
2966 | |
2967 The clause @samp{return @var{form}} is equivalent to | |
2968 @samp{do (return @var{form})} (or @code{return-from} if the loop | |
2969 was named). The @code{return} clause is implemented a bit more | |
2970 efficiently, though. | |
2971 @end table | |
2972 | |
2973 While there is no high-level way to add user extensions to @code{loop} | |
2974 (comparable to @code{defsetf} for @code{setf}, say), this package | |
2975 does offer two properties called @code{cl-loop-handler} and | |
2976 @code{cl-loop-for-handler} which are functions to be called when | |
2977 a given symbol is encountered as a top-level loop clause or | |
2978 @code{for} clause, respectively. Consult the source code in | |
2979 file @file{cl-macs.el} for details. | |
2980 | |
2981 This package's @code{loop} macro is compatible with that of Common | |
2982 Lisp, except that a few features are not implemented: @code{loop-finish} | |
2983 and data-type specifiers. Naturally, the @code{for} clauses which | |
2984 iterate over keymaps, overlays, intervals, frames, windows, and | |
2985 buffers are Emacs-specific extensions. | |
2986 | |
2987 @node Multiple Values, , Loop Facility, Control Structure | |
2988 @section Multiple Values | |
2989 | |
2990 @noindent | |
2991 Common Lisp functions can return zero or more results. Emacs Lisp | |
2992 functions, by contrast, always return exactly one result. This | |
2993 package makes no attempt to emulate Common Lisp multiple return | |
2994 values; Emacs versions of Common Lisp functions that return more | |
2995 than one value either return just the first value (as in | |
2996 @code{compiler-macroexpand}) or return a list of values (as in | |
2997 @code{get-setf-method}). This package @emph{does} define placeholders | |
2998 for the Common Lisp functions that work with multiple values, but | |
2999 in Emacs Lisp these functions simply operate on lists instead. | |
3000 The @code{values} form, for example, is a synonym for @code{list} | |
3001 in Emacs. | |
3002 | |
3003 @defspec multiple-value-bind (var@dots{}) values-form forms@dots{} | |
3004 This form evaluates @var{values-form}, which must return a list of | |
3005 values. It then binds the @var{var}s to these respective values, | |
3006 as if by @code{let}, and then executes the body @var{forms}. | |
3007 If there are more @var{var}s than values, the extra @var{var}s | |
3008 are bound to @code{nil}. If there are fewer @var{var}s than | |
3009 values, the excess values are ignored. | |
3010 @end defspec | |
3011 | |
3012 @defspec multiple-value-setq (var@dots{}) form | |
3013 This form evaluates @var{form}, which must return a list of values. | |
3014 It then sets the @var{var}s to these respective values, as if by | |
3015 @code{setq}. Extra @var{var}s or values are treated the same as | |
3016 in @code{multiple-value-bind}. | |
3017 @end defspec | |
3018 | |
3019 The older Quiroz package attempted a more faithful (but still | |
3020 imperfect) emulation of Common Lisp multiple values. The old | |
3021 method ``usually'' simulated true multiple values quite well, | |
3022 but under certain circumstances would leave spurious return | |
3023 values in memory where a later, unrelated @code{multiple-value-bind} | |
3024 form would see them. | |
3025 | |
3026 Since a perfect emulation is not feasible in Emacs Lisp, this | |
3027 package opts to keep it as simple and predictable as possible. | |
3028 | |
3029 @node Macros, Declarations, Control Structure, Top | |
3030 @chapter Macros | |
3031 | |
3032 @noindent | |
3033 This package implements the various Common Lisp features of | |
3034 @code{defmacro}, such as destructuring, @code{&environment}, | |
3035 and @code{&body}. Top-level @code{&whole} is not implemented | |
3036 for @code{defmacro} due to technical difficulties. | |
3037 @xref{Argument Lists}. | |
3038 | |
3039 Destructuring is made available to the user by way of the | |
3040 following macro: | |
3041 | |
3042 @defspec destructuring-bind arglist expr forms@dots{} | |
3043 This macro expands to code which executes @var{forms}, with | |
3044 the variables in @var{arglist} bound to the list of values | |
3045 returned by @var{expr}. The @var{arglist} can include all | |
3046 the features allowed for @code{defmacro} argument lists, | |
3047 including destructuring. (The @code{&environment} keyword | |
3048 is not allowed.) The macro expansion will signal an error | |
3049 if @var{expr} returns a list of the wrong number of arguments | |
3050 or with incorrect keyword arguments. | |
3051 @end defspec | |
3052 | |
3053 This package also includes the Common Lisp @code{define-compiler-macro} | |
3054 facility, which allows you to define compile-time expansions and | |
3055 optimizations for your functions. | |
3056 | |
3057 @defspec define-compiler-macro name arglist forms@dots{} | |
3058 This form is similar to @code{defmacro}, except that it only expands | |
3059 calls to @var{name} at compile-time; calls processed by the Lisp | |
3060 interpreter are not expanded, nor are they expanded by the | |
3061 @code{macroexpand} function. | |
3062 | |
3063 The argument list may begin with a @code{&whole} keyword and a | |
3064 variable. This variable is bound to the macro-call form itself, | |
3065 i.e., to a list of the form @samp{(@var{name} @var{args}@dots{})}. | |
3066 If the macro expander returns this form unchanged, then the | |
3067 compiler treats it as a normal function call. This allows | |
3068 compiler macros to work as optimizers for special cases of a | |
3069 function, leaving complicated cases alone. | |
3070 | |
3071 For example, here is a simplified version of a definition that | |
3072 appears as a standard part of this package: | |
3073 | |
3074 @example | |
3075 (define-compiler-macro member* (&whole form a list &rest keys) | |
3076 (if (and (null keys) | |
3077 (eq (car-safe a) 'quote) | |
3078 (not (floatp-safe (cadr a)))) | |
3079 (list 'memq a list) | |
3080 form)) | |
3081 @end example | |
3082 | |
3083 @noindent | |
3084 This definition causes @code{(member* @var{a} @var{list})} to change | |
3085 to a call to the faster @code{memq} in the common case where @var{a} | |
3086 is a non-floating-point constant; if @var{a} is anything else, or | |
3087 if there are any keyword arguments in the call, then the original | |
3088 @code{member*} call is left intact. (The actual compiler macro | |
3089 for @code{member*} optimizes a number of other cases, including | |
3090 common @code{:test} predicates.) | |
3091 @end defspec | |
3092 | |
3093 @defun compiler-macroexpand form | |
3094 This function is analogous to @code{macroexpand}, except that it | |
3095 expands compiler macros rather than regular macros. It returns | |
3096 @var{form} unchanged if it is not a call to a function for which | |
3097 a compiler macro has been defined, or if that compiler macro | |
3098 decided to punt by returning its @code{&whole} argument. Like | |
3099 @code{macroexpand}, it expands repeatedly until it reaches a form | |
3100 for which no further expansion is possible. | |
3101 @end defun | |
3102 | |
3103 @xref{Macro Bindings}, for descriptions of the @code{macrolet} | |
3104 and @code{symbol-macrolet} forms for making ``local'' macro | |
3105 definitions. | |
3106 | |
3107 @node Declarations, Symbols, Macros, Top | |
3108 @chapter Declarations | |
3109 | |
3110 @noindent | |
3111 Common Lisp includes a complex and powerful ``declaration'' | |
3112 mechanism that allows you to give the compiler special hints | |
3113 about the types of data that will be stored in particular variables, | |
3114 and about the ways those variables and functions will be used. This | |
3115 package defines versions of all the Common Lisp declaration forms: | |
3116 @code{declare}, @code{locally}, @code{proclaim}, @code{declaim}, | |
3117 and @code{the}. | |
3118 | |
3119 Most of the Common Lisp declarations are not currently useful in | |
3120 Emacs Lisp, as the byte-code system provides little opportunity | |
3121 to benefit from type information, and @code{special} declarations | |
3122 are redundant in a fully dynamically-scoped Lisp. A few | |
3123 declarations are meaningful when the optimizing Emacs 19 byte | |
3124 compiler is being used, however. Under the earlier non-optimizing | |
3125 compiler, these declarations will effectively be ignored. | |
3126 | |
3127 @defun proclaim decl-spec | |
3128 This function records a ``global'' declaration specified by | |
3129 @var{decl-spec}. Since @code{proclaim} is a function, @var{decl-spec} | |
3130 is evaluated and thus should normally be quoted. | |
3131 @end defun | |
3132 | |
3133 @defspec declaim decl-specs@dots{} | |
3134 This macro is like @code{proclaim}, except that it takes any number | |
3135 of @var{decl-spec} arguments, and the arguments are unevaluated and | |
3136 unquoted. The @code{declaim} macro also puts an @code{(eval-when | |
3137 (compile load eval) ...)} around the declarations so that they will | |
3138 be registered at compile-time as well as at run-time. (This is vital, | |
3139 since normally the declarations are meant to influence the way the | |
3140 compiler treats the rest of the file that contains the @code{declaim} | |
3141 form.) | |
3142 @end defspec | |
3143 | |
3144 @defspec declare decl-specs@dots{} | |
3145 This macro is used to make declarations within functions and other | |
3146 code. Common Lisp allows declarations in various locations, generally | |
3147 at the beginning of any of the many ``implicit @code{progn}s'' | |
3148 throughout Lisp syntax, such as function bodies, @code{let} bodies, | |
3149 etc. Currently the only declaration understood by @code{declare} | |
3150 is @code{special}. | |
3151 @end defspec | |
3152 | |
3153 @defspec locally declarations@dots{} forms@dots{} | |
3154 In this package, @code{locally} is no different from @code{progn}. | |
3155 @end defspec | |
3156 | |
3157 @defspec the type form | |
3158 Type information provided by @code{the} is ignored in this package; | |
3159 in other words, @code{(the @var{type} @var{form})} is equivalent | |
3160 to @var{form}. Future versions of the optimizing byte-compiler may | |
3161 make use of this information. | |
3162 | |
3163 For example, @code{mapcar} can map over both lists and arrays. It is | |
3164 hard for the compiler to expand @code{mapcar} into an in-line loop | |
3165 unless it knows whether the sequence will be a list or an array ahead | |
3166 of time. With @code{(mapcar 'car (the vector foo))}, a future | |
3167 compiler would have enough information to expand the loop in-line. | |
3168 For now, Emacs Lisp will treat the above code as exactly equivalent | |
3169 to @code{(mapcar 'car foo)}. | |
3170 @end defspec | |
3171 | |
3172 Each @var{decl-spec} in a @code{proclaim}, @code{declaim}, or | |
3173 @code{declare} should be a list beginning with a symbol that says | |
3174 what kind of declaration it is. This package currently understands | |
3175 @code{special}, @code{inline}, @code{notinline}, @code{optimize}, | |
3176 and @code{warn} declarations. (The @code{warn} declaration is an | |
3177 extension of standard Common Lisp.) Other Common Lisp declarations, | |
3178 such as @code{type} and @code{ftype}, are silently ignored. | |
3179 | |
3180 @table @code | |
3181 @item special | |
3182 Since all variables in Emacs Lisp are ``special'' (in the Common | |
3183 Lisp sense), @code{special} declarations are only advisory. They | |
3184 simply tell the optimizing byte compiler that the specified | |
3185 variables are intentionally being referred to without being | |
3186 bound in the body of the function. The compiler normally emits | |
3187 warnings for such references, since they could be typographical | |
3188 errors for references to local variables. | |
3189 | |
3190 The declaration @code{(declare (special @var{var1} @var{var2}))} is | |
3191 equivalent to @code{(defvar @var{var1}) (defvar @var{var2})} in the | |
3192 optimizing compiler, or to nothing at all in older compilers (which | |
3193 do not warn for non-local references). | |
3194 | |
3195 In top-level contexts, it is generally better to write | |
3196 @code{(defvar @var{var})} than @code{(declaim (special @var{var}))}, | |
3197 since @code{defvar} makes your intentions clearer. But the older | |
3198 byte compilers can not handle @code{defvar}s appearing inside of | |
3199 functions, while @code{(declare (special @var{var}))} takes care | |
3200 to work correctly with all compilers. | |
3201 | |
3202 @item inline | |
3203 The @code{inline} @var{decl-spec} lists one or more functions | |
3204 whose bodies should be expanded ``in-line'' into calling functions | |
3205 whenever the compiler is able to arrange for it. For example, | |
3206 the Common Lisp function @code{cadr} is declared @code{inline} | |
3207 by this package so that the form @code{(cadr @var{x})} will | |
3208 expand directly into @code{(car (cdr @var{x}))} when it is called | |
3209 in user functions, for a savings of one (relatively expensive) | |
3210 function call. | |
3211 | |
3212 The following declarations are all equivalent. Note that the | |
3213 @code{defsubst} form is a convenient way to define a function | |
3214 and declare it inline all at once, but it is available only in | |
3215 Emacs 19. | |
3216 | |
3217 @example | |
3218 (declaim (inline foo bar)) | |
3219 (eval-when (compile load eval) (proclaim '(inline foo bar))) | |
442 | 3220 (proclaim-inline foo bar) ; XEmacs only |
3221 (defsubst foo (...) ...) ; instead of defun; Emacs 19 only | |
428 | 3222 @end example |
3223 | |
3224 @strong{Please note:} This declaration remains in effect after the | |
3225 containing source file is done. It is correct to use it to | |
3226 request that a function you have defined should be inlined, | |
3227 but it is impolite to use it to request inlining of an external | |
3228 function. | |
3229 | |
3230 In Common Lisp, it is possible to use @code{(declare (inline @dots{}))} | |
3231 before a particular call to a function to cause just that call to | |
3232 be inlined; the current byte compilers provide no way to implement | |
3233 this, so @code{(declare (inline @dots{}))} is currently ignored by | |
3234 this package. | |
3235 | |
3236 @item notinline | |
3237 The @code{notinline} declaration lists functions which should | |
3238 not be inlined after all; it cancels a previous @code{inline} | |
3239 declaration. | |
3240 | |
3241 @item optimize | |
3242 This declaration controls how much optimization is performed by | |
3243 the compiler. Naturally, it is ignored by the earlier non-optimizing | |
3244 compilers. | |
3245 | |
3246 The word @code{optimize} is followed by any number of lists like | |
3247 @code{(speed 3)} or @code{(safety 2)}. Common Lisp defines several | |
3248 optimization ``qualities''; this package ignores all but @code{speed} | |
3249 and @code{safety}. The value of a quality should be an integer from | |
3250 0 to 3, with 0 meaning ``unimportant'' and 3 meaning ``very important.'' | |
3251 The default level for both qualities is 1. | |
3252 | |
3253 In this package, with the Emacs 19 optimizing compiler, the | |
3254 @code{speed} quality is tied to the @code{byte-compile-optimize} | |
3255 flag, which is set to @code{nil} for @code{(speed 0)} and to | |
3256 @code{t} for higher settings; and the @code{safety} quality is | |
3257 tied to the @code{byte-compile-delete-errors} flag, which is | |
3258 set to @code{t} for @code{(safety 3)} and to @code{nil} for all | |
3259 lower settings. (The latter flag controls whether the compiler | |
3260 is allowed to optimize out code whose only side-effect could | |
3261 be to signal an error, e.g., rewriting @code{(progn foo bar)} to | |
3262 @code{bar} when it is not known whether @code{foo} will be bound | |
3263 at run-time.) | |
3264 | |
3265 Note that even compiling with @code{(safety 0)}, the Emacs | |
3266 byte-code system provides sufficient checking to prevent real | |
3267 harm from being done. For example, barring serious bugs in | |
3268 Emacs itself, Emacs will not crash with a segmentation fault | |
3269 just because of an error in a fully-optimized Lisp program. | |
3270 | |
3271 The @code{optimize} declaration is normally used in a top-level | |
3272 @code{proclaim} or @code{declaim} in a file; Common Lisp allows | |
3273 it to be used with @code{declare} to set the level of optimization | |
3274 locally for a given form, but this will not work correctly with the | |
3275 current version of the optimizing compiler. (The @code{declare} | |
3276 will set the new optimization level, but that level will not | |
3277 automatically be unset after the enclosing form is done.) | |
3278 | |
3279 @item warn | |
3280 This declaration controls what sorts of warnings are generated | |
3281 by the byte compiler. Again, only the optimizing compiler | |
3282 generates warnings. The word @code{warn} is followed by any | |
3283 number of ``warning qualities,'' similar in form to optimization | |
3284 qualities. The currently supported warning types are | |
3285 @code{redefine}, @code{callargs}, @code{unresolved}, and | |
3286 @code{free-vars}; in the current system, a value of 0 will | |
