Commit | Line | Data |
---|---|---|
b8d4c8d0 GM |
1 | @c -*-texinfo-*- |
2 | @c This is part of the GNU Emacs Lisp Reference Manual. | |
3 | @c Copyright (C) 1990, 1991, 1992, 1993, 1994, 1995, 1998, 1999, 2001, 2002, | |
5df4f04c | 4 | @c 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010, 2011 Free Software Foundation, Inc. |
b8d4c8d0 | 5 | @c See the file elisp.texi for copying conditions. |
6336d8c3 | 6 | @setfilename ../../info/control |
b8d4c8d0 GM |
7 | @node Control Structures, Variables, Evaluation, Top |
8 | @chapter Control Structures | |
9 | @cindex special forms for control structures | |
10 | @cindex control structures | |
11 | ||
12 | A Lisp program consists of expressions or @dfn{forms} (@pxref{Forms}). | |
13 | We control the order of execution of these forms by enclosing them in | |
14 | @dfn{control structures}. Control structures are special forms which | |
15 | control when, whether, or how many times to execute the forms they | |
16 | contain. | |
17 | ||
13e31e2b | 18 | @cindex textual order |
b8d4c8d0 GM |
19 | The simplest order of execution is sequential execution: first form |
20 | @var{a}, then form @var{b}, and so on. This is what happens when you | |
21 | write several forms in succession in the body of a function, or at top | |
22 | level in a file of Lisp code---the forms are executed in the order | |
23 | written. We call this @dfn{textual order}. For example, if a function | |
24 | body consists of two forms @var{a} and @var{b}, evaluation of the | |
25 | function evaluates first @var{a} and then @var{b}. The result of | |
26 | evaluating @var{b} becomes the value of the function. | |
27 | ||
28 | Explicit control structures make possible an order of execution other | |
29 | than sequential. | |
30 | ||
31 | Emacs Lisp provides several kinds of control structure, including | |
32 | other varieties of sequencing, conditionals, iteration, and (controlled) | |
33 | jumps---all discussed below. The built-in control structures are | |
34 | special forms since their subforms are not necessarily evaluated or not | |
35 | evaluated sequentially. You can use macros to define your own control | |
36 | structure constructs (@pxref{Macros}). | |
37 | ||
38 | @menu | |
39 | * Sequencing:: Evaluation in textual order. | |
40 | * Conditionals:: @code{if}, @code{cond}, @code{when}, @code{unless}. | |
41 | * Combining Conditions:: @code{and}, @code{or}, @code{not}. | |
42 | * Iteration:: @code{while} loops. | |
43 | * Nonlocal Exits:: Jumping out of a sequence. | |
44 | @end menu | |
45 | ||
46 | @node Sequencing | |
47 | @section Sequencing | |
48 | ||
49 | Evaluating forms in the order they appear is the most common way | |
50 | control passes from one form to another. In some contexts, such as in a | |
51 | function body, this happens automatically. Elsewhere you must use a | |
52 | control structure construct to do this: @code{progn}, the simplest | |
53 | control construct of Lisp. | |
54 | ||
55 | A @code{progn} special form looks like this: | |
56 | ||
57 | @example | |
58 | @group | |
59 | (progn @var{a} @var{b} @var{c} @dots{}) | |
60 | @end group | |
61 | @end example | |
62 | ||
63 | @noindent | |
64 | and it says to execute the forms @var{a}, @var{b}, @var{c}, and so on, in | |
65 | that order. These forms are called the @dfn{body} of the @code{progn} form. | |
66 | The value of the last form in the body becomes the value of the entire | |
67 | @code{progn}. @code{(progn)} returns @code{nil}. | |
68 | ||
69 | @cindex implicit @code{progn} | |
70 | In the early days of Lisp, @code{progn} was the only way to execute | |
71 | two or more forms in succession and use the value of the last of them. | |
72 | But programmers found they often needed to use a @code{progn} in the | |
73 | body of a function, where (at that time) only one form was allowed. So | |
74 | the body of a function was made into an ``implicit @code{progn}'': | |
75 | several forms are allowed just as in the body of an actual @code{progn}. | |