3287 disable these warnings and any higher value will enable them. | |
3288 See the documentation for the optimizing byte compiler for details. | |
3289 @end table | |
3290 | |
3291 @node Symbols, Numbers, Declarations, Top | |
3292 @chapter Symbols | |
3293 | |
3294 @noindent | |
3295 This package defines several symbol-related features that were | |
3296 missing from Emacs Lisp. | |
3297 | |
3298 @menu | |
442 | 3299 * Property Lists:: `getf', `remf' |
428 | 3300 * Creating Symbols:: `gensym', `gentemp' |
3301 @end menu | |
3302 | |
3303 @node Property Lists, Creating Symbols, Symbols, Symbols | |
3304 @section Property Lists | |
3305 | |
3306 @noindent | |
3307 These functions augment the standard Emacs Lisp functions @code{get} | |
442 | 3308 and @code{put} for operating on properties attached to objects. |
428 | 3309 There are also functions for working with property lists as |
442 | 3310 first-class data structures not attached to particular objects. |
428 | 3311 |
3312 @defun getf place property &optional default | |
3313 This function scans the list @var{place} as if it were a property | |
3314 list, i.e., a list of alternating property names and values. If | |
3315 an even-numbered element of @var{place} is found which is @code{eq} | |
3316 to @var{property}, the following odd-numbered element is returned. | |
3317 Otherwise, @var{default} is returned (or @code{nil} if no default | |
3318 is given). | |
3319 | |
3320 In particular, | |
3321 | |
3322 @example | |
3323 (get sym prop) @equiv{} (getf (symbol-plist sym) prop) | |
3324 @end example | |
3325 | |
3326 It is legal to use @code{getf} as a @code{setf} place, in which case | |
3327 its @var{place} argument must itself be a legal @code{setf} place. | |
3328 The @var{default} argument, if any, is ignored in this context. | |
3329 The effect is to change (via @code{setcar}) the value cell in the | |
3330 list that corresponds to @var{property}, or to cons a new property-value | |
3331 pair onto the list if the property is not yet present. | |
3332 | |
3333 @example | |
3334 (put sym prop val) @equiv{} (setf (getf (symbol-plist sym) prop) val) | |
3335 @end example | |
3336 | |
440 | 3337 The @code{get} function is also @code{setf}-able. The fact that |
3338 @code{default} is ignored can sometimes be useful: | |
428 | 3339 |
3340 @example | |
440 | 3341 (incf (get 'foo 'usage-count 0)) |
428 | 3342 @end example |
3343 | |
3344 Here, symbol @code{foo}'s @code{usage-count} property is incremented | |
3345 if it exists, or set to 1 (an incremented 0) otherwise. | |
3346 | |
3347 When not used as a @code{setf} form, @code{getf} is just a regular | |
3348 function and its @var{place} argument can actually be any Lisp | |
3349 expression. | |
3350 @end defun | |
3351 | |
3352 @defspec remf place property | |
3353 This macro removes the property-value pair for @var{property} from | |
3354 the property list stored at @var{place}, which is any @code{setf}-able | |
3355 place expression. It returns true if the property was found. Note | |
3356 that if @var{property} happens to be first on the list, this will | |
3357 effectively do a @code{(setf @var{place} (cddr @var{place}))}, | |
3358 whereas if it occurs later, this simply uses @code{setcdr} to splice | |
3359 out the property and value cells. | |
3360 @end defspec | |
3361 | |
3362 @iftex | |
3363 @secno=2 | |
3364 @end iftex | |
3365 | |
3366 @node Creating Symbols, , Property Lists, Symbols | |
3367 @section Creating Symbols | |
3368 | |
3369 @noindent | |
3370 These functions create unique symbols, typically for use as | |
3371 temporary variables. | |
3372 | |
3373 @defun gensym &optional x | |
3374 This function creates a new, uninterned symbol (using @code{make-symbol}) | |
3375 with a unique name. (The name of an uninterned symbol is relevant | |
3376 only if the symbol is printed.) By default, the name is generated | |
3377 from an increasing sequence of numbers, @samp{G1000}, @samp{G1001}, | |
3378 @samp{G1002}, etc. If the optional argument @var{x} is a string, that | |
3379 string is used as a prefix instead of @samp{G}. Uninterned symbols | |
3380 are used in macro expansions for temporary variables, to ensure that | |
3381 their names will not conflict with ``real'' variables in the user's | |
3382 code. | |
3383 @end defun | |
3384 | |
3385 @defvar *gensym-counter* | |
3386 This variable holds the counter used to generate @code{gensym} names. | |
3387 It is incremented after each use by @code{gensym}. In Common Lisp | |
3388 this is initialized with 0, but this package initializes it with a | |
3389 random (time-dependent) value to avoid trouble when two files that | |
3390 each used @code{gensym} in their compilation are loaded together. | |
3391 | |
3392 @strong{XEmacs note:} As of XEmacs 21.0, an uninterned symbol remains | |
3393 uninterned even after being dumped to bytecode. Older versions of Emacs | |
3394 didn't distinguish the printed representation of interned and uninterned | |
3395 symbols, so their names had to be treated more carefully. | |
3396 @end defvar | |
3397 | |
3398 @defun gentemp &optional x | |
3399 This function is like @code{gensym}, except that it produces a new | |
3400 @emph{interned} symbol. If the symbol that is generated already | |
3401 exists, the function keeps incrementing the counter and trying | |
3402 again until a new symbol is generated. | |
3403 @end defun | |
3404 | |
3405 The Quiroz @file{cl.el} package also defined a @code{defkeyword} | |
3406 form for creating self-quoting keyword symbols. This package | |
3407 automatically creates all keywords that are called for by | |
3408 @code{&key} argument specifiers, and discourages the use of | |
3409 keywords as data unrelated to keyword arguments, so the | |
3410 @code{defkeyword} form has been discontinued. | |
3411 | |
3412 @iftex | |
3413 @chapno=11 | |
3414 @end iftex | |
3415 | |
3416 @node Numbers, Sequences, Symbols, Top | |
3417 @chapter Numbers | |
3418 | |
3419 @noindent | |
3420 This section defines a few simple Common Lisp operations on numbers | |
3421 which were left out of Emacs Lisp. | |
3422 | |
3423 @menu | |
3424 * Predicates on Numbers:: `plusp', `oddp', `floatp-safe', etc. | |
3425 * Numerical Functions:: `abs', `expt', `floor*', etc. | |
3426 * Random Numbers:: `random*', `make-random-state' | |
3427 * Implementation Parameters:: `most-positive-fixnum', `most-positive-float' | |
3428 @end menu | |
3429 | |
3430 @iftex | |
3431 @secno=1 | |
3432 @end iftex | |
3433 | |
3434 @node Predicates on Numbers, Numerical Functions, Numbers, Numbers | |
3435 @section Predicates on Numbers | |
3436 | |
3437 @noindent | |
3438 These functions return @code{t} if the specified condition is | |
3439 true of the numerical argument, or @code{nil} otherwise. | |
3440 | |
3441 @defun plusp number | |
3442 This predicate tests whether @var{number} is positive. It is an | |
3443 error if the argument is not a number. | |
3444 @end defun | |
3445 | |
3446 @defun minusp number | |
3447 This predicate tests whether @var{number} is negative. It is an | |
3448 error if the argument is not a number. | |
3449 @end defun | |
3450 | |
3451 @defun oddp integer | |
3452 This predicate tests whether @var{integer} is odd. It is an | |
3453 error if the argument is not an integer. | |
3454 @end defun | |
3455 | |
3456 @defun evenp integer | |
3457 This predicate tests whether @var{integer} is even. It is an | |
3458 error if the argument is not an integer. | |
3459 @end defun | |
3460 | |
3461 @defun floatp-safe object | |
3462 This predicate tests whether @var{object} is a floating-point | |
3463 number. On systems that support floating-point, this is equivalent | |
3464 to @code{floatp}. On other systems, this always returns @code{nil}. | |
3465 @end defun | |
3466 | |
3467 @iftex | |
3468 @secno=3 | |
3469 @end iftex | |
3470 | |
3471 @node Numerical Functions, Random Numbers, Predicates on Numbers, Numbers | |
3472 @section Numerical Functions | |
3473 | |
3474 @noindent | |
3475 These functions perform various arithmetic operations on numbers. | |
3476 | |
3477 @defun abs number | |
3478 This function returns the absolute value of @var{number}. (Newer | |
3479 versions of Emacs provide this as a built-in function; this package | |
3480 defines @code{abs} only for Emacs 18 versions which don't provide | |
3481 it as a primitive.) | |
3482 @end defun | |
3483 | |
3484 @defun expt base power | |
3485 This function returns @var{base} raised to the power of @var{number}. | |
3486 (Newer versions of Emacs provide this as a built-in function; this | |
3487 package defines @code{expt} only for Emacs 18 versions which don't | |
3488 provide it as a primitive.) | |
3489 @end defun | |
3490 | |
3491 @defun gcd &rest integers | |
3492 This function returns the Greatest Common Divisor of the arguments. | |
3493 For one argument, it returns the absolute value of that argument. | |
3494 For zero arguments, it returns zero. | |
3495 @end defun | |
3496 | |
3497 @defun lcm &rest integers | |
3498 This function returns the Least Common Multiple of the arguments. | |
3499 For one argument, it returns the absolute value of that argument. | |
3500 For zero arguments, it returns one. | |
3501 @end defun | |
3502 | |
3503 @defun isqrt integer | |
3504 This function computes the ``integer square root'' of its integer | |
3505 argument, i.e., the greatest integer less than or equal to the true | |
3506 square root of the argument. | |
3507 @end defun | |
3508 | |
3509 @defun floor* number &optional divisor | |
3510 This function implements the Common Lisp @code{floor} function. | |
3511 It is called @code{floor*} to avoid name conflicts with the | |
3512 simpler @code{floor} function built-in to Emacs 19. | |
3513 | |
3514 With one argument, @code{floor*} returns a list of two numbers: | |
3515 The argument rounded down (toward minus infinity) to an integer, | |
3516 and the ``remainder'' which would have to be added back to the | |
3517 first return value to yield the argument again. If the argument | |
3518 is an integer @var{x}, the result is always the list @code{(@var{x} 0)}. | |
3519 If the argument is an Emacs 19 floating-point number, the first | |
3520 result is a Lisp integer and the second is a Lisp float between | |
3521 0 (inclusive) and 1 (exclusive). | |
3522 | |
3523 With two arguments, @code{floor*} divides @var{number} by | |
3524 @var{divisor}, and returns the floor of the quotient and the | |
3525 corresponding remainder as a list of two numbers. If | |
3526 @code{(floor* @var{x} @var{y})} returns @code{(@var{q} @var{r})}, | |
3527 then @code{@var{q}*@var{y} + @var{r} = @var{x}}, with @var{r} | |
3528 between 0 (inclusive) and @var{r} (exclusive). Also, note | |
3529 that @code{(floor* @var{x})} is exactly equivalent to | |
3530 @code{(floor* @var{x} 1)}. | |
3531 | |
3532 This function is entirely compatible with Common Lisp's @code{floor} | |
3533 function, except that it returns the two results in a list since | |
3534 Emacs Lisp does not support multiple-valued functions. | |
3535 @end defun | |
3536 | |
3537 @defun ceiling* number &optional divisor | |
3538 This function implements the Common Lisp @code{ceiling} function, | |
3539 which is analogous to @code{floor} except that it rounds the | |
3540 argument or quotient of the arguments up toward plus infinity. | |
3541 The remainder will be between 0 and minus @var{r}. | |
3542 @end defun | |
3543 | |
3544 @defun truncate* number &optional divisor | |
3545 This function implements the Common Lisp @code{truncate} function, | |
3546 which is analogous to @code{floor} except that it rounds the | |
3547 argument or quotient of the arguments toward zero. Thus it is | |
3548 equivalent to @code{floor*} if the argument or quotient is | |
3549 positive, or to @code{ceiling*} otherwise. The remainder has | |
3550 the same sign as @var{number}. | |
3551 @end defun | |
3552 | |
3553 @defun round* number &optional divisor | |
3554 This function implements the Common Lisp @code{round} function, | |
3555 which is analogous to @code{floor} except that it rounds the | |
3556 argument or quotient of the arguments to the nearest integer. | |
3557 In the case of a tie (the argument or quotient is exactly | |
3558 halfway between two integers), it rounds to the even integer. | |
3559 @end defun | |
3560 | |
3561 @defun mod* number divisor | |
3562 This function returns the same value as the second return value | |
3563 of @code{floor}. | |
3564 @end defun | |
3565 | |
3566 @defun rem* number divisor | |
3567 This function returns the same value as the second return value | |
3568 of @code{truncate}. | |
3569 @end defun | |
3570 | |
3571 These definitions are compatible with those in the Quiroz | |
3572 @file{cl.el} package, except that this package appends @samp{*} | |
3573 to certain function names to avoid conflicts with existing | |
3574 Emacs 19 functions, and that the mechanism for returning | |
3575 multiple values is different. | |
3576 | |
3577 @iftex | |
3578 @secno=8 | |
3579 @end iftex | |
3580 | |
3581 @node Random Numbers, Implementation Parameters, Numerical Functions, Numbers | |
3582 @section Random Numbers | |
3583 | |
3584 @noindent | |
3585 This package also provides an implementation of the Common Lisp | |
3586 random number generator. It uses its own additive-congruential | |
3587 algorithm, which is much more likely to give statistically clean | |
3588 random numbers than the simple generators supplied by many | |
3589 operating systems. | |
3590 | |
3591 @defun random* number &optional state | |
3592 This function returns a random nonnegative number less than | |
3593 @var{number}, and of the same type (either integer or floating-point). | |
3594 The @var{state} argument should be a @code{random-state} object | |
3595 which holds the state of the random number generator. The | |
3596 function modifies this state object as a side effect. If | |
3597 @var{state} is omitted, it defaults to the variable | |
3598 @code{*random-state*}, which contains a pre-initialized | |
3599 @code{random-state} object. | |
3600 @end defun | |
3601 | |
3602 @defvar *random-state* | |
3603 This variable contains the system ``default'' @code{random-state} | |
3604 object, used for calls to @code{random*} that do not specify an | |
3605 alternative state object. Since any number of programs in the | |
3606 Emacs process may be accessing @code{*random-state*} in interleaved | |
3607 fashion, the sequence generated from this variable will be | |
3608 irreproducible for all intents and purposes. | |
3609 @end defvar | |
3610 | |
3611 @defun make-random-state &optional state | |
3612 This function creates or copies a @code{random-state} object. | |
3613 If @var{state} is omitted or @code{nil}, it returns a new copy of | |
3614 @code{*random-state*}. This is a copy in the sense that future | |
3615 sequences of calls to @code{(random* @var{n})} and | |
3616 @code{(random* @var{n} @var{s})} (where @var{s} is the new | |
3617 random-state object) will return identical sequences of random | |
3618 numbers. | |
3619 | |
3620 If @var{state} is a @code{random-state} object, this function | |
3621 returns a copy of that object. If @var{state} is @code{t}, this | |
3622 function returns a new @code{random-state} object seeded from the | |
3623 date and time. As an extension to Common Lisp, @var{state} may also | |
3624 be an integer in which case the new object is seeded from that | |
3625 integer; each different integer seed will result in a completely | |
3626 different sequence of random numbers. | |
3627 | |
3628 It is legal to print a @code{random-state} object to a buffer or | |
3629 file and later read it back with @code{read}. If a program wishes | |
3630 to use a sequence of pseudo-random numbers which can be reproduced | |
3631 later for debugging, it can call @code{(make-random-state t)} to | |
3632 get a new sequence, then print this sequence to a file. When the | |
3633 program is later rerun, it can read the original run's random-state | |
3634 from the file. | |
3635 @end defun | |
3636 | |
3637 @defun random-state-p object | |
3638 This predicate returns @code{t} if @var{object} is a | |
3639 @code{random-state} object, or @code{nil} otherwise. | |
3640 @end defun | |
3641 | |
3642 @node Implementation Parameters, , Random Numbers, Numbers | |
3643 @section Implementation Parameters | |
3644 | |
3645 @noindent | |
3646 This package defines several useful constants having to with numbers. | |
3647 | |
3648 @defvar most-positive-fixnum | |
3649 This constant equals the largest value a Lisp integer can hold. | |
3650 It is typically @code{2^23-1} or @code{2^25-1}. | |
3651 @end defvar | |
3652 | |
3653 @defvar most-negative-fixnum | |
3654 This constant equals the smallest (most negative) value a Lisp | |
3655 integer can hold. | |
3656 @end defvar | |
3657 | |
3658 The following parameters have to do with floating-point numbers. | |
3659 This package determines their values by exercising the computer's | |
3660 floating-point arithmetic in various ways. Because this operation | |
3661 might be slow, the code for initializing them is kept in a separate | |
3662 function that must be called before the parameters can be used. | |
3663 | |
3664 @defun cl-float-limits | |
3665 This function makes sure that the Common Lisp floating-point | |
3666 parameters like @code{most-positive-float} have been initialized. | |
3667 Until it is called, these parameters will be @code{nil}. If this | |
3668 version of Emacs does not support floats (e.g., most versions of | |
3669 Emacs 18), the parameters will remain @code{nil}. If the parameters | |
3670 have already been initialized, the function returns immediately. | |
3671 | |
3672 The algorithm makes assumptions that will be valid for most modern | |
3673 machines, but will fail if the machine's arithmetic is extremely | |
3674 unusual, e.g., decimal. | |
3675 @end defun | |
3676 | |
3677 Since true Common Lisp supports up to four different floating-point | |
3678 precisions, it has families of constants like | |
3679 @code{most-positive-single-float}, @code{most-positive-double-float}, | |
3680 @code{most-positive-long-float}, and so on. Emacs has only one | |
3681 floating-point precision, so this package omits the precision word | |
3682 from the constants' names. | |
3683 | |
3684 @defvar most-positive-float | |
3685 This constant equals the largest value a Lisp float can hold. | |
3686 For those systems whose arithmetic supports infinities, this is | |
3687 the largest @emph{finite} value. For IEEE machines, the value | |
3688 is approximately @code{1.79e+308}. | |
3689 @end defvar | |
3690 | |
3691 @defvar most-negative-float | |
3692 This constant equals the most-negative value a Lisp float can hold. | |
3693 (It is assumed to be equal to @code{(- most-positive-float)}.) | |
3694 @end defvar | |
3695 | |
3696 @defvar least-positive-float | |
3697 This constant equals the smallest Lisp float value greater than zero. | |
3698 For IEEE machines, it is about @code{4.94e-324} if denormals are | |
3699 supported or @code{2.22e-308} if not. | |
3700 @end defvar | |
3701 | |
3702 @defvar least-positive-normalized-float | |
3703 This constant equals the smallest @emph{normalized} Lisp float greater | |
3704 than zero, i.e., the smallest value for which IEEE denormalization | |
3705 will not result in a loss of precision. For IEEE machines, this | |
3706 value is about @code{2.22e-308}. For machines that do not support | |
3707 the concept of denormalization and gradual underflow, this constant | |
3708 will always equal @code{least-positive-float}. | |
3709 @end defvar | |
3710 | |
3711 @defvar least-negative-float | |
3712 This constant is the negative counterpart of @code{least-positive-float}. | |
3713 @end defvar | |
3714 | |
3715 @defvar least-negative-normalized-float | |
3716 This constant is the negative counterpart of | |
3717 @code{least-positive-normalized-float}. | |
3718 @end defvar | |
3719 | |
3720 @defvar float-epsilon | |
3721 This constant is the smallest positive Lisp float that can be added | |
3722 to 1.0 to produce a distinct value. Adding a smaller number to 1.0 | |
3723 will yield 1.0 again due to roundoff. For IEEE machines, epsilon | |
3724 is about @code{2.22e-16}. | |
3725 @end defvar | |
3726 | |
3727 @defvar float-negative-epsilon | |
3728 This is the smallest positive value that can be subtracted from | |
3729 1.0 to produce a distinct value. For IEEE machines, it is about | |
3730 @code{1.11e-16}. | |
3731 @end defvar | |
3732 | |
3733 @iftex | |
3734 @chapno=13 | |
3735 @end iftex | |
3736 | |
3737 @node Sequences, Lists, Numbers, Top | |
3738 @chapter Sequences | |
3739 | |
3740 @noindent | |
3741 Common Lisp defines a number of functions that operate on | |
3742 @dfn{sequences}, which are either lists, strings, or vectors. | |
3743 Emacs Lisp includes a few of these, notably @code{elt} and | |
3744 @code{length}; this package defines most of the rest. | |
3745 | |
3746 @menu | |
3747 * Sequence Basics:: Arguments shared by all sequence functions | |
3748 * Mapping over Sequences:: `mapcar*', `mapcan', `map', `every', etc. | |
3749 * Sequence Functions:: `subseq', `remove*', `substitute', etc. | |
3750 * Searching Sequences:: `find', `position', `count', `search', etc. | |
3751 * Sorting Sequences:: `sort*', `stable-sort', `merge' | |
3752 @end menu | |
3753 | |
3754 @node Sequence Basics, Mapping over Sequences, Sequences, Sequences | |
3755 @section Sequence Basics | |
3756 | |
778 | 3757 @noindent Many of the sequence functions take keyword arguments; |
3758 @pxref{Argument Lists}. All keyword arguments are optional and, if | |
3759 specified, may appear in any order. | |
428 | 3760 |
3761 The @code{:key} argument should be passed either @code{nil}, or a | |
3762 function of one argument. This key function is used as a filter | |
3763 through which the elements of the sequence are seen; for example, | |
3764 @code{(find x y :key 'car)} is similar to @code{(assoc* x y)}: | |
3765 It searches for an element of the list whose @code{car} equals | |
3766 @code{x}, rather than for an element which equals @code{x} itself. | |
3767 If @code{:key} is omitted or @code{nil}, the filter is effectively | |
3768 the identity function. | |
3769 | |
3770 The @code{:test} and @code{:test-not} arguments should be either | |
3771 @code{nil}, or functions of two arguments. The test function is | |
3772 used to compare two sequence elements, or to compare a search value | |
3773 with sequence elements. (The two values are passed to the test | |
3774 function in the same order as the original sequence function | |
3775 arguments from which they are derived, or, if they both come from | |
3776 the same sequence, in the same order as they appear in that sequence.) | |
3777 The @code{:test} argument specifies a function which must return | |
3778 true (non-@code{nil}) to indicate a match; instead, you may use | |
3779 @code{:test-not} to give a function which returns @emph{false} to | |
3780 indicate a match. The default test function is @code{:test 'eql}. | |
3781 | |
3782 Many functions which take @var{item} and @code{:test} or @code{:test-not} | |
3783 arguments also come in @code{-if} and @code{-if-not} varieties, | |
3784 where a @var{predicate} function is passed instead of @var{item}, | |
3785 and sequence elements match if the predicate returns true on them | |
3786 (or false in the case of @code{-if-not}). For example: | |
3787 | |
3788 @example | |
3789 (remove* 0 seq :test '=) @equiv{} (remove-if 'zerop seq) | |
3790 @end example | |
3791 | |
3792 @noindent | |
3793 to remove all zeros from sequence @code{seq}. | |
3794 | |
3795 Some operations can work on a subsequence of the argument sequence; | |
3796 these function take @code{:start} and @code{:end} arguments which | |
3797 default to zero and the length of the sequence, respectively. | |
3798 Only elements between @var{start} (inclusive) and @var{end} | |
3799 (exclusive) are affected by the operation. The @var{end} argument | |
3800 may be passed @code{nil} to signify the length of the sequence; | |
3801 otherwise, both @var{start} and @var{end} must be integers, with | |
3802 @code{0 <= @var{start} <= @var{end} <= (length @var{seq})}. | |
3803 If the function takes two sequence arguments, the limits are | |
3804 defined by keywords @code{:start1} and @code{:end1} for the first, | |
3805 and @code{:start2} and @code{:end2} for the second. | |
3806 | |
3807 A few functions accept a @code{:from-end} argument, which, if | |
3808 non-@code{nil}, causes the operation to go from right-to-left | |
3809 through the sequence instead of left-to-right, and a @code{:count} | |
3810 argument, which specifies an integer maximum number of elements | |
3811 to be removed or otherwise processed. | |
3812 | |
3813 The sequence functions make no guarantees about the order in | |
3814 which the @code{:test}, @code{:test-not}, and @code{:key} functions | |
3815 are called on various elements. Therefore, it is a bad idea to depend | |
3816 on side effects of these functions. For example, @code{:from-end} | |
3817 may cause the sequence to be scanned actually in reverse, or it may | |
3818 be scanned forwards but computing a result ``as if'' it were scanned | |
3819 backwards. (Some functions, like @code{mapcar*} and @code{every}, | |
3820 @emph{do} specify exactly the order in which the function is called | |
3821 so side effects are perfectly acceptable in those cases.) | |
3822 | |
3823 Strings in GNU Emacs 19 may contain ``text properties'' as well | |
3824 as character data. Except as noted, it is undefined whether or | |
3825 not text properties are preserved by sequence functions. For | |
3826 example, @code{(remove* ?A @var{str})} may or may not preserve | |
3827 the properties of the characters copied from @var{str} into the | |
3828 result. | |
3829 | |
3830 @node Mapping over Sequences, Sequence Functions, Sequence Basics, Sequences | |
3831 @section Mapping over Sequences | |
3832 | |
3833 @noindent | |
3834 These functions ``map'' the function you specify over the elements | |
3835 of lists or arrays. They are all variations on the theme of the | |
3836 built-in function @code{mapcar}. | |
3837 | |
3838 @defun mapcar* function seq &rest more-seqs | |
3839 This function calls @var{function} on successive parallel sets of | |
3840 elements from its argument sequences. Given a single @var{seq} | |
3841 argument it is equivalent to @code{mapcar}; given @var{n} sequences, | |
3842 it calls the function with the first elements of each of the sequences | |
3843 as the @var{n} arguments to yield the first element of the result | |
3844 list, then with the second elements, and so on. The mapping stops as | |
3845 soon as the shortest sequence runs out. The argument sequences may | |
3846 be any mixture of lists, strings, and vectors; the return sequence | |
3847 is always a list. | |
3848 | |
3849 Common Lisp's @code{mapcar} accepts multiple arguments but works | |
3850 only on lists; Emacs Lisp's @code{mapcar} accepts a single sequence | |
3851 argument. This package's @code{mapcar*} works as a compatible | |
3852 superset of both. | |
3853 @end defun | |
3854 | |
3855 @defun map result-type function seq &rest more-seqs | |
3856 This function maps @var{function} over the argument sequences, | |
3857 just like @code{mapcar*}, but it returns a sequence of type | |
3858 @var{result-type} rather than a list. @var{result-type} must | |
3859 be one of the following symbols: @code{vector}, @code{string}, | |
3860 @code{list} (in which case the effect is the same as for | |
3861 @code{mapcar*}), or @code{nil} (in which case the results are | |
3862 thrown away and @code{map} returns @code{nil}). | |
3863 @end defun | |
3864 | |
3865 @defun maplist function list &rest more-lists | |
3866 This function calls @var{function} on each of its argument lists, | |
3867 then on the @code{cdr}s of those lists, and so on, until the | |
3868 shortest list runs out. The results are returned in the form | |
3869 of a list. Thus, @code{maplist} is like @code{mapcar*} except | |
3870 that it passes in the list pointers themselves rather than the | |
3871 @code{car}s of the advancing pointers. | |
3872 @end defun | |
3873 | |
3874 @defun mapc function seq &rest more-seqs | |
3875 This function is like @code{mapcar*}, except that the values | |
3876 returned by @var{function} are ignored and thrown away rather | |
3877 than being collected into a list. The return value of @code{mapc} | |
3878 is @var{seq}, the first sequence. | |
3879 @end defun | |
3880 | |
3881 @defun mapl function list &rest more-lists | |
3882 This function is like @code{maplist}, except that it throws away | |
3883 the values returned by @var{function}. | |
3884 @end defun | |
3885 | |
3886 @defun mapcan function seq &rest more-seqs | |
3887 This function is like @code{mapcar*}, except that it concatenates | |
3888 the return values (which must be lists) using @code{nconc}, | |
3889 rather than simply collecting them into a list. | |
3890 @end defun | |
3891 | |
3892 @defun mapcon function list &rest more-lists | |
3893 This function is like @code{maplist}, except that it concatenates | |
3894 the return values using @code{nconc}. | |
3895 @end defun | |
3896 | |
3897 @defun some predicate seq &rest more-seqs | |
3898 This function calls @var{predicate} on each element of @var{seq} | |
3899 in turn; if @var{predicate} returns a non-@code{nil} value, | |
3900 @code{some} returns that value, otherwise it returns @code{nil}. | |
3901 Given several sequence arguments, it steps through the sequences | |
3902 in parallel until the shortest one runs out, just as in | |
3903 @code{mapcar*}. You can rely on the left-to-right order in which | |
3904 the elements are visited, and on the fact that mapping stops | |
3905 immediately as soon as @var{predicate} returns non-@code{nil}. | |
3906 @end defun | |
3907 | |
3908 @defun every predicate seq &rest more-seqs | |
3909 This function calls @var{predicate} on each element of the sequence(s) | |
3910 in turn; it returns @code{nil} as soon as @var{predicate} returns | |
3911 @code{nil} for any element, or @code{t} if the predicate was true | |
3912 for all elements. | |
3913 @end defun | |
3914 | |
3915 @defun notany predicate seq &rest more-seqs | |
3916 This function calls @var{predicate} on each element of the sequence(s) | |
3917 in turn; it returns @code{nil} as soon as @var{predicate} returns | |
3918 a non-@code{nil} value for any element, or @code{t} if the predicate | |
3919 was @code{nil} for all elements. | |
3920 @end defun | |
3921 | |
3922 @defun notevery predicate seq &rest more-seqs | |
3923 This function calls @var{predicate} on each element of the sequence(s) | |
3924 in turn; it returns a non-@code{nil} value as soon as @var{predicate} | |
3925 returns @code{nil} for any element, or @code{t} if the predicate was | |
3926 true for all elements. | |
3927 @end defun | |
3928 | |
3929 @defun reduce function seq @t{&key :from-end :start :end :initial-value :key} | |
3930 This function combines the elements of @var{seq} using an associative | |
3931 binary operation. Suppose @var{function} is @code{*} and @var{seq} is | |
3932 the list @code{(2 3 4 5)}. The first two elements of the list are | |
3933 combined with @code{(* 2 3) = 6}; this is combined with the next | |
3934 element, @code{(* 6 4) = 24}, and that is combined with the final | |
3935 element: @code{(* 24 5) = 120}. Note that the @code{*} function happens | |
3936 to be self-reducing, so that @code{(* 2 3 4 5)} has the same effect as | |
3937 an explicit call to @code{reduce}. | |
3938 | |
3939 If @code{:from-end} is true, the reduction is right-associative instead | |
3940 of left-associative: | |
3941 | |
3942 @example | |
3943 (reduce '- '(1 2 3 4)) | |
3944 @equiv{} (- (- (- 1 2) 3) 4) @result{} -8 | |
3945 (reduce '- '(1 2 3 4) :from-end t) | |
3946 @equiv{} (- 1 (- 2 (- 3 4))) @result{} -2 | |
3947 @end example | |
3948 | |
3949 If @code{:key} is specified, it is a function of one argument which | |
3950 is called on each of the sequence elements in turn. | |
3951 | |
3952 If @code{:initial-value} is specified, it is effectively added to the | |
3953 front (or rear in the case of @code{:from-end}) of the sequence. | |
3954 The @code{:key} function is @emph{not} applied to the initial value. | |
3955 | |
3956 If the sequence, including the initial value, has exactly one element | |
3957 then that element is returned without ever calling @var{function}. | |
3958 If the sequence is empty (and there is no initial value), then | |
3959 @var{function} is called with no arguments to obtain the return value. | |
3960 @end defun | |
3961 | |
3962 All of these mapping operations can be expressed conveniently in | |
3963 terms of the @code{loop} macro. In compiled code, @code{loop} will | |
3964 be faster since it generates the loop as in-line code with no | |
3965 function calls. | |
3966 | |
3967 @node Sequence Functions, Searching Sequences, Mapping over Sequences, Sequences | |
3968 @section Sequence Functions | |
3969 | |
3970 @noindent | |
3971 This section describes a number of Common Lisp functions for | |
3972 operating on sequences. | |
3973 | |
3974 @defun subseq sequence start &optional end | |
3975 This function returns a given subsequence of the argument | |
3976 @var{sequence}, which may be a list, string, or vector. | |
3977 The indices @var{start} and @var{end} must be in range, and | |
3978 @var{start} must be no greater than @var{end}. If @var{end} | |
3979 is omitted, it defaults to the length of the sequence. The | |
3980 return value is always a copy; it does not share structure | |
3981 with @var{sequence}. | |
3982 | |
3983 As an extension to Common Lisp, @var{start} and/or @var{end} | |
3984 may be negative, in which case they represent a distance back | |
3985 from the end of the sequence. This is for compatibility with | |
3986 Emacs' @code{substring} function. Note that @code{subseq} is | |
3987 the @emph{only} sequence function that allows negative | |
3988 @var{start} and @var{end}. | |
3989 | |
3990 You can use @code{setf} on a @code{subseq} form to replace a | |
3991 specified range of elements with elements from another sequence. | |