76 | Many other control structures likewise contain an implicit @code{progn}. | |
77 | As a result, @code{progn} is not used as much as it was many years ago. | |
78 | It is needed now most often inside an @code{unwind-protect}, @code{and}, | |
79 | @code{or}, or in the @var{then}-part of an @code{if}. | |
80 | ||
81 | @defspec progn forms@dots{} | |
82 | This special form evaluates all of the @var{forms}, in textual | |
83 | order, returning the result of the final form. | |
84 | ||
85 | @example | |
86 | @group | |
87 | (progn (print "The first form") | |
88 | (print "The second form") | |
89 | (print "The third form")) | |
90 | @print{} "The first form" | |
91 | @print{} "The second form" | |
92 | @print{} "The third form" | |
93 | @result{} "The third form" | |
94 | @end group | |
95 | @end example | |
96 | @end defspec | |
97 | ||
98 | Two other control constructs likewise evaluate a series of forms but return | |
99 | a different value: | |
100 | ||
101 | @defspec prog1 form1 forms@dots{} | |
102 | This special form evaluates @var{form1} and all of the @var{forms}, in | |
103 | textual order, returning the result of @var{form1}. | |
104 | ||
105 | @example | |
106 | @group | |
107 | (prog1 (print "The first form") | |
108 | (print "The second form") | |
109 | (print "The third form")) | |
110 | @print{} "The first form" | |
111 | @print{} "The second form" | |
112 | @print{} "The third form" | |
113 | @result{} "The first form" | |
114 | @end group | |
115 | @end example | |
116 | ||
117 | Here is a way to remove the first element from a list in the variable | |
118 | @code{x}, then return the value of that former element: | |
119 | ||
120 | @example | |
121 | (prog1 (car x) (setq x (cdr x))) | |
122 | @end example | |
123 | @end defspec | |
124 | ||
125 | @defspec prog2 form1 form2 forms@dots{} | |
126 | This special form evaluates @var{form1}, @var{form2}, and all of the | |
127 | following @var{forms}, in textual order, returning the result of | |
128 | @var{form2}. | |
129 | ||
130 | @example | |
131 | @group | |
132 | (prog2 (print "The first form") | |
133 | (print "The second form") | |
134 | (print "The third form")) | |
135 | @print{} "The first form" | |
136 | @print{} "The second form" | |
137 | @print{} "The third form" | |
138 | @result{} "The second form" | |
139 | @end group | |
140 | @end example | |
141 | @end defspec | |
142 | ||
143 | @node Conditionals | |
144 | @section Conditionals | |
145 | @cindex conditional evaluation | |
146 | ||
147 | Conditional control structures choose among alternatives. Emacs Lisp | |
148 | has four conditional forms: @code{if}, which is much the same as in | |
149 | other languages; @code{when} and @code{unless}, which are variants of | |
150 | @code{if}; and @code{cond}, which is a generalized case statement. | |
151 | ||
152 | @defspec if condition then-form else-forms@dots{} | |
153 | @code{if} chooses between the @var{then-form} and the @var{else-forms} | |
154 | based on the value of @var{condition}. If the evaluated @var{condition} is | |
155 | non-@code{nil}, @var{then-form} is evaluated and the result returned. | |
156 | Otherwise, the @var{else-forms} are evaluated in textual order, and the | |
157 | value of the last one is returned. (The @var{else} part of @code{if} is | |
158 | an example of an implicit @code{progn}. @xref{Sequencing}.) | |
159 | ||
160 | If @var{condition} has the value @code{nil}, and no @var{else-forms} are | |
161 | given, @code{if} returns @code{nil}. | |
162 | ||
163 | @code{if} is a special form because the branch that is not selected is | |
164 | never evaluated---it is ignored. Thus, in the example below, | |
165 | @code{true} is not printed because @code{print} is never called. | |
166 | ||
167 | @example | |
168 | @group | |
169 | (if nil | |
170 | (print 'true) | |
171 | 'very-false) | |
172 | @result{} very-false | |
173 | @end group | |
174 | @end example | |
175 | @end defspec | |
176 | ||
177 | @defmac when condition then-forms@dots{} | |
178 | This is a variant of @code{if} where there are no @var{else-forms}, | |