3992 The replacement is done as if by @code{replace}, described below. | |
3993 @end defun | |
3994 | |
3995 @defun concatenate result-type &rest seqs | |
3996 This function concatenates the argument sequences together to | |
3997 form a result sequence of type @var{result-type}, one of the | |
3998 symbols @code{vector}, @code{string}, or @code{list}. The | |
3999 arguments are always copied, even in cases such as | |
4000 @code{(concatenate 'list '(1 2 3))} where the result is | |
4001 identical to an argument. | |
4002 @end defun | |
4003 | |
4004 @defun fill seq item @t{&key :start :end} | |
4005 This function fills the elements of the sequence (or the specified | |
4006 part of the sequence) with the value @var{item}. | |
4007 @end defun | |
4008 | |
4009 @defun replace seq1 seq2 @t{&key :start1 :end1 :start2 :end2} | |
4010 This function copies part of @var{seq2} into part of @var{seq1}. | |
4011 The sequence @var{seq1} is not stretched or resized; the amount | |
4012 of data copied is simply the shorter of the source and destination | |
4013 (sub)sequences. The function returns @var{seq1}. | |
4014 | |
4015 If @var{seq1} and @var{seq2} are @code{eq}, then the replacement | |
4016 will work correctly even if the regions indicated by the start | |
4017 and end arguments overlap. However, if @var{seq1} and @var{seq2} | |
4018 are lists which share storage but are not @code{eq}, and the | |
4019 start and end arguments specify overlapping regions, the effect | |
4020 is undefined. | |
4021 @end defun | |
4022 | |
4023 @defun remove* item seq @t{&key :test :test-not :key :count :start :end :from-end} | |
4024 This returns a copy of @var{seq} with all elements matching | |
4025 @var{item} removed. The result may share storage with or be | |
4026 @code{eq} to @var{seq} in some circumstances, but the original | |
4027 @var{seq} will not be modified. The @code{:test}, @code{:test-not}, | |
4028 and @code{:key} arguments define the matching test that is used; | |
4029 by default, elements @code{eql} to @var{item} are removed. The | |
4030 @code{:count} argument specifies the maximum number of matching | |
4031 elements that can be removed (only the leftmost @var{count} matches | |
4032 are removed). The @code{:start} and @code{:end} arguments specify | |
4033 a region in @var{seq} in which elements will be removed; elements | |
4034 outside that region are not matched or removed. The @code{:from-end} | |
4035 argument, if true, says that elements should be deleted from the | |
4036 end of the sequence rather than the beginning (this matters only | |
4037 if @var{count} was also specified). | |
4038 @end defun | |
4039 | |
4040 @defun delete* item seq @t{&key :test :test-not :key :count :start :end :from-end} | |
4041 This deletes all elements of @var{seq} which match @var{item}. | |
4042 It is a destructive operation. Since Emacs Lisp does not support | |
4043 stretchable strings or vectors, this is the same as @code{remove*} | |
4044 for those sequence types. On lists, @code{remove*} will copy the | |
4045 list if necessary to preserve the original list, whereas | |
4046 @code{delete*} will splice out parts of the argument list. | |
4047 Compare @code{append} and @code{nconc}, which are analogous | |
4048 non-destructive and destructive list operations in Emacs Lisp. | |
4049 @end defun | |
4050 | |
4051 @findex remove-if | |
4052 @findex remove-if-not | |
4053 @findex delete-if | |
4054 @findex delete-if-not | |
4055 The predicate-oriented functions @code{remove-if}, @code{remove-if-not}, | |
4056 @code{delete-if}, and @code{delete-if-not} are defined similarly. | |
4057 | |
4058 @defun delete item list | |
4059 This MacLisp-compatible function deletes from @var{list} all elements | |
4060 which are @code{equal} to @var{item}. The @code{delete} function is | |
4061 built-in to Emacs 19; this package defines it equivalently in Emacs 18. | |
4062 @end defun | |
4063 | |
4064 @defun remove item list | |
4065 This function removes from @var{list} all elements which are | |
4066 @code{equal} to @var{item}. This package defines it for symmetry | |
4067 with @code{delete}, even though @code{remove} is not built-in to | |
4068 Emacs 19. | |
4069 @end defun | |
4070 | |
4071 @defun remq item list | |
4072 This function removes from @var{list} all elements which are | |
4073 @code{eq} to @var{item}. This package defines it for symmetry | |
4074 with @code{delq}, even though @code{remq} is not built-in to | |
4075 Emacs 19. | |
4076 @end defun | |
4077 | |
4078 @defun remove-duplicates seq @t{&key :test :test-not :key :start :end :from-end} | |
4079 This function returns a copy of @var{seq} with duplicate elements | |
4080 removed. Specifically, if two elements from the sequence match | |
4081 according to the @code{:test}, @code{:test-not}, and @code{:key} | |
4082 arguments, only the rightmost one is retained. If @code{:from-end} | |
4083 is true, the leftmost one is retained instead. If @code{:start} or | |
4084 @code{:end} is specified, only elements within that subsequence are | |
4085 examined or removed. | |
4086 @end defun | |
4087 | |
4088 @defun delete-duplicates seq @t{&key :test :test-not :key :start :end :from-end} | |
4089 This function deletes duplicate elements from @var{seq}. It is | |
4090 a destructive version of @code{remove-duplicates}. | |
4091 @end defun | |
4092 | |
4093 @defun substitute new old seq @t{&key :test :test-not :key :count :start :end :from-end} | |
4094 This function returns a copy of @var{seq}, with all elements | |
4095 matching @var{old} replaced with @var{new}. The @code{:count}, | |
4096 @code{:start}, @code{:end}, and @code{:from-end} arguments may be | |
4097 used to limit the number of substitutions made. | |
4098 @end defun | |
4099 | |
4100 @defun nsubstitute new old seq @t{&key :test :test-not :key :count :start :end :from-end} | |
4101 This is a destructive version of @code{substitute}; it performs | |
4102 the substitution using @code{setcar} or @code{aset} rather than | |
4103 by returning a changed copy of the sequence. | |
4104 @end defun | |
4105 | |
4106 @findex substitute-if | |
4107 @findex substitute-if-not | |
4108 @findex nsubstitute-if | |
4109 @findex nsubstitute-if-not | |
4110 The @code{substitute-if}, @code{substitute-if-not}, @code{nsubstitute-if}, | |
4111 and @code{nsubstitute-if-not} functions are defined similarly. For | |
4112 these, a @var{predicate} is given in place of the @var{old} argument. | |
4113 | |
4114 @node Searching Sequences, Sorting Sequences, Sequence Functions, Sequences | |
4115 @section Searching Sequences | |
4116 | |
4117 @noindent | |
4118 These functions search for elements or subsequences in a sequence. | |
4119 (See also @code{member*} and @code{assoc*}; @pxref{Lists}.) | |
4120 | |
4121 @defun find item seq @t{&key :test :test-not :key :start :end :from-end} | |
4122 This function searches @var{seq} for an element matching @var{item}. | |
4123 If it finds a match, it returns the matching element. Otherwise, | |
4124 it returns @code{nil}. It returns the leftmost match, unless | |
4125 @code{:from-end} is true, in which case it returns the rightmost | |
4126 match. The @code{:start} and @code{:end} arguments may be used to | |
4127 limit the range of elements that are searched. | |
4128 @end defun | |
4129 | |
4130 @defun position item seq @t{&key :test :test-not :key :start :end :from-end} | |
4131 This function is like @code{find}, except that it returns the | |
4132 integer position in the sequence of the matching item rather than | |
4133 the item itself. The position is relative to the start of the | |
4134 sequence as a whole, even if @code{:start} is non-zero. The function | |
4135 returns @code{nil} if no matching element was found. | |
4136 @end defun | |
4137 | |
4138 @defun count item seq @t{&key :test :test-not :key :start :end} | |
4139 This function returns the number of elements of @var{seq} which | |
4140 match @var{item}. The result is always a nonnegative integer. | |
4141 @end defun | |
4142 | |
4143 @findex find-if | |
4144 @findex find-if-not | |
4145 @findex position-if | |
4146 @findex position-if-not | |
4147 @findex count-if | |
4148 @findex count-if-not | |
4149 The @code{find-if}, @code{find-if-not}, @code{position-if}, | |
4150 @code{position-if-not}, @code{count-if}, and @code{count-if-not} | |
4151 functions are defined similarly. | |
4152 | |
4153 @defun mismatch seq1 seq2 @t{&key :test :test-not :key :start1 :end1 :start2 :end2 :from-end} | |
4154 This function compares the specified parts of @var{seq1} and | |
4155 @var{seq2}. If they are the same length and the corresponding | |
4156 elements match (according to @code{:test}, @code{:test-not}, | |
4157 and @code{:key}), the function returns @code{nil}. If there is | |
4158 a mismatch, the function returns the index (relative to @var{seq1}) | |
4159 of the first mismatching element. This will be the leftmost pair of | |
4160 elements which do not match, or the position at which the shorter of | |
4161 the two otherwise-matching sequences runs out. | |
4162 | |
4163 If @code{:from-end} is true, then the elements are compared from right | |
4164 to left starting at @code{(1- @var{end1})} and @code{(1- @var{end2})}. | |
4165 If the sequences differ, then one plus the index of the rightmost | |
4166 difference (relative to @var{seq1}) is returned. | |
4167 | |
4168 An interesting example is @code{(mismatch str1 str2 :key 'upcase)}, | |
4169 which compares two strings case-insensitively. | |
4170 @end defun | |
4171 | |
4172 @defun search seq1 seq2 @t{&key :test :test-not :key :from-end :start1 :end1 :start2 :end2} | |
4173 This function searches @var{seq2} for a subsequence that matches | |
4174 @var{seq1} (or part of it specified by @code{:start1} and | |
4175 @code{:end1}.) Only matches which fall entirely within the region | |
4176 defined by @code{:start2} and @code{:end2} will be considered. | |
4177 The return value is the index of the leftmost element of the | |
4178 leftmost match, relative to the start of @var{seq2}, or @code{nil} | |
4179 if no matches were found. If @code{:from-end} is true, the | |
4180 function finds the @emph{rightmost} matching subsequence. | |
4181 @end defun | |
4182 | |
4183 @node Sorting Sequences, , Searching Sequences, Sequences | |
4184 @section Sorting Sequences | |
4185 | |
4186 @defun sort* seq predicate @t{&key :key} | |
4187 This function sorts @var{seq} into increasing order as determined | |
4188 by using @var{predicate} to compare pairs of elements. @var{predicate} | |
4189 should return true (non-@code{nil}) if and only if its first argument | |
4190 is less than (not equal to) its second argument. For example, | |
4191 @code{<} and @code{string-lessp} are suitable predicate functions | |
4192 for sorting numbers and strings, respectively; @code{>} would sort | |
4193 numbers into decreasing rather than increasing order. | |
4194 | |
4195 This function differs from Emacs' built-in @code{sort} in that it | |
4196 can operate on any type of sequence, not just lists. Also, it | |
4197 accepts a @code{:key} argument which is used to preprocess data | |
4198 fed to the @var{predicate} function. For example, | |
4199 | |
4200 @example | |
4201 (setq data (sort data 'string-lessp :key 'downcase)) | |
4202 @end example | |
4203 | |
4204 @noindent | |
4205 sorts @var{data}, a sequence of strings, into increasing alphabetical | |
4206 order without regard to case. A @code{:key} function of @code{car} | |
4207 would be useful for sorting association lists. | |
4208 | |
4209 The @code{sort*} function is destructive; it sorts lists by actually | |
4210 rearranging the @code{cdr} pointers in suitable fashion. | |
4211 @end defun | |
4212 | |
4213 @defun stable-sort seq predicate @t{&key :key} | |
4214 This function sorts @var{seq} @dfn{stably}, meaning two elements | |
4215 which are equal in terms of @var{predicate} are guaranteed not to | |
4216 be rearranged out of their original order by the sort. | |
4217 | |
4218 In practice, @code{sort*} and @code{stable-sort} are equivalent | |
4219 in Emacs Lisp because the underlying @code{sort} function is | |
4220 stable by default. However, this package reserves the right to | |
4221 use non-stable methods for @code{sort*} in the future. | |
4222 @end defun | |
4223 | |
4224 @defun merge type seq1 seq2 predicate @t{&key :key} | |
4225 This function merges two sequences @var{seq1} and @var{seq2} by | |
4226 interleaving their elements. The result sequence, of type @var{type} | |
4227 (in the sense of @code{concatenate}), has length equal to the sum | |
4228 of the lengths of the two input sequences. The sequences may be | |
4229 modified destructively. Order of elements within @var{seq1} and | |
4230 @var{seq2} is preserved in the interleaving; elements of the two | |
4231 sequences are compared by @var{predicate} (in the sense of | |
4232 @code{sort}) and the lesser element goes first in the result. | |
4233 When elements are equal, those from @var{seq1} precede those from | |
4234 @var{seq2} in the result. Thus, if @var{seq1} and @var{seq2} are | |
4235 both sorted according to @var{predicate}, then the result will be | |
4236 a merged sequence which is (stably) sorted according to | |
4237 @var{predicate}. | |
4238 @end defun | |
4239 | |
4240 @node Lists, Hash Tables, Sequences, Top | |
4241 @chapter Lists | |
4242 | |
4243 @noindent | |
4244 The functions described here operate on lists. | |
4245 | |
4246 @menu | |
4247 * List Functions:: `caddr', `first', `last', `list*', etc. | |
4248 * Substitution of Expressions:: `subst', `sublis', etc. | |
4249 * Lists as Sets:: `member*', `adjoin', `union', etc. | |
4250 * Association Lists:: `assoc*', `rassoc*', `acons', `pairlis' | |
4251 @end menu | |
4252 | |
4253 @node List Functions, Substitution of Expressions, Lists, Lists | |
4254 @section List Functions | |
4255 | |
4256 @noindent | |
4257 This section describes a number of simple operations on lists, | |
4258 i.e., chains of cons cells. | |
4259 | |
4260 @defun caddr x | |
4261 This function is equivalent to @code{(car (cdr (cdr @var{x})))}. | |
4262 Likewise, this package defines all 28 @code{c@var{xxx}r} functions | |
4263 where @var{xxx} is up to four @samp{a}s and/or @samp{d}s. | |
4264 All of these functions are @code{setf}-able, and calls to them | |
4265 are expanded inline by the byte-compiler for maximum efficiency. | |
4266 @end defun | |
4267 | |
4268 @defun first x | |
4269 This function is a synonym for @code{(car @var{x})}. Likewise, | |
4270 the functions @code{second}, @code{third}, @dots{}, through | |
4271 @code{tenth} return the given element of the list @var{x}. | |
4272 @end defun | |
4273 | |
4274 @defun rest x | |
4275 This function is a synonym for @code{(cdr @var{x})}. | |
4276 @end defun | |
4277 | |
4278 @defun endp x | |
4279 Common Lisp defines this function to act like @code{null}, but | |
4280 signalling an error if @code{x} is neither a @code{nil} nor a | |
4281 cons cell. This package simply defines @code{endp} as a synonym | |
4282 for @code{null}. | |
4283 @end defun | |
4284 | |
4285 @defun list-length x | |
4286 This function returns the length of list @var{x}, exactly like | |
4287 @code{(length @var{x})}, except that if @var{x} is a circular | |
4288 list (where the cdr-chain forms a loop rather than terminating | |
4289 with @code{nil}), this function returns @code{nil}. (The regular | |
4290 @code{length} function would get stuck if given a circular list.) | |
4291 @end defun | |
4292 | |
4293 @defun last x &optional n | |
4294 This function returns the last cons, or the @var{n}th-to-last cons, | |
4295 of the list @var{x}. If @var{n} is omitted it defaults to 1. | |
4296 The ``last cons'' means the first cons cell of the list whose | |
4297 @code{cdr} is not another cons cell. (For normal lists, the | |
4298 @code{cdr} of the last cons will be @code{nil}.) This function | |
4299 returns @code{nil} if @var{x} is @code{nil} or shorter than | |
4300 @var{n}. Note that the last @emph{element} of the list is | |
4301 @code{(car (last @var{x}))}. | |
4302 @end defun | |
4303 | |
4304 @defun butlast x &optional n | |
4305 This function returns the list @var{x} with the last element, | |
4306 or the last @var{n} elements, removed. If @var{n} is greater | |
4307 than zero it makes a copy of the list so as not to damage the | |
4308 original list. In general, @code{(append (butlast @var{x} @var{n}) | |
4309 (last @var{x} @var{n}))} will return a list equal to @var{x}. | |
4310 @end defun | |
4311 | |
4312 @defun nbutlast x &optional n | |
4313 This is a version of @code{butlast} that works by destructively | |
4314 modifying the @code{cdr} of the appropriate element, rather than | |
4315 making a copy of the list. | |
4316 @end defun | |
4317 | |
4318 @defun list* arg &rest others | |
4319 This function constructs a list of its arguments. The final | |
4320 argument becomes the @code{cdr} of the last cell constructed. | |
4321 Thus, @code{(list* @var{a} @var{b} @var{c})} is equivalent to | |
4322 @code{(cons @var{a} (cons @var{b} @var{c}))}, and | |
4323 @code{(list* @var{a} @var{b} nil)} is equivalent to | |
4324 @code{(list @var{a} @var{b})}. | |
4325 | |
4326 (Note that this function really is called @code{list*} in Common | |
4327 Lisp; it is not a name invented for this package like @code{member*} | |
4328 or @code{defun*}.) | |
4329 @end defun | |
4330 | |
4331 @defun ldiff list sublist | |
4332 If @var{sublist} is a sublist of @var{list}, i.e., is @code{eq} to | |
4333 one of the cons cells of @var{list}, then this function returns | |
4334 a copy of the part of @var{list} up to but not including | |
4335 @var{sublist}. For example, @code{(ldiff x (cddr x))} returns | |
4336 the first two elements of the list @code{x}. The result is a | |
4337 copy; the original @var{list} is not modified. If @var{sublist} | |
4338 is not a sublist of @var{list}, a copy of the entire @var{list} | |
4339 is returned. | |
4340 @end defun | |
4341 | |
4342 @defun copy-list list | |
4343 This function returns a copy of the list @var{list}. It copies | |
4344 dotted lists like @code{(1 2 . 3)} correctly. | |
4345 @end defun | |
4346 | |
4347 @defun copy-tree x &optional vecp | |
4348 This function returns a copy of the tree of cons cells @var{x}. | |
4349 Unlike @code{copy-sequence} (and its alias @code{copy-list}), | |
4350 which copies only along the @code{cdr} direction, this function | |
4351 copies (recursively) along both the @code{car} and the @code{cdr} | |
4352 directions. If @var{x} is not a cons cell, the function simply | |
4353 returns @var{x} unchanged. If the optional @var{vecp} argument | |
4354 is true, this function copies vectors (recursively) as well as | |
4355 cons cells. | |
4356 @end defun | |
4357 | |
4358 @defun tree-equal x y @t{&key :test :test-not :key} | |
4359 This function compares two trees of cons cells. If @var{x} and | |
4360 @var{y} are both cons cells, their @code{car}s and @code{cdr}s are | |
4361 compared recursively. If neither @var{x} nor @var{y} is a cons | |
4362 cell, they are compared by @code{eql}, or according to the | |
4363 specified test. The @code{:key} function, if specified, is | |
4364 applied to the elements of both trees. @xref{Sequences}. | |
4365 @end defun | |
4366 | |
4367 @iftex | |
4368 @secno=3 | |
4369 @end iftex | |
4370 | |
4371 @node Substitution of Expressions, Lists as Sets, List Functions, Lists | |
4372 @section Substitution of Expressions | |
4373 | |
4374 @noindent | |
4375 These functions substitute elements throughout a tree of cons | |
4376 cells. (@xref{Sequence Functions}, for the @code{substitute} | |
4377 function, which works on just the top-level elements of a list.) | |
4378 | |
4379 @defun subst new old tree @t{&key :test :test-not :key} | |
4380 This function substitutes occurrences of @var{old} with @var{new} | |
4381 in @var{tree}, a tree of cons cells. It returns a substituted | |
4382 tree, which will be a copy except that it may share storage with | |
4383 the argument @var{tree} in parts where no substitutions occurred. | |
4384 The original @var{tree} is not modified. This function recurses | |
4385 on, and compares against @var{old}, both @code{car}s and @code{cdr}s | |
4386 of the component cons cells. If @var{old} is itself a cons cell, | |
4387 then matching cells in the tree are substituted as usual without | |
4388 recursively substituting in that cell. Comparisons with @var{old} | |
4389 are done according to the specified test (@code{eql} by default). | |
4390 The @code{:key} function is applied to the elements of the tree | |
4391 but not to @var{old}. | |
4392 @end defun | |
4393 | |
4394 @defun nsubst new old tree @t{&key :test :test-not :key} | |
4395 This function is like @code{subst}, except that it works by | |
4396 destructive modification (by @code{setcar} or @code{setcdr}) | |
4397 rather than copying. | |
4398 @end defun | |
4399 | |
4400 @findex subst-if | |
4401 @findex subst-if-not | |
4402 @findex nsubst-if | |
4403 @findex nsubst-if-not | |
4404 The @code{subst-if}, @code{subst-if-not}, @code{nsubst-if}, and | |
4405 @code{nsubst-if-not} functions are defined similarly. | |
4406 | |
4407 @defun sublis alist tree @t{&key :test :test-not :key} | |
4408 This function is like @code{subst}, except that it takes an | |
4409 association list @var{alist} of @var{old}-@var{new} pairs. | |
4410 Each element of the tree (after applying the @code{:key} | |
4411 function, if any), is compared with the @code{car}s of | |
4412 @var{alist}; if it matches, it is replaced by the corresponding | |
4413 @code{cdr}. | |
4414 @end defun | |
4415 | |
4416 @defun nsublis alist tree @t{&key :test :test-not :key} | |
4417 This is a destructive version of @code{sublis}. | |
4418 @end defun | |
4419 | |
4420 @node Lists as Sets, Association Lists, Substitution of Expressions, Lists | |
4421 @section Lists as Sets | |
4422 | |
4423 @noindent | |
4424 These functions perform operations on lists which represent sets | |
4425 of elements. | |
4426 | |
4427 @defun member item list | |
4428 This MacLisp-compatible function searches @var{list} for an element | |
4429 which is @code{equal} to @var{item}. The @code{member} function is | |
4430 built-in to Emacs 19; this package defines it equivalently in Emacs 18. | |
4431 See the following function for a Common-Lisp compatible version. | |
4432 @end defun | |
4433 | |
4434 @defun member* item list @t{&key :test :test-not :key} | |
4435 This function searches @var{list} for an element matching @var{item}. | |
4436 If a match is found, it returns the cons cell whose @code{car} was | |
4437 the matching element. Otherwise, it returns @code{nil}. Elements | |
4438 are compared by @code{eql} by default; you can use the @code{:test}, | |
4439 @code{:test-not}, and @code{:key} arguments to modify this behavior. | |
4440 @xref{Sequences}. | |
4441 | |
4442 Note that this function's name is suffixed by @samp{*} to avoid | |
4443 the incompatible @code{member} function defined in Emacs 19. | |
4444 (That function uses @code{equal} for comparisons; it is equivalent | |
4445 to @code{(member* @var{item} @var{list} :test 'equal)}.) | |
4446 @end defun | |
4447 | |
4448 @findex member-if | |
4449 @findex member-if-not | |
4450 The @code{member-if} and @code{member-if-not} functions | |
4451 analogously search for elements which satisfy a given predicate. | |
4452 | |
4453 @defun tailp sublist list | |
4454 This function returns @code{t} if @var{sublist} is a sublist of | |
4455 @var{list}, i.e., if @var{sublist} is @code{eql} to @var{list} or to | |
4456 any of its @code{cdr}s. | |
4457 @end defun | |
4458 | |
4459 @defun adjoin item list @t{&key :test :test-not :key} | |
4460 This function conses @var{item} onto the front of @var{list}, | |
4461 like @code{(cons @var{item} @var{list})}, but only if @var{item} | |
4462 is not already present on the list (as determined by @code{member*}). | |
4463 If a @code{:key} argument is specified, it is applied to | |
4464 @var{item} as well as to the elements of @var{list} during | |
4465 the search, on the reasoning that @var{item} is ``about'' to | |
4466 become part of the list. | |
4467 @end defun | |
4468 | |
4469 @defun union list1 list2 @t{&key :test :test-not :key} | |
4470 This function combines two lists which represent sets of items, | |
4471 returning a list that represents the union of those two sets. | |
4472 The result list will contain all items which appear in @var{list1} | |
4473 or @var{list2}, and no others. If an item appears in both | |
4474 @var{list1} and @var{list2} it will be copied only once. If | |
4475 an item is duplicated in @var{list1} or @var{list2}, it is | |
4476 undefined whether or not that duplication will survive in the | |
4477 result list. The order of elements in the result list is also | |
4478 undefined. | |
4479 @end defun | |
4480 | |
4481 @defun nunion list1 list2 @t{&key :test :test-not :key} | |
4482 This is a destructive version of @code{union}; rather than copying, | |
4483 it tries to reuse the storage of the argument lists if possible. | |
4484 @end defun | |
4485 | |
4486 @defun intersection list1 list2 @t{&key :test :test-not :key} | |
4487 This function computes the intersection of the sets represented | |
4488 by @var{list1} and @var{list2}. It returns the list of items | |
4489 which appear in both @var{list1} and @var{list2}. | |
4490 @end defun | |
4491 | |
4492 @defun nintersection list1 list2 @t{&key :test :test-not :key} | |
4493 This is a destructive version of @code{intersection}. It | |
4494 tries to reuse storage of @var{list1} rather than copying. | |
4495 It does @emph{not} reuse the storage of @var{list2}. | |
4496 @end defun | |
4497 | |
4498 @defun set-difference list1 list2 @t{&key :test :test-not :key} | |
4499 This function computes the ``set difference'' of @var{list1} | |
4500 and @var{list2}, i.e., the set of elements that appear in | |
4501 @var{list1} but @emph{not} in @var{list2}. | |
4502 @end defun | |
4503 | |
4504 @defun nset-difference list1 list2 @t{&key :test :test-not :key} | |
4505 This is a destructive @code{set-difference}, which will try | |
4506 to reuse @var{list1} if possible. | |
4507 @end defun | |
4508 | |
4509 @defun set-exclusive-or list1 list2 @t{&key :test :test-not :key} | |
4510 This function computes the ``set exclusive or'' of @var{list1} | |
4511 and @var{list2}, i.e., the set of elements that appear in | |
4512 exactly one of @var{list1} and @var{list2}. | |
4513 @end defun | |
4514 | |
4515 @defun nset-exclusive-or list1 list2 @t{&key :test :test-not :key} | |
4516 This is a destructive @code{set-exclusive-or}, which will try | |
4517 to reuse @var{list1} and @var{list2} if possible. | |
4518 @end defun | |
4519 | |
4520 @defun subsetp list1 list2 @t{&key :test :test-not :key} | |
4521 This function checks whether @var{list1} represents a subset | |
4522 of @var{list2}, i.e., whether every element of @var{list1} | |
4523 also appears in @var{list2}. | |
4524 @end defun | |
4525 | |
4526 @node Association Lists, , Lists as Sets, Lists | |
4527 @section Association Lists | |
4528 | |
4529 @noindent | |
4530 An @dfn{association list} is a list representing a mapping from | |
4531 one set of values to another; any list whose elements are cons | |
4532 cells is an association list. | |
4533 | |
4534 @defun assoc* item a-list @t{&key :test :test-not :key} | |
4535 This function searches the association list @var{a-list} for an | |
4536 element whose @code{car} matches (in the sense of @code{:test}, | |
4537 @code{:test-not}, and @code{:key}, or by comparison with @code{eql}) | |
4538 a given @var{item}. It returns the matching element, if any, | |
4539 otherwise @code{nil}. It ignores elements of @var{a-list} which | |
4540 are not cons cells. (This corresponds to the behavior of | |
4541 @code{assq} and @code{assoc} in Emacs Lisp; Common Lisp's | |
4542 @code{assoc} ignores @code{nil}s but considers any other non-cons | |
4543 elements of @var{a-list} to be an error.) | |
4544 @end defun | |
4545 | |
4546 @defun rassoc* item a-list @t{&key :test :test-not :key} | |
4547 This function searches for an element whose @code{cdr} matches | |
4548 @var{item}. If @var{a-list} represents a mapping, this applies | |
4549 the inverse of the mapping to @var{item}. | |
4550 @end defun | |
4551 | |
4552 @defun rassoc item a-list | |
4553 This function searches like @code{rassoc*} with a @code{:test} | |
4554 argument of @code{equal}. It is analogous to Emacs Lisp's | |
4555 standard @code{assoc} function, which derives from the MacLisp | |
4556 rather than the Common Lisp tradition. | |
4557 @end defun | |
4558 | |
4559 @findex assoc-if | |
4560 @findex assoc-if-not | |
4561 @findex rassoc-if | |
4562 @findex rassoc-if-not | |
4563 The @code{assoc-if}, @code{assoc-if-not}, @code{rassoc-if}, | |
4564 and @code{rassoc-if-not} functions are defined similarly. | |
4565 | |
4566 Two simple functions for constructing association lists are: | |
4567 | |
4568 @defun acons key value alist | |
4569 This is equivalent to @code{(cons (cons @var{key} @var{value}) @var{alist})}. | |
4570 @end defun | |
4571 | |
4572 @defun pairlis keys values &optional alist | |
4573 This is equivalent to @code{(nconc (mapcar* 'cons @var{keys} @var{values}) | |
4574 @var{alist})}. | |
4575 @end defun | |
4576 | |
4577 @node Hash Tables, Structures, Lists, Top | |
4578 @chapter Hash Tables | |
4579 | |
4580 @noindent | |
4581 Hash tables are now implemented directly in the C code and documented in | |
446 | 4582 @ref{Hash Tables,,, lispref, XEmacs Lisp Reference Manual}. |
428 | 4583 |
4584 @ignore | |
4585 A @dfn{hash table} is a data structure that maps ``keys'' onto | |
4586 ``values.'' Keys and values can be arbitrary Lisp data objects. | |
4587 Hash tables have the property that the time to search for a given | |
4588 key is roughly constant; simpler data structures like association | |
4589 lists take time proportional to the number of entries in the list. | |
4590 | |
4591 @defun make-hash-table @t{&key :test :size} | |
4592 This function creates and returns a hash-table object whose | |
4593 function for comparing elements is @code{:test} (@code{eql} | |
4594 by default), and which is allocated to fit about @code{:size} | |
4595 elements. The @code{:size} argument is purely advisory; the | |
4596 table will stretch automatically if you store more elements in | |
4597 it. If @code{:size} is omitted, a reasonable default is used. | |
4598 | |
4599 Common Lisp allows only @code{eq}, @code{eql}, @code{equal}, | |
4600 and @code{equalp} as legal values for the @code{:test} argument. | |
4601 In this package, any reasonable predicate function will work, | |
4602 though if you use something else you should check the details of | |
4603 the hashing function described below to make sure it is suitable | |
4604 for your predicate. | |
4605 | |
4606 Some versions of Emacs (like XEmacs) include a built-in hash | |
4607 table type; in these versions, @code{make-hash-table} with a test of | |
4608 @code{eq}, @code{eql}, or @code{equal} will use these built-in hash | |
4609 tables. In all other cases, it will return a hash-table object which | |
4610 takes the form of a list with an identifying ``tag'' symbol at the | |
4611 front. All of the hash table functions in this package can operate on | |
4612 both types of hash table; normally you will never know which type is | |
4613 being used. | |
4614 | |
4615 This function accepts the additional Common Lisp keywords | |
4616 @code{:rehash-size} and @code{:rehash-threshold}, but it ignores | |
4617 their values. | |
4618 @end defun | |
4619 | |
4620 @defun gethash key table &optional default | |
4621 This function looks up @var{key} in @var{table}. If @var{key} | |
4622 exists in the table, in the sense that it matches any of the existing | |
4623 keys according to the table's test function, then the associated value | |
4624 is returned. Otherwise, @var{default} (or @code{nil}) is returned. | |
4625 | |
4626 To store new data in the hash table, use @code{setf} on a call to | |
4627 @code{gethash}. If @var{key} already exists in the table, the | |
4628 corresponding value is changed to the stored value. If @var{key} | |
4629 does not already exist, a new entry is added to the table and the | |
4630 table is reallocated to a larger size if necessary. The @var{default} | |
4631 argument is allowed but ignored in this case. The situation is | |
440 | 4632 exactly analogous to that of @code{get}; @pxref{Property Lists}. |
428 | 4633 @end defun |
4634 | |
4635 @defun remhash key table | |
4636 This function removes the entry for @var{key} from @var{table}. | |
4637 If an entry was removed, it returns @code{t}. If @var{key} does | |
4638 not appear in the table, it does nothing and returns @code{nil}. | |
4639 @end defun | |
4640 | |
4641 @defun clrhash table | |
4642 This function removes all the entries from @var{table}, leaving | |
4643 an empty hash table. | |
4644 @end defun | |
4645 | |
4646 @defun maphash function table | |
4647 This function calls @var{function} for each entry in @var{table}. | |
4648 It passes two arguments to @var{function}, the key and the value | |
4649 of the given entry. The return value of @var{function} is ignored; | |
4650 @var{maphash} itself returns @code{nil}. @xref{Loop Facility}, for | |
4651 an alternate way of iterating over hash tables. | |
4652 @end defun | |
4653 | |
4654 @defun hash-table-count table This function returns the number of | |
4655 entries in @var{table}. @strong{Warning:} The current implementation of | |
4656 XEmacs hash-tables does not decrement the stored @code{count} | |
4657 when @code{remhash} removes an entry. Therefore, the return value of | |
4658 this function is not dependable if you have used @code{remhash} on the | |
4659 table and the table's test is @code{eq}, @code{eql}, or @code{equal}. | |
4660 A slower, but reliable, way to count the entries is | |
4661 @code{(loop for x being the hash-keys of @var{table} count t)}. | |
4662 @end defun | |
4663 | |
4664 @defun hash-table-p object This function returns @code{t} if | |
4665 @var{object} is a hash table, @code{nil} otherwise. It recognizes both | |
4666 types of hash tables (both XEmacs built-in tables and tables implemented | |