179 | and possibly several @var{then-forms}. In particular, | |
180 | ||
181 | @example | |
182 | (when @var{condition} @var{a} @var{b} @var{c}) | |
183 | @end example | |
184 | ||
185 | @noindent | |
186 | is entirely equivalent to | |
187 | ||
188 | @example | |
189 | (if @var{condition} (progn @var{a} @var{b} @var{c}) nil) | |
190 | @end example | |
191 | @end defmac | |
192 | ||
193 | @defmac unless condition forms@dots{} | |
194 | This is a variant of @code{if} where there is no @var{then-form}: | |
195 | ||
196 | @example | |
197 | (unless @var{condition} @var{a} @var{b} @var{c}) | |
198 | @end example | |
199 | ||
200 | @noindent | |
201 | is entirely equivalent to | |
202 | ||
203 | @example | |
204 | (if @var{condition} nil | |
205 | @var{a} @var{b} @var{c}) | |
206 | @end example | |
207 | @end defmac | |
208 | ||
209 | @defspec cond clause@dots{} | |
210 | @code{cond} chooses among an arbitrary number of alternatives. Each | |
211 | @var{clause} in the @code{cond} must be a list. The @sc{car} of this | |
212 | list is the @var{condition}; the remaining elements, if any, the | |
213 | @var{body-forms}. Thus, a clause looks like this: | |
214 | ||
215 | @example | |
216 | (@var{condition} @var{body-forms}@dots{}) | |
217 | @end example | |
218 | ||
219 | @code{cond} tries the clauses in textual order, by evaluating the | |
220 | @var{condition} of each clause. If the value of @var{condition} is | |
221 | non-@code{nil}, the clause ``succeeds''; then @code{cond} evaluates its | |
222 | @var{body-forms}, and the value of the last of @var{body-forms} becomes | |
223 | the value of the @code{cond}. The remaining clauses are ignored. | |
224 | ||
225 | If the value of @var{condition} is @code{nil}, the clause ``fails,'' so | |
226 | the @code{cond} moves on to the following clause, trying its | |
227 | @var{condition}. | |
228 | ||
229 | If every @var{condition} evaluates to @code{nil}, so that every clause | |
230 | fails, @code{cond} returns @code{nil}. | |
231 | ||
232 | A clause may also look like this: | |
233 | ||
234 | @example | |
235 | (@var{condition}) | |
236 | @end example | |
237 | ||
238 | @noindent | |
239 | Then, if @var{condition} is non-@code{nil} when tested, the value of | |
240 | @var{condition} becomes the value of the @code{cond} form. | |
241 | ||
242 | The following example has four clauses, which test for the cases where | |
243 | the value of @code{x} is a number, string, buffer and symbol, | |
244 | respectively: | |
245 | ||
246 | @example | |
247 | @group | |
248 | (cond ((numberp x) x) | |
249 | ((stringp x) x) | |
250 | ((bufferp x) | |
251 | (setq temporary-hack x) ; @r{multiple body-forms} | |
252 | (buffer-name x)) ; @r{in one clause} | |
253 | ((symbolp x) (symbol-value x))) | |
254 | @end group | |
255 | @end example | |
256 | ||
257 | Often we want to execute the last clause whenever none of the previous | |
258 | clauses was successful. To do this, we use @code{t} as the | |
259 | @var{condition} of the last clause, like this: @code{(t | |
260 | @var{body-forms})}. The form @code{t} evaluates to @code{t}, which is | |
261 | never @code{nil}, so this clause never fails, provided the @code{cond} | |
262 | gets to it at all. | |
263 | ||
264 | For example, | |
265 | ||
266 | @example | |
267 | @group | |
268 | (setq a 5) | |
269 | (cond ((eq a 'hack) 'foo) | |
270 | (t "default")) | |
271 | @result{} "default" | |
272 | @end group | |
273 | @end example | |
274 | ||
275 | @noindent | |
276 | This @code{cond} expression returns @code{foo} if the value of @code{a} | |
277 | is @code{hack}, and returns the string @code{"default"} otherwise. | |
278 | @end defspec | |
279 | ||
280 | Any conditional construct can be expressed with @code{cond} or with | |
281 | @code{if}. Therefore, the choice between them is a matter of style. | |
282 | For example: | |
283 | ||
284 | @example | |
285 | @group | |
286 | (if @var{a} @var{b} @var{c}) | |
287 | @equiv{} | |
288 | (cond (@var{a} @var{b}) (t @var{c})) | |
289 | @end group | |
290 | @end example | |
291 | ||
292 | @node Combining Conditions | |
293 | @section Constructs for Combining Conditions | |
294 | ||
295 | This section describes three constructs that are often used together | |
296 | with @code{if} and @code{cond} to express complicated conditions. The | |
297 | constructs @code{and} and @code{or} can also be used individually as | |
298 | kinds of multiple conditional constructs. | |
299 | ||
300 | @defun not condition | |
301 | This function tests for the falsehood of @var{condition}. It returns | |
302 | @code{t} if @var{condition} is @code{nil}, and @code{nil} otherwise. | |
303 | The function @code{not} is identical to @code{null}, and we recommend | |
304 | using the name @code{null} if you are testing for an empty list. | |
305 | @end defun | |
306 | ||
307 | @defspec and conditions@dots{} | |
308 | The @code{and} special form tests whether all the @var{conditions} are | |
309 | true. It works by evaluating the @var{conditions} one by one in the | |
310 | order written. | |
311 | ||
312 | If any of the @var{conditions} evaluates to @code{nil}, then the result | |
313 | of the @code{and} must be @code{nil} regardless of the remaining | |
314 | @var{conditions}; so @code{and} returns @code{nil} right away, ignoring | |
315 | the remaining @var{conditions}. | |
316 | ||
317 | If all the @var{conditions} turn out non-@code{nil}, then the value of | |
318 | the last of them becomes the value of the @code{and} form. Just | |
319 | @code{(and)}, with no @var{conditions}, returns @code{t}, appropriate | |
320 | because all the @var{conditions} turned out non-@code{nil}. (Think | |
321 | about it; which one did not?) | |
322 | ||
323 | Here is an example. The first condition returns the integer 1, which is | |
324 | not @code{nil}. Similarly, the second condition returns the integer 2, | |
325 | which is not @code{nil}. The third condition is @code{nil}, so the | |
326 | remaining condition is never evaluated. | |
327 | ||
328 | @example | |
329 | @group | |
330 | (and (print 1) (print 2) nil (print 3)) | |
331 | @print{} 1 | |
332 | @print{} 2 | |
333 | @result{} nil | |
334 | @end group | |
335 | @end example | |
336 | ||
337 | Here is a more realistic example of using @code{and}: | |
338 | ||
339 | @example | |
340 | @group | |
341 | (if (and (consp foo) (eq (car foo) 'x)) | |
342 | (message "foo is a list starting with x")) | |
343 | @end group | |
344 | @end example | |
345 | ||
346 | @noindent | |
347 | Note that @code{(car foo)} is not executed if @code{(consp foo)} returns | |
348 | @code{nil}, thus avoiding an error. | |
349 | ||
350 | @code{and} expressions can also be written using either @code{if} or | |
351 | @code{cond}. Here's how: | |
352 | ||
353 | @example | |
354 | @group | |
355 | (and @var{arg1} @var{arg2} @var{arg3}) | |
356 | @equiv{} | |
357 | (if @var{arg1} (if @var{arg2} @var{arg3})) | |
358 | @equiv{} | |
359 | (cond (@var{arg1} (cond (@var{arg2} @var{arg3})))) | |
360 | @end group | |
361 | @end example | |
362 | @end defspec | |
363 | ||
364 | @defspec or conditions@dots{} | |
365 | The @code{or} special form tests whether at least one of the | |
366 | @var{conditions} is true. It works by evaluating all the | |
367 | @var{conditions} one by one in the order written. | |
368 | ||
369 | If any of the @var{conditions} evaluates to a non-@code{nil} value, then | |
370 | the result of the @code{or} must be non-@code{nil}; so @code{or} returns | |
371 | right away, ignoring the remaining @var{conditions}. The value it | |
372 | returns is the non-@code{nil} value of the condition just evaluated. | |
373 | ||
374 | If all the @var{conditions} turn out @code{nil}, then the @code{or} | |
375 | expression returns @code{nil}. Just @code{(or)}, with no | |
376 | @var{conditions}, returns @code{nil}, appropriate because all the | |
377 | @var{conditions} turned out @code{nil}. (Think about it; which one | |
378 | did not?) | |
379 | ||
380 | For example, this expression tests whether @code{x} is either | |
381 | @code{nil} or the integer zero: | |