4667 with special lists.) | |
4668 @end defun | |
4669 | |
4670 Sometimes when dealing with hash tables it is useful to know the | |
4671 exact ``hash function'' that is used. This package implements | |
4672 hash tables using Emacs Lisp ``obarrays,'' which are the same | |
4673 data structure that Emacs Lisp uses to keep track of symbols. | |
4674 Each hash table includes an embedded obarray. Key values given | |
4675 to @code{gethash} are converted by various means into strings, | |
4676 which are then looked up in the obarray using @code{intern} and | |
4677 @code{intern-soft}. The symbol, or ``bucket,'' corresponding to | |
4678 a given key string includes as its @code{symbol-value} an association | |
4679 list of all key-value pairs which hash to that string. Depending | |
4680 on the test function, it is possible for many entries to hash to | |
4681 the same bucket. For example, if the test is @code{eql}, then the | |
4682 symbol @code{foo} and two separately built strings @code{"foo"} will | |
4683 create three entries in the same bucket. Search time is linear | |
4684 within buckets, so hash tables will be most effective if you arrange | |
4685 not to store too many things that hash the same. | |
4686 | |
4687 The following algorithm is used to convert Lisp objects to hash | |
4688 strings: | |
4689 | |
4690 @itemize @bullet | |
4691 @item | |
4692 Strings are used directly as hash strings. (However, if the test | |
4693 function is @code{equalp}, strings are @code{downcase}d first.) | |
4694 | |
4695 @item | |
4696 Symbols are hashed according to their @code{symbol-name}. | |
4697 | |
4698 @item | |
4699 Integers are hashed into one of 16 buckets depending on their value | |
4700 modulo 16. Floating-point numbers are truncated to integers and | |
4701 hashed modulo 16. | |
4702 | |
4703 @item | |
4704 Cons cells are hashed according to their @code{car}s; nonempty vectors | |
4705 are hashed according to their first element. | |
4706 | |
4707 @item | |
4708 All other types of objects hash into a single bucket named @code{"*"}. | |
4709 @end itemize | |
4710 | |
4711 @noindent | |
4712 Thus, for example, searching among many buffer objects in a hash table | |
4713 will devolve to a (still fairly fast) linear-time search through a | |
4714 single bucket, whereas searching for different symbols will be very | |
4715 fast since each symbol will, in general, hash into its own bucket. | |
4716 | |
4717 The size of the obarray in a hash table is automatically adjusted | |
4718 as the number of elements increases. | |
4719 | |
4720 As a special case, @code{make-hash-table} with a @code{:size} argument | |
4721 of 0 or 1 will create a hash-table object that uses a single association | |
4722 list rather than an obarray of many lists. For very small tables this | |
4723 structure will be more efficient since lookup does not require | |
4724 converting the key to a string or looking it up in an obarray. | |
4725 However, such tables are guaranteed to take time proportional to | |
4726 their size to do a search. | |
4727 @end ignore | |
4728 | |
4729 @iftex | |
4730 @chapno=18 | |
4731 @end iftex | |
4732 | |
4733 @node Structures, Assertions, Hash Tables, Top | |
4734 @chapter Structures | |
4735 | |
4736 @noindent | |
4737 The Common Lisp @dfn{structure} mechanism provides a general way | |
4738 to define data types similar to C's @code{struct} types. A | |
4739 structure is a Lisp object containing some number of @dfn{slots}, | |
4740 each of which can hold any Lisp data object. Functions are | |
4741 provided for accessing and setting the slots, creating or copying | |
4742 structure objects, and recognizing objects of a particular structure | |
4743 type. | |
4744 | |
4745 In true Common Lisp, each structure type is a new type distinct | |
4746 from all existing Lisp types. Since the underlying Emacs Lisp | |
4747 system provides no way to create new distinct types, this package | |
4748 implements structures as vectors (or lists upon request) with a | |
4749 special ``tag'' symbol to identify them. | |
4750 | |
4751 @defspec defstruct name slots@dots{} | |
4752 The @code{defstruct} form defines a new structure type called | |
4753 @var{name}, with the specified @var{slots}. (The @var{slots} | |
4754 may begin with a string which documents the structure type.) | |
4755 In the simplest case, @var{name} and each of the @var{slots} | |
4756 are symbols. For example, | |
4757 | |
4758 @example | |
4759 (defstruct person name age sex) | |
4760 @end example | |
4761 | |
4762 @noindent | |
4763 defines a struct type called @code{person} which contains three | |
4764 slots. Given a @code{person} object @var{p}, you can access those | |
4765 slots by calling @code{(person-name @var{p})}, @code{(person-age @var{p})}, | |
4766 and @code{(person-sex @var{p})}. You can also change these slots by | |
4767 using @code{setf} on any of these place forms: | |
4768 | |
4769 @example | |
4770 (incf (person-age birthday-boy)) | |
4771 @end example | |
4772 | |
4773 You can create a new @code{person} by calling @code{make-person}, | |
4774 which takes keyword arguments @code{:name}, @code{:age}, and | |
4775 @code{:sex} to specify the initial values of these slots in the | |
4776 new object. (Omitting any of these arguments leaves the corresponding | |
4777 slot ``undefined,'' according to the Common Lisp standard; in Emacs | |
4778 Lisp, such uninitialized slots are filled with @code{nil}.) | |
4779 | |
4780 Given a @code{person}, @code{(copy-person @var{p})} makes a new | |
4781 object of the same type whose slots are @code{eq} to those of @var{p}. | |
4782 | |
4783 Given any Lisp object @var{x}, @code{(person-p @var{x})} returns | |
4784 true if @var{x} looks like a @code{person}, false otherwise. (Again, | |
4785 in Common Lisp this predicate would be exact; in Emacs Lisp the | |
4786 best it can do is verify that @var{x} is a vector of the correct | |
4787 length which starts with the correct tag symbol.) | |
4788 | |
4789 Accessors like @code{person-name} normally check their arguments | |
4790 (effectively using @code{person-p}) and signal an error if the | |
4791 argument is the wrong type. This check is affected by | |
4792 @code{(optimize (safety @dots{}))} declarations. Safety level 1, | |
4793 the default, uses a somewhat optimized check that will detect all | |
4794 incorrect arguments, but may use an uninformative error message | |
4795 (e.g., ``expected a vector'' instead of ``expected a @code{person}''). | |
4796 Safety level 0 omits all checks except as provided by the underlying | |
4797 @code{aref} call; safety levels 2 and 3 do rigorous checking that will | |
4798 always print a descriptive error message for incorrect inputs. | |
4799 @xref{Declarations}. | |
4800 | |
4801 @example | |
4802 (setq dave (make-person :name "Dave" :sex 'male)) | |
4803 @result{} [cl-struct-person "Dave" nil male] | |
4804 (setq other (copy-person dave)) | |
4805 @result{} [cl-struct-person "Dave" nil male] | |
4806 (eq dave other) | |
4807 @result{} nil | |
4808 (eq (person-name dave) (person-name other)) | |
4809 @result{} t | |
4810 (person-p dave) | |
4811 @result{} t | |
4812 (person-p [1 2 3 4]) | |
4813 @result{} nil | |
4814 (person-p "Bogus") | |
4815 @result{} nil | |
4816 (person-p '[cl-struct-person counterfeit person object]) | |
4817 @result{} t | |
4818 @end example | |
4819 | |
4820 In general, @var{name} is either a name symbol or a list of a name | |
4821 symbol followed by any number of @dfn{struct options}; each @var{slot} | |
4822 is either a slot symbol or a list of the form @samp{(@var{slot-name} | |
4823 @var{default-value} @var{slot-options}@dots{})}. The @var{default-value} | |
4824 is a Lisp form which is evaluated any time an instance of the | |
4825 structure type is created without specifying that slot's value. | |
4826 | |
4827 Common Lisp defines several slot options, but the only one | |
4828 implemented in this package is @code{:read-only}. A non-@code{nil} | |
4829 value for this option means the slot should not be @code{setf}-able; | |
4830 the slot's value is determined when the object is created and does | |
4831 not change afterward. | |
4832 | |
4833 @example | |
4834 (defstruct person | |
4835 (name nil :read-only t) | |
4836 age | |
4837 (sex 'unknown)) | |
4838 @end example | |
4839 | |
4840 Any slot options other than @code{:read-only} are ignored. | |
4841 | |
4842 For obscure historical reasons, structure options take a different | |
4843 form than slot options. A structure option is either a keyword | |
4844 symbol, or a list beginning with a keyword symbol possibly followed | |
4845 by arguments. (By contrast, slot options are key-value pairs not | |
4846 enclosed in lists.) | |
4847 | |
4848 @example | |
4849 (defstruct (person (:constructor create-person) | |
4850 (:type list) | |
4851 :named) | |
4852 name age sex) | |
4853 @end example | |
4854 | |
4855 The following structure options are recognized. | |
4856 | |
4857 @table @code | |
4858 @iftex | |
4859 @itemmax=0 in | |
4860 @advance@leftskip-.5@tableindent | |
4861 @end iftex | |
4862 @item :conc-name | |
4863 The argument is a symbol whose print name is used as the prefix for | |
4864 the names of slot accessor functions. The default is the name of | |
4865 the struct type followed by a hyphen. The option @code{(:conc-name p-)} | |
4866 would change this prefix to @code{p-}. Specifying @code{nil} as an | |
4867 argument means no prefix, so that the slot names themselves are used | |
4868 to name the accessor functions. | |
4869 | |
4870 @item :constructor | |
4871 In the simple case, this option takes one argument which is an | |
4872 alternate name to use for the constructor function. The default | |
4873 is @code{make-@var{name}}, e.g., @code{make-person}. The above | |
4874 example changes this to @code{create-person}. Specifying @code{nil} | |
4875 as an argument means that no standard constructor should be | |
4876 generated at all. | |
4877 | |
4878 In the full form of this option, the constructor name is followed | |
4879 by an arbitrary argument list. @xref{Program Structure}, for a | |
4880 description of the format of Common Lisp argument lists. All | |
4881 options, such as @code{&rest} and @code{&key}, are supported. | |
4882 The argument names should match the slot names; each slot is | |
4883 initialized from the corresponding argument. Slots whose names | |
4884 do not appear in the argument list are initialized based on the | |
4885 @var{default-value} in their slot descriptor. Also, @code{&optional} | |
4886 and @code{&key} arguments which don't specify defaults take their | |
4887 defaults from the slot descriptor. It is legal to include arguments | |
4888 which don't correspond to slot names; these are useful if they are | |
4889 referred to in the defaults for optional, keyword, or @code{&aux} | |
4890 arguments which @emph{do} correspond to slots. | |
4891 | |
4892 You can specify any number of full-format @code{:constructor} | |
4893 options on a structure. The default constructor is still generated | |
4894 as well unless you disable it with a simple-format @code{:constructor} | |
4895 option. | |
4896 | |
4897 @example | |
4898 (defstruct | |
4899 (person | |
4900 (:constructor nil) ; no default constructor | |
4901 (:constructor new-person (name sex &optional (age 0))) | |
4902 (:constructor new-hound (&key (name "Rover") | |
4903 (dog-years 0) | |
4904 &aux (age (* 7 dog-years)) | |
4905 (sex 'canine)))) | |
4906 name age sex) | |
4907 @end example | |
4908 | |
4909 The first constructor here takes its arguments positionally rather | |
4910 than by keyword. (In official Common Lisp terminology, constructors | |
4911 that work By Order of Arguments instead of by keyword are called | |
4912 ``BOA constructors.'' No, I'm not making this up.) For example, | |
4913 @code{(new-person "Jane" 'female)} generates a person whose slots | |
4914 are @code{"Jane"}, 0, and @code{female}, respectively. | |
4915 | |
4916 The second constructor takes two keyword arguments, @code{:name}, | |
4917 which initializes the @code{name} slot and defaults to @code{"Rover"}, | |
4918 and @code{:dog-years}, which does not itself correspond to a slot | |
4919 but which is used to initialize the @code{age} slot. The @code{sex} | |
4920 slot is forced to the symbol @code{canine} with no syntax for | |
4921 overriding it. | |
4922 | |
4923 @item :copier | |
4924 The argument is an alternate name for the copier function for | |
4925 this type. The default is @code{copy-@var{name}}. @code{nil} | |
4926 means not to generate a copier function. (In this implementation, | |
4927 all copier functions are simply synonyms for @code{copy-sequence}.) | |
4928 | |
4929 @item :predicate | |
4930 The argument is an alternate name for the predicate which recognizes | |
4931 objects of this type. The default is @code{@var{name}-p}. @code{nil} | |
4932 means not to generate a predicate function. (If the @code{:type} | |
4933 option is used without the @code{:named} option, no predicate is | |
4934 ever generated.) | |
4935 | |
4936 In true Common Lisp, @code{typep} is always able to recognize a | |
4937 structure object even if @code{:predicate} was used. In this | |
4938 package, @code{typep} simply looks for a function called | |
4939 @code{@var{typename}-p}, so it will work for structure types | |
4940 only if they used the default predicate name. | |
4941 | |
4942 @item :include | |
4943 This option implements a very limited form of C++-style inheritance. | |
4944 The argument is the name of another structure type previously | |
4945 created with @code{defstruct}. The effect is to cause the new | |
4946 structure type to inherit all of the included structure's slots | |
4947 (plus, of course, any new slots described by this struct's slot | |
4948 descriptors). The new structure is considered a ``specialization'' | |
4949 of the included one. In fact, the predicate and slot accessors | |
4950 for the included type will also accept objects of the new type. | |
4951 | |
4952 If there are extra arguments to the @code{:include} option after | |
4953 the included-structure name, these options are treated as replacement | |
4954 slot descriptors for slots in the included structure, possibly with | |
4955 modified default values. Borrowing an example from Steele: | |
4956 | |
4957 @example | |
4958 (defstruct person name (age 0) sex) | |
4959 @result{} person | |
4960 (defstruct (astronaut (:include person (age 45))) | |
4961 helmet-size | |
4962 (favorite-beverage 'tang)) | |
4963 @result{} astronaut | |
4964 | |
4965 (setq joe (make-person :name "Joe")) | |
4966 @result{} [cl-struct-person "Joe" 0 nil] | |
4967 (setq buzz (make-astronaut :name "Buzz")) | |
4968 @result{} [cl-struct-astronaut "Buzz" 45 nil nil tang] | |
4969 | |
4970 (list (person-p joe) (person-p buzz)) | |
4971 @result{} (t t) | |
4972 (list (astronaut-p joe) (astronaut-p buzz)) | |
4973 @result{} (nil t) | |
4974 | |
4975 (person-name buzz) | |
4976 @result{} "Buzz" | |
4977 (astronaut-name joe) | |
4978 @result{} error: "astronaut-name accessing a non-astronaut" | |
4979 @end example | |
4980 | |
4981 Thus, if @code{astronaut} is a specialization of @code{person}, | |
4982 then every @code{astronaut} is also a @code{person} (but not the | |
4983 other way around). Every @code{astronaut} includes all the slots | |
4984 of a @code{person}, plus extra slots that are specific to | |
4985 astronauts. Operations that work on people (like @code{person-name}) | |
4986 work on astronauts just like other people. | |
4987 | |
4988 @item :print-function | |
4989 In full Common Lisp, this option allows you to specify a function | |
4990 which is called to print an instance of the structure type. The | |
4991 Emacs Lisp system offers no hooks into the Lisp printer which would | |
4992 allow for such a feature, so this package simply ignores | |
4993 @code{:print-function}. | |
4994 | |
4995 @item :type | |
4996 The argument should be one of the symbols @code{vector} or @code{list}. | |
4997 This tells which underlying Lisp data type should be used to implement | |
4998 the new structure type. Vectors are used by default, but | |
4999 @code{(:type list)} will cause structure objects to be stored as | |
5000 lists instead. | |
5001 | |
5002 The vector representation for structure objects has the advantage | |
5003 that all structure slots can be accessed quickly, although creating | |
5004 vectors is a bit slower in Emacs Lisp. Lists are easier to create, | |
5005 but take a relatively long time accessing the later slots. | |
5006 | |
5007 @item :named | |
5008 This option, which takes no arguments, causes a characteristic ``tag'' | |
5009 symbol to be stored at the front of the structure object. Using | |
5010 @code{:type} without also using @code{:named} will result in a | |
5011 structure type stored as plain vectors or lists with no identifying | |
5012 features. | |
5013 | |
5014 The default, if you don't specify @code{:type} explicitly, is to | |
5015 use named vectors. Therefore, @code{:named} is only useful in | |
5016 conjunction with @code{:type}. | |