382 | ||
383 | @example | |
384 | (or (eq x nil) (eq x 0)) | |
385 | @end example | |
386 | ||
387 | Like the @code{and} construct, @code{or} can be written in terms of | |
388 | @code{cond}. For example: | |
389 | ||
390 | @example | |
391 | @group | |
392 | (or @var{arg1} @var{arg2} @var{arg3}) | |
393 | @equiv{} | |
394 | (cond (@var{arg1}) | |
395 | (@var{arg2}) | |
396 | (@var{arg3})) | |
397 | @end group | |
398 | @end example | |
399 | ||
400 | You could almost write @code{or} in terms of @code{if}, but not quite: | |
401 | ||
402 | @example | |
403 | @group | |
404 | (if @var{arg1} @var{arg1} | |
405 | (if @var{arg2} @var{arg2} | |
406 | @var{arg3})) | |
407 | @end group | |
408 | @end example | |
409 | ||
410 | @noindent | |
411 | This is not completely equivalent because it can evaluate @var{arg1} or | |
412 | @var{arg2} twice. By contrast, @code{(or @var{arg1} @var{arg2} | |
413 | @var{arg3})} never evaluates any argument more than once. | |
414 | @end defspec | |
415 | ||
416 | @node Iteration | |
417 | @section Iteration | |
418 | @cindex iteration | |
419 | @cindex recursion | |
420 | ||
421 | Iteration means executing part of a program repetitively. For | |
422 | example, you might want to repeat some computation once for each element | |
423 | of a list, or once for each integer from 0 to @var{n}. You can do this | |
424 | in Emacs Lisp with the special form @code{while}: | |
425 | ||
426 | @defspec while condition forms@dots{} | |
427 | @code{while} first evaluates @var{condition}. If the result is | |
428 | non-@code{nil}, it evaluates @var{forms} in textual order. Then it | |
429 | reevaluates @var{condition}, and if the result is non-@code{nil}, it | |
430 | evaluates @var{forms} again. This process repeats until @var{condition} | |
431 | evaluates to @code{nil}. | |
432 | ||
433 | There is no limit on the number of iterations that may occur. The loop | |
434 | will continue until either @var{condition} evaluates to @code{nil} or | |
435 | until an error or @code{throw} jumps out of it (@pxref{Nonlocal Exits}). | |
436 | ||
437 | The value of a @code{while} form is always @code{nil}. | |
438 | ||
439 | @example | |
440 | @group | |
441 | (setq num 0) | |
442 | @result{} 0 | |
443 | @end group | |
444 | @group | |
445 | (while (< num 4) | |
446 | (princ (format "Iteration %d." num)) | |
447 | (setq num (1+ num))) | |
448 | @print{} Iteration 0. | |
449 | @print{} Iteration 1. | |
450 | @print{} Iteration 2. | |
451 | @print{} Iteration 3. | |
452 | @result{} nil | |
453 | @end group | |
454 | @end example | |
455 | ||
456 | To write a ``repeat...until'' loop, which will execute something on each | |
457 | iteration and then do the end-test, put the body followed by the | |
458 | end-test in a @code{progn} as the first argument of @code{while}, as | |
459 | shown here: | |
460 | ||
461 | @example | |
462 | @group | |
463 | (while (progn | |
464 | (forward-line 1) | |
465 | (not (looking-at "^$")))) | |
466 | @end group | |
467 | @end example | |
468 | ||
469 | @noindent | |
470 | This moves forward one line and continues moving by lines until it | |
471 | reaches an empty line. It is peculiar in that the @code{while} has no | |
472 | body, just the end test (which also does the real work of moving point). | |
473 | @end defspec | |
474 | ||
475 | The @code{dolist} and @code{dotimes} macros provide convenient ways to | |
476 | write two common kinds of loops. | |
477 | ||
478 | @defmac dolist (var list [result]) body@dots{} | |
479 | This construct executes @var{body} once for each element of | |
480 | @var{list}, binding the variable @var{var} locally to hold the current | |
481 | element. Then it returns the value of evaluating @var{result}, or | |
482 | @code{nil} if @var{result} is omitted. For example, here is how you | |
483 | could use @code{dolist} to define the @code{reverse} function: | |
484 | ||
485 | @example | |
486 | (defun reverse (list) | |
487 | (let (value) | |
488 | (dolist (elt list value) | |
489 | (setq value (cons elt value))))) | |
490 | @end example | |
491 | @end defmac | |
492 | ||
493 | @defmac dotimes (var count [result]) body@dots{} | |
494 | This construct executes @var{body} once for each integer from 0 | |
495 | (inclusive) to @var{count} (exclusive), binding the variable @var{var} | |
496 | to the integer for the current iteration. Then it returns the value | |
497 | of evaluating @var{result}, or @code{nil} if @var{result} is omitted. | |
498 | Here is an example of using @code{dotimes} to do something 100 times: | |
499 | ||
500 | @example | |
501 | (dotimes (i 100) | |
502 | (insert "I will not obey absurd orders\n")) | |
503 | @end example | |
504 | @end defmac | |
505 | ||
506 | @node Nonlocal Exits | |
507 | @section Nonlocal Exits | |
508 | @cindex nonlocal exits | |
509 | ||
510 | A @dfn{nonlocal exit} is a transfer of control from one point in a | |
511 | program to another remote point. Nonlocal exits can occur in Emacs Lisp | |
512 | as a result of errors; you can also use them under explicit control. | |
513 | Nonlocal exits unbind all variable bindings made by the constructs being | |
514 | exited. | |
515 | ||
516 | @menu | |
517 | * Catch and Throw:: Nonlocal exits for the program's own purposes. | |
518 | * Examples of Catch:: Showing how such nonlocal exits can be written. | |
519 | * Errors:: How errors are signaled and handled. | |
520 | * Cleanups:: Arranging to run a cleanup form if an error happens. | |
521 | @end menu | |
522 | ||
523 | @node Catch and Throw | |
524 | @subsection Explicit Nonlocal Exits: @code{catch} and @code{throw} | |
525 | ||
526 | Most control constructs affect only the flow of control within the | |
527 | construct itself. The function @code{throw} is the exception to this | |
528 | rule of normal program execution: it performs a nonlocal exit on | |
529 | request. (There are other exceptions, but they are for error handling | |
530 | only.) @code{throw} is used inside a @code{catch}, and jumps back to | |
531 | that @code{catch}. For example: | |
532 | ||
533 | @example | |
534 | @group | |
535 | (defun foo-outer () | |
536 | (catch 'foo | |
537 | (foo-inner))) | |
538 | ||
539 | (defun foo-inner () | |
540 | @dots{} | |
541 | (if x | |
542 | (throw 'foo t)) | |
543 | @dots{}) | |
544 | @end group | |
545 | @end example | |
546 | ||
547 | @noindent | |
548 | The @code{throw} form, if executed, transfers control straight back to | |
549 | the corresponding @code{catch}, which returns immediately. The code | |
550 | following the @code{throw} is not executed. The second argument of | |
551 | @code{throw} is used as the return value of the @code{catch}. | |
552 | ||
553 | The function @code{throw} finds the matching @code{catch} based on the | |
554 | first argument: it searches for a @code{catch} whose first argument is | |
555 | @code{eq} to the one specified in the @code{throw}. If there is more | |
556 | than one applicable @code{catch}, the innermost one takes precedence. | |
557 | Thus, in the above example, the @code{throw} specifies @code{foo}, and | |
558 | the @code{catch} in @code{foo-outer} specifies the same symbol, so that | |
559 | @code{catch} is the applicable one (assuming there is no other matching | |
560 | @code{catch} in between). | |
561 | ||
562 | Executing @code{throw} exits all Lisp constructs up to the matching | |
563 | @code{catch}, including function calls. When binding constructs such as | |
564 | @code{let} or function calls are exited in this way, the bindings are | |
565 | unbound, just as they are when these constructs exit normally | |
566 | (@pxref{Local Variables}). Likewise, @code{throw} restores the buffer | |
567 | and position saved by @code{save-excursion} (@pxref{Excursions}), and | |
568 | the narrowing status saved by @code{save-restriction} and the window | |
569 | selection saved by @code{save-window-excursion} (@pxref{Window | |
570 | Configurations}). It also runs any cleanups established with the | |
571 | @code{unwind-protect} special form when it exits that form | |
572 | (@pxref{Cleanups