5017 | |
5018 @example | |
5019 (defstruct (person1) name age sex) | |
5020 (defstruct (person2 (:type list) :named) name age sex) | |
5021 (defstruct (person3 (:type list)) name age sex) | |
5022 | |
5023 (setq p1 (make-person1)) | |
5024 @result{} [cl-struct-person1 nil nil nil] | |
5025 (setq p2 (make-person2)) | |
5026 @result{} (person2 nil nil nil) | |
5027 (setq p3 (make-person3)) | |
5028 @result{} (nil nil nil) | |
5029 | |
5030 (person1-p p1) | |
5031 @result{} t | |
5032 (person2-p p2) | |
5033 @result{} t | |
5034 (person3-p p3) | |
5035 @result{} error: function person3-p undefined | |
5036 @end example | |
5037 | |
5038 Since unnamed structures don't have tags, @code{defstruct} is not | |
5039 able to make a useful predicate for recognizing them. Also, | |
5040 accessors like @code{person3-name} will be generated but they | |
5041 will not be able to do any type checking. The @code{person3-name} | |
5042 function, for example, will simply be a synonym for @code{car} in | |
5043 this case. By contrast, @code{person2-name} is able to verify | |
5044 that its argument is indeed a @code{person2} object before | |
5045 proceeding. | |
5046 | |
5047 @item :initial-offset | |
5048 The argument must be a nonnegative integer. It specifies a | |
5049 number of slots to be left ``empty'' at the front of the | |
5050 structure. If the structure is named, the tag appears at the | |
5051 specified position in the list or vector; otherwise, the first | |
5052 slot appears at that position. Earlier positions are filled | |
5053 with @code{nil} by the constructors and ignored otherwise. If | |
5054 the type @code{:include}s another type, then @code{:initial-offset} | |
5055 specifies a number of slots to be skipped between the last slot | |
5056 of the included type and the first new slot. | |
5057 @end table | |
5058 @end defspec | |
5059 | |
5060 Except as noted, the @code{defstruct} facility of this package is | |
5061 entirely compatible with that of Common Lisp. | |
5062 | |
5063 @iftex | |
5064 @chapno=23 | |
5065 @end iftex | |
5066 | |
5067 @node Assertions, Efficiency Concerns, Structures, Top | |
5068 @chapter Assertions and Errors | |
5069 | |
5070 @noindent | |
5071 This section describes two macros that test @dfn{assertions}, i.e., | |
5072 conditions which must be true if the program is operating correctly. | |
5073 Assertions never add to the behavior of a Lisp program; they simply | |
5074 make ``sanity checks'' to make sure everything is as it should be. | |
5075 | |
5076 If the optimization property @code{speed} has been set to 3, and | |
5077 @code{safety} is less than 3, then the byte-compiler will optimize | |
5078 away the following assertions. Because assertions might be optimized | |
5079 away, it is a bad idea for them to include side-effects. | |
5080 | |
5081 @defspec assert test-form [show-args string args@dots{}] | |
5082 This form verifies that @var{test-form} is true (i.e., evaluates to | |
5083 a non-@code{nil} value). If so, it returns @code{nil}. If the test | |
5084 is not satisfied, @code{assert} signals an error. | |
5085 | |
5086 A default error message will be supplied which includes @var{test-form}. | |
5087 You can specify a different error message by including a @var{string} | |
5088 argument plus optional extra arguments. Those arguments are simply | |
5089 passed to @code{error} to signal the error. | |
5090 | |
5091 If the optional second argument @var{show-args} is @code{t} instead | |
5092 of @code{nil}, then the error message (with or without @var{string}) | |
5093 will also include all non-constant arguments of the top-level | |
5094 @var{form}. For example: | |
5095 | |
5096 @example | |
5097 (assert (> x 10) t "x is too small: %d") | |
5098 @end example | |
5099 | |
446 | 5100 This usage of @var{show-args} is a change to Common Lisp. In |
428 | 5101 true Common Lisp, the second argument gives a list of @var{places} |
5102 which can be @code{setf}'d by the user before continuing from the | |
446 | 5103 error. |
428 | 5104 @end defspec |
5105 | |
446 | 5106 @defspec check-type place type &optional string |
5107 This form verifies that @var{place} evaluates to a value of type | |
428 | 5108 @var{type}. If so, it returns @code{nil}. If not, @code{check-type} |
446 | 5109 signals a continuable @code{wrong-type-argument} error. The default |
5110 error message lists the erroneous value along with @var{type} and | |
5111 @var{place} themselves. If @var{string} is specified, it is included in | |
5112 the error message in place of @var{type}. For example: | |
428 | 5113 |
5114 @example | |
5115 (check-type x (integer 1 *) "a positive integer") | |
5116 @end example | |
5117 | |
5118 @xref{Type Predicates}, for a description of the type specifiers | |
5119 that may be used for @var{type}. | |
5120 | |
446 | 5121 Note that as in Common Lisp, the first argument to @code{check-type} |
5122 should be a @var{place} suitable for use by @code{setf}, because | |
5123 @code{check-type} signals a continuable error that allows the user to | |
5124 modify @var{place}, most simply by returning a value from the debugger. | |
428 | 5125 @end defspec |
5126 | |
5127 The following error-related macro is also defined: | |
5128 | |
5129 @defspec ignore-errors forms@dots{} | |
5130 This executes @var{forms} exactly like a @code{progn}, except that | |
5131 errors are ignored during the @var{forms}. More precisely, if | |
5132 an error is signalled then @code{ignore-errors} immediately | |
5133 aborts execution of the @var{forms} and returns @code{nil}. | |
5134 If the @var{forms} complete successfully, @code{ignore-errors} | |
5135 returns the result of the last @var{form}. | |
5136 @end defspec | |
5137 | |
5138 @node Efficiency Concerns, Common Lisp Compatibility, Assertions, Top | |
5139 @appendix Efficiency Concerns | |
5140 | |
5141 @appendixsec Macros | |
5142 | |
5143 @noindent | |
5144 Many of the advanced features of this package, such as @code{defun*}, | |
5145 @code{loop}, and @code{setf}, are implemented as Lisp macros. In | |
5146 byte-compiled code, these complex notations will be expanded into | |
5147 equivalent Lisp code which is simple and efficient. For example, | |
5148 the forms | |
5149 | |
5150 @example | |
5151 (incf i n) | |
5152 (push x (car p)) | |
5153 @end example | |
5154 | |
5155 @noindent | |
5156 are expanded at compile-time to the Lisp forms | |
5157 | |
5158 @example | |
5159 (setq i (+ i n)) | |
5160 (setcar p (cons x (car p))) | |
5161 @end example | |
5162 | |
5163 @noindent | |
5164 which are the most efficient ways of doing these respective operations | |
5165 in Lisp. Thus, there is no performance penalty for using the more | |
5166 readable @code{incf} and @code{push} forms in your compiled code. | |
5167 | |
5168 @emph{Interpreted} code, on the other hand, must expand these macros | |
5169 every time they are executed. For this reason it is strongly | |
5170 recommended that code making heavy use of macros be compiled. | |
5171 (The features labelled ``Special Form'' instead of ``Function'' in | |
5172 this manual are macros.) A loop using @code{incf} a hundred times | |
5173 will execute considerably faster if compiled, and will also | |
5174 garbage-collect less because the macro expansion will not have | |
5175 to be generated, used, and thrown away a hundred times. | |
5176 | |
5177 You can find out how a macro expands by using the | |
5178 @code{cl-prettyexpand} function. | |
5179 | |
5180 @defun cl-prettyexpand form &optional full | |
5181 This function takes a single Lisp form as an argument and inserts | |
5182 a nicely formatted copy of it in the current buffer (which must be | |
5183 in Lisp mode so that indentation works properly). It also expands | |
5184 all Lisp macros which appear in the form. The easiest way to use | |
5185 this function is to go to the @code{*scratch*} buffer and type, say, | |
5186 | |
5187 @example | |
5188 (cl-prettyexpand '(loop for x below 10 collect x)) | |
5189 @end example | |
5190 | |
5191 @noindent | |
5192 and type @kbd{C-x C-e} immediately after the closing parenthesis; | |
5193 the expansion | |
5194 | |
5195 @example | |
5196 (block nil | |
5197 (let* ((x 0) | |
5198 (G1004 nil)) | |
5199 (while (< x 10) | |
5200 (setq G1004 (cons x G1004)) | |
5201 (setq x (+ x 1))) | |
5202 (nreverse G1004))) | |
5203 @end example | |
5204 | |
5205 @noindent | |
5206 will be inserted into the buffer. (The @code{block} macro is | |
5207 expanded differently in the interpreter and compiler, so | |
5208 @code{cl-prettyexpand} just leaves it alone. The temporary | |
5209 variable @code{G1004} was created by @code{gensym}.) | |
5210 | |
5211 If the optional argument @var{full} is true, then @emph{all} | |
5212 macros are expanded, including @code{block}, @code{eval-when}, | |
5213 and compiler macros. Expansion is done as if @var{form} were | |
5214 a top-level form in a file being compiled. For example, | |
5215 | |
5216 @example | |
5217 (cl-prettyexpand '(pushnew 'x list)) | |
5218 @print{} (setq list (adjoin 'x list)) | |
5219 (cl-prettyexpand '(pushnew 'x list) t) | |
5220 @print{} (setq list (if (memq 'x list) list (cons 'x list))) | |
5221 (cl-prettyexpand '(caddr (member* 'a list)) t) | |
5222 @print{} (car (cdr (cdr (memq 'a list)))) | |
5223 @end example | |
5224 | |
5225 Note that @code{adjoin}, @code{caddr}, and @code{member*} all | |
5226 have built-in compiler macros to optimize them in common cases. | |
5227 @end defun | |
5228 | |
5229 @ifinfo | |
5230 @example | |
5231 | |
5232 @end example | |
5233 @end ifinfo | |
5234 @appendixsec Error Checking | |
5235 | |
5236 @noindent | |
5237 Common Lisp compliance has in general not been sacrificed for the | |
5238 sake of efficiency. A few exceptions have been made for cases | |
5239 where substantial gains were possible at the expense of marginal | |
5240 incompatibility. One example is the use of @code{memq} (which is | |
5241 treated very efficiently by the byte-compiler) to scan for keyword | |
5242 arguments; this can become confused in rare cases when keyword | |
5243 symbols are used as both keywords and data values at once. This | |
5244 is extremely unlikely to occur in practical code, and the use of | |
5245 @code{memq} allows functions with keyword arguments to be nearly | |
5246 as fast as functions that use @code{&optional} arguments. | |
5247 | |
5248 The Common Lisp standard (as embodied in Steele's book) uses the | |
5249 phrase ``it is an error if'' to indicate a situation which is not | |
5250 supposed to arise in complying programs; implementations are strongly | |
5251 encouraged but not required to signal an error in these situations. | |
5252 This package sometimes omits such error checking in the interest of | |
5253 compactness and efficiency. For example, @code{do} variable | |
5254 specifiers are supposed to be lists of one, two, or three forms; | |
5255 extra forms are ignored by this package rather than signalling a | |
5256 syntax error. The @code{endp} function is simply a synonym for | |
5257 @code{null} in this package. Functions taking keyword arguments | |
5258 will accept an odd number of arguments, treating the trailing | |
5259 keyword as if it were followed by the value @code{nil}. | |
5260 | |
5261 Argument lists (as processed by @code{defun*} and friends) | |
5262 @emph{are} checked rigorously except for the minor point just | |
5263 mentioned; in particular, keyword arguments are checked for | |
5264 validity, and @code{&allow-other-keys} and @code{:allow-other-keys} | |
5265 are fully implemented. Keyword validity checking is slightly | |
5266 time consuming (though not too bad in byte-compiled code); | |
5267 you can use @code{&allow-other-keys} to omit this check. Functions | |
5268 defined in this package such as @code{find} and @code{member*} | |
5269 do check their keyword arguments for validity. | |
5270 | |
5271 @ifinfo | |
5272 @example | |
5273 | |
5274 @end example | |
5275 @end ifinfo | |
5276 @appendixsec Optimizing Compiler | |
5277 | |
5278 @noindent | |
5279 The byte-compiler that comes with Emacs 18 normally fails to expand | |
5280 macros that appear in top-level positions in the file (i.e., outside | |
5281 of @code{defun}s or other enclosing forms). This would have | |
5282 disastrous consequences to programs that used such top-level macros | |
5283 as @code{defun*}, @code{eval-when}, and @code{defstruct}. To | |
5284 work around this problem, the @dfn{CL} package patches the Emacs | |
5285 18 compiler to expand top-level macros. This patch will apply to | |
5286 your own macros, too, if they are used in a top-level context. | |
5287 The patch will not harm versions of the Emacs 18 compiler which | |
5288 have already had a similar patch applied, nor will it affect the | |
5289 optimizing Emacs 19 byte-compiler written by Jamie Zawinski and | |
5290 Hallvard Furuseth. The patch is applied to the byte compiler's | |
5291 code in Emacs' memory, @emph{not} to the @file{bytecomp.elc} file | |
5292 stored on disk. | |
5293 | |
5294 The Emacs 19 compiler (for Emacs 18) is available from various | |
5295 Emacs Lisp archive sites such as @code{archive.cis.ohio-state.edu}. | |
5296 Its use is highly recommended; many of the Common Lisp macros emit | |
5297 code which can be improved by optimization. In particular, | |
5298 @code{block}s (whether explicit or implicit in constructs like | |
5299 @code{defun*} and @code{loop}) carry a fair run-time penalty; the | |
5300 optimizing compiler removes @code{block}s which are not actually | |
5301 referenced by @code{return} or @code{return-from} inside the block. | |
5302 | |
5303 @node Common Lisp Compatibility, Old CL Compatibility, Efficiency Concerns, Top | |
5304 @appendix Common Lisp Compatibility | |
5305 | |
5306 @noindent | |
5307 Following is a list of all known incompatibilities between this | |
5308 package and Common Lisp as documented in Steele (2nd edition). | |
5309 | |
5310 Certain function names, such as @code{member}, @code{assoc}, and | |
5311 @code{floor}, were already taken by (incompatible) Emacs Lisp | |
5312 functions; this package appends @samp{*} to the names of its | |
5313 Common Lisp versions of these functions. | |
5314 | |
5315 The word @code{defun*} is required instead of @code{defun} in order | |
5316 to use extended Common Lisp argument lists in a function. Likewise, | |
5317 @code{defmacro*} and @code{function*} are versions of those forms | |
5318 which understand full-featured argument lists. The @code{&whole} | |
5319 keyword does not work in @code{defmacro} argument lists (except | |
5320 inside recursive argument lists). | |
5321 | |
5322 In order to allow an efficient implementation, keyword arguments use | |
5323 a slightly cheesy parser which may be confused if a keyword symbol | |
5324 is passed as the @emph{value} of another keyword argument. | |
5325 (Specifically, @code{(memq :@var{keyword} @var{rest-of-arguments})} | |
5326 is used to scan for @code{:@var{keyword}} among the supplied | |
5327 keyword arguments.) | |
5328 | |
5329 The @code{eql} and @code{equal} predicates do not distinguish | |
5330 between IEEE floating-point plus and minus zero. The @code{equalp} | |
5331 predicate has several differences with Common Lisp; @pxref{Predicates}. | |
5332 | |
5333 The @code{setf} mechanism is entirely compatible, except that | |
5334 setf-methods return a list of five values rather than five | |
5335 values directly. Also, the new ``@code{setf} function'' concept | |
5336 (typified by @code{(defun (setf foo) @dots{})}) is not implemented. | |
5337 | |
5338 The @code{do-all-symbols} form is the same as @code{do-symbols} | |
5339 with no @var{obarray} argument. In Common Lisp, this form would | |
5340 iterate over all symbols in all packages. Since Emacs obarrays | |
5341 are not a first-class package mechanism, there is no way for | |
5342 @code{do-all-symbols} to locate any but the default obarray. | |
5343 | |
5344 The @code{loop} macro is complete except that @code{loop-finish} | |
5345 and type specifiers are unimplemented. | |
5346 | |
5347 Many Common Lisp declarations are ignored, and others match | |
5348 the Common Lisp standard in concept but not in detail. For | |
5349 example, local @code{special} declarations, which are purely | |
5350 advisory in Emacs Lisp, do not rigorously obey the scoping rules | |
5351 set down in Steele's book. | |
5352 | |
5353 The variable @code{*gensym-counter*} starts out with a pseudo-random | |
5354 value rather than with zero. This is to cope with the fact that | |
5355 generated symbols become interned when they are written to and | |
5356 loaded back from a file. | |
5357 | |
5358 The @code{defstruct} facility is compatible, except that structures | |
5359 are of type @code{:type vector :named} by default rather than some | |
5360 special, distinct type. Also, the @code{:type} slot option is ignored. | |
5361 | |
5362 The second argument of @code{check-type} is treated differently. | |
5363 | |
5364 @node Old CL Compatibility, Porting Common Lisp, Common Lisp Compatibility, Top | |
5365 @appendix Old CL Compatibility | |
5366 | |
5367 @noindent | |
5368 Following is a list of all known incompatibilities between this package | |
5369 and the older Quiroz @file{cl.el} package. | |
5370 | |
5371 The @code{defkeyword} form and @code{keywordp} function are not | |
5372 implemented in this package. | |
5373 | |
5374 The @code{member}, @code{floor}, @code{ceiling}, @code{truncate}, | |
5375 @code{round}, @code{mod}, and @code{rem} functions are suffixed | |
5376 by @samp{*} in this package to avoid collision with existing | |
5377 functions in Emacs 18 or Emacs 19. The older package simply | |
5378 redefined these functions, overwriting the built-in meanings and | |
5379 causing serious portability problems with Emacs 19. (Some more | |
5380 recent versions of the Quiroz package changed the names to | |
5381 @code{cl-member}, etc.; this package defines the latter names as | |
5382 aliases for @code{member*}, etc.) | |
5383 | |
5384 Certain functions in the old package which were buggy or inconsistent | |
5385 with the Common Lisp standard are incompatible with the conforming | |
5386 versions in this package. For example, @code{eql} and @code{member} | |
5387 were synonyms for @code{eq} and @code{memq} in that package, @code{setf} | |
5388 failed to preserve correct order of evaluation of its arguments, etc. | |
5389 | |
5390 Finally, unlike the older package, this package is careful to | |
5391 prefix all of its internal names with @code{cl-}. Except for a | |
5392 few functions which are explicitly defined as additional features | |
5393 (such as @code{floatp-safe} and @code{letf}), this package does not | |
5394 export any non-@samp{cl-} symbols which are not also part of Common | |
5395 Lisp. | |
5396 | |
5397 @ifinfo | |
5398 @example | |
5399 | |
5400 @end example | |
5401 @end ifinfo | |
5402 @appendixsec The @code{cl-compat} package | |
5403 | |
5404 @noindent | |
5405 The @dfn{CL} package includes emulations of some features of the | |
5406 old @file{cl.el}, in the form of a compatibility package | |
5407 @code{cl-compat}. To use it, put @code{(require 'cl-compat)} in | |
5408 your program. | |
5409 | |
5410 The old package defined a number of internal routines without | |
5411 @code{cl-} prefixes or other annotations. Call to these routines | |
5412 may have crept into existing Lisp code. @code{cl-compat} | |
5413 provides emulations of the following internal routines: | |
5414 @code{pair-with-newsyms}, @code{zip-lists}, @code{unzip-lists}, | |
5415 @code{reassemble-arglists}, @code{duplicate-symbols-p}, | |
5416 @code{safe-idiv}. | |
5417 | |
5418 Some @code{setf} forms translated into calls to internal | |
5419 functions that user code might call directly. The functions | |
5420 @code{setnth}, @code{setnthcdr}, and @code{setelt} fall in | |
5421 this category; they are defined by @code{cl-compat}, but the | |
5422 best fix is to change to use @code{setf} properly. | |
5423 | |
5424 The @code{cl-compat} file defines the keyword functions | |
5425 @code{keywordp}, @code{keyword-of}, and @code{defkeyword}, | |
5426 which are not defined by the new @dfn{CL} package because the | |
5427 use of keywords as data is discouraged. | |
5428 | |
5429 The @code{build-klist} mechanism for parsing keyword arguments | |
5430 is emulated by @code{cl-compat}; the @code{with-keyword-args} | |
5431 macro is not, however, and in any case it's best to change to | |
5432 use the more natural keyword argument processing offered by | |
5433 @code{defun*}. | |
5434 | |
5435 The functions @code{cl-floor}, @code{cl-ceiling}, @code{cl-truncate}, | |
5436 and @code{cl-round} are defined by @code{cl-compat} to use the | |
5437 old-style multiple-value mechanism, just as they did in the old | |
5438 package. The newer @code{floor*} and friends return their two | |
5439 results in a list rather than as multiple values. Note that | |
5440 older versions of the old package used the unadorned names | |
5441 @code{floor}, @code{ceiling}, etc.; @code{cl-compat} cannot use | |
5442 these names because they conflict with Emacs 19 built-ins. | |
5443 | |
5444 @node Porting Common Lisp, Function Index, Old CL Compatibility, Top | |
5445 @appendix Porting Common Lisp | |
5446 | |
5447 @noindent | |
5448 This package is meant to be used as an extension to Emacs Lisp, | |
5449 not as an Emacs implementation of true Common Lisp. Some of the | |
5450 remaining differences between Emacs Lisp and Common Lisp make it | |
5451 difficult to port large Common Lisp applications to Emacs. For | |
5452 one, some of the features in this package are not fully compliant | |
5453 with ANSI or Steele; @pxref{Common Lisp Compatibility}. But there | |
5454 are also quite a few features that this package does not provide | |
5455 at all. Here are some major omissions that you will want watch out | |
5456 for when bringing Common Lisp code into Emacs. | |
5457 | |
5458 @itemize @bullet | |
5459 @item | |
5460 Case-insensitivity. Symbols in Common Lisp are case-insensitive | |
5461 by default. Some programs refer to a function or variable as | |
5462 @code{foo} in one place and @code{Foo} or @code{FOO} in another. | |
5463 Emacs Lisp will treat these as three distinct symbols. | |
5464 | |
5465 Some Common Lisp code is written in all upper-case. While Emacs | |
5466 is happy to let the program's own functions and variables use | |
5467 this convention, calls to Lisp builtins like @code{if} and | |
5468 @code{defun} will have to be changed to lower-case. | |
5469 | |
5470 @item | |
5471 Lexical scoping. In Common Lisp, function arguments and @code{let} | |
5472 bindings apply only to references physically within their bodies | |
5473 (or within macro expansions in their bodies). Emacs Lisp, by | |
5474 contrast, uses @dfn{dynamic scoping} wherein a binding to a | |
5475 variable is visible even inside functions called from the body. | |
5476 | |
5477 Variables in Common Lisp can be made dynamically scoped by | |
5478 declaring them @code{special} or using @code{defvar}. In Emacs | |
5479 Lisp it is as if all variables were declared @code{special}. | |
5480 | |
5481 Often you can use code that was written for lexical scoping | |
5482 even in a dynamically scoped Lisp, but not always. Here is | |
5483 an example of a Common Lisp code fragment that would fail in | |
5484 Emacs Lisp: | |
5485 | |
5486 @example | |
5487 (defun map-odd-elements (func list) | |
5488 (loop for x in list | |
5489 for flag = t then (not flag) | |
5490 collect (if flag x (funcall func x)))) | |
5491 | |
5492 (defun add-odd-elements (list x) | |
5493 (map-odd-elements (function (lambda (a) (+ a x))) list)) | |
5494 @end example | |
5495 | |
5496 @noindent | |
5497 In Common Lisp, the two functions' usages of @code{x} are completely | |
5498 independent. In Emacs Lisp, the binding to @code{x} made by | |
5499 @code{add-odd-elements} will have been hidden by the binding | |
5500 in @code{map-odd-elements} by the time the @code{(+ a x)} function | |
5501 is called. | |
5502 | |
5503 (This package avoids such problems in its own mapping functions | |
5504 by using names like @code{cl-x} instead of @code{x} internally; | |
5505 as long as you don't use the @code{cl-} prefix for your own | |
5506 variables no collision can occur.) | |
5507 | |
5508 @xref{Lexical Bindings}, for a description of the @code{lexical-let} | |
5509 form which establishes a Common Lisp-style lexical binding, and some | |
5510 examples of how it differs from Emacs' regular @code{let}. | |
5511 | |
5512 @item | |
5513 Common Lisp allows the shorthand @code{#'x} to stand for | |
5514 @code{(function x)}, just as @code{'x} stands for @code{(quote x)}. | |
5515 In Common Lisp, one traditionally uses @code{#'} notation when | |
5516 referring to the name of a function. In Emacs Lisp, it works | |
5517 just as well to use a regular quote: | |
5518 | |
5519 @example | |
442 | 5520 (loop for x in y by #'cddr collect (mapcar #'plusp x)) ; Common Lisp |
5521 (loop for x in y by 'cddr collect (mapcar 'plusp x)) ; Emacs Lisp | |
428 | 5522 @end example |
5523 | |
5524 When @code{#'} introduces a @code{lambda} form, it is best to | |
5525 write out @code{(function ...)} longhand in Emacs Lisp. You can | |
5526 use a regular quote, but then the byte-compiler won't know that | |
5527 the @code{lambda} expression is code that can be compiled. | |
5528 | |
5529 @example | |
5530 (mapcar #'(lambda (x) (* x 2)) list) ; Common Lisp | |
5531 (mapcar (function (lambda (x) (* x 2))) list) ; Emacs Lisp | |
5532 @end example | |
5533 | |
5534 XEmacs supports @code{#'} notation starting with version 19.8. | |
5535 | |
5536 @item | |
5537 Reader macros. Common Lisp includes a second type of macro that | |
5538 works at the level of individual characters. For example, Common | |
5539 Lisp implements the quote notation by a reader macro called @code{'}, | |
5540 whereas Emacs Lisp's parser just treats quote as a special case. | |
5541 Some Lisp packages use reader macros to create special syntaxes | |
5542 for themselves, which the Emacs parser is incapable of reading. | |
5543 | |
5544 @item | |
5545 Other syntactic features. Common Lisp provides a number of | |
5546 notations beginning with @code{#} that the Emacs Lisp parser | |
5547 won't understand. For example, @samp{#| ... |#} is an | |
5548 alternate comment notation, and @samp{#+lucid (foo)} tells | |
5549 the parser to ignore the @code{(foo)} except in Lucid Common | |
5550 Lisp. | |
5551 | |
5552 The number prefixes `#b', `#o', and `#x', however, are supported | |
5553 by the Emacs Lisp parser to represent numbers in binary, octal, | |
5554 and hexadecimal notation (or radix), just like in Common Lisp. | |
5555 | |
5556 @item | |
5557 Packages. In Common Lisp, symbols are divided into @dfn{packages}. | |
5558 Symbols that are Lisp built-ins are typically stored in one package; | |
5559 symbols that are vendor extensions are put in another, and each | |
5560 application program would have a package for its own symbols. | |
5561 Certain symbols are ``exported'' by a package and others are | |
5562 internal; certain packages ``use'' or import the exported symbols | |
5563 of other packages. To access symbols that would not normally be | |
5564 visible due to this importing and exporting, Common Lisp provides | |
5565 a syntax like @code{package:symbol} or @code{package::symbol}. | |
5566 | |
5567 Emacs Lisp has a single namespace for all interned symbols, and | |
5568 then uses a naming convention of putting a prefix like @code{cl-} | |
5569 in front of the name. Some Emacs packages adopt the Common Lisp-like | |
5570 convention of using @code{cl:} or @code{cl::} as the prefix. | |
5571 However, the Emacs parser does not understand colons and just | |
5572 treats them as part of the symbol name. Thus, while @code{mapcar} | |
5573 and @code{lisp:mapcar} may refer to the same symbol in Common | |
5574 Lisp, they are totally distinct in Emacs Lisp. Common Lisp | |
5575 programs which refer to a symbol by the full name sometimes | |
5576 and the short name other times will not port cleanly to Emacs. | |
5577 | |
5578 Emacs Lisp does have a concept of ``obarrays,'' which are | |
5579 package-like collections of symbols, but this feature is not | |
5580 strong enough to be used as a true package mechanism. | |
5581 | |
5582 @item | |
5583 Keywords. The notation @code{:test-not} in Common Lisp really | |
5584 is a shorthand for @code{keyword:test-not}; keywords are just | |
5585 symbols in a built-in @code{keyword} package with the special | |
5586 property that all its symbols are automatically self-evaluating. | |
5587 Common Lisp programs often use keywords liberally to avoid | |
5588 having to use quotes. | |
5589 | |
5590 In Emacs Lisp a keyword is just a symbol whose name begins with | |
5591 a colon; since the Emacs parser does not treat them specially, | |
5592 they have to be explicitly made self-evaluating by a statement | |
5593 like @code{(setq :test-not ':test-not)}. This package arranges | |
5594 to execute such a statement whenever @code{defun*} or some | |
5595 other form sees a keyword being used as an argument. Common | |
5596 Lisp code that assumes that a symbol @code{:mumble} will be | |
5597 self-evaluating even though it was never introduced by a | |
5598 @code{defun*} will have to be fixed. | |
5599 | |
5600 @item | |
5601 The @code{format} function is quite different between Common | |
5602 Lisp and Emacs Lisp. It takes an additional ``destination'' | |
5603 argument before the format string. A destination of @code{nil} | |
5604 means to format to a string as in Emacs Lisp; a destination | |
5605 of @code{t} means to write to the terminal (similar to | |
5606 @code{message} in Emacs). Also, format control strings are | |
5607 utterly different; @code{~} is used instead of @code{%} to | |
5608 introduce format codes, and the set of available codes is | |
5609 much richer. There are no notations like @code{\n} for | |
5610 string literals; instead, @code{format} is used with the | |
5611 ``newline'' format code, @code{~%}. More advanced formatting | |
5612 codes provide such features as paragraph filling, case | |
5613 conversion, and even loops and conditionals. | |
5614 | |
5615 While it would have been possible to implement most of Common | |
5616 Lisp @code{format} in this package (under the name @code{format*}, | |
5617 of course), it was not deemed worthwhile. It would have required | |
5618 a huge amount of code to implement even a decent subset of | |
5619 @code{format*}, yet the functionality it would provide over | |
5620 Emacs Lisp's @code{format} would rarely be useful. | |
5621 | |
5622 @item | |
5623 Vector constants use square brackets in Emacs Lisp, but | |
5624 @code{#(a b c)} notation in Common Lisp. To further complicate | |
5625 matters, Emacs 19 introduces its own @code{#(} notation for | |
5626 something entirely different---strings with properties. | |
5627 | |
5628 @item | |
5629 Characters are distinct from integers in Common Lisp. The | |
5630 notation for character constants is also different: @code{#\A} | |
5631 instead of @code{?A}. Also, @code{string=} and @code{string-equal} | |
5632 are synonyms in Emacs Lisp whereas the latter is case-insensitive | |
5633 in Common Lisp. | |
5634 | |
5635 @item | |
5636 Data types. Some Common Lisp data types do not exist in Emacs | |
5637 Lisp. Rational numbers and complex numbers are not present, | |
5638 nor are large integers (all integers are ``fixnums''). All | |
5639 arrays are one-dimensional. There are no readtables or pathnames; | |
5640 streams are a set of existing data types rather than a new data | |
5641 type of their own. Hash tables, random-states, structures, and | |
5642 packages (obarrays) are built from Lisp vectors or lists rather | |
5643 than being distinct types. | |
5644 | |
5645 @item | |
5646 The Common Lisp Object System (CLOS) is not implemented, | |
5647 nor is the Common Lisp Condition System. | |
5648 | |
5649 @item | |
5650 Common Lisp features that are completely redundant with Emacs | |
5651 Lisp features of a different name generally have not been | |
5652 implemented. For example, Common Lisp writes @code{defconstant} | |
5653 where Emacs Lisp uses @code{defconst}. Similarly, @code{make-list} | |
5654 takes its arguments in different ways in the two Lisps but does | |
5655 exactly the same thing, so this package has not bothered to | |
5656 implement a Common Lisp-style @code{make-list}. | |
5657 | |
5658 @item | |
5659 A few more notable Common Lisp features not included in this | |
5660 package: @code{compiler-let}, @code{tagbody}, @code{prog}, | |
5661 @code{ldb/dpb}, @code{parse-integer}, @code{cerror}. | |
5662 | |
5663 @item | |
5664 Recursion. While recursion works in Emacs Lisp just like it | |
5665 does in Common Lisp, various details of the Emacs Lisp system | |
5666 and compiler make recursion much less efficient than it is in | |
5667 most Lisps. Some schools of thought prefer to use recursion | |
5668 in Lisp over other techniques; they would sum a list of | |
5669 numbers using something like | |
5670 | |
5671 @example | |
5672 (defun sum-list (list) | |
5673 (if list | |
5674 (+ (car list) (sum-list (cdr list))) | |
5675 0)) | |
5676 @end example | |
5677 | |
5678 @noindent | |
5679 where a more iteratively-minded programmer might write one of | |
5680 these forms: | |
5681 | |
5682 @example | |
5683 (let ((total 0)) (dolist (x my-list) (incf total x)) total) | |
5684 (loop for x in my-list sum x) | |
5685 @end example | |
5686 | |
5687 While this would be mainly a stylistic choice in most Common Lisps, | |
5688 in Emacs Lisp you should be aware that the iterative forms are | |
5689 much faster than recursion. Also, Lisp programmers will want to | |
5690 note that the current Emacs Lisp compiler does not optimize tail | |
5691 recursion. | |
5692 @end itemize | |
5693 | |
5694 @node Function Index, Variable Index, Porting Common Lisp, Top | |
5695 @unnumbered Function Index | |
5696 | |
5697 @printindex fn | |
5698 | |
5699 @node Variable Index, , Function Index, Top | |
5700 @unnumbered Variable Index | |
5701 | |
5702 @printindex vr | |
5703 | |
5704 @contents | |
5705 @bye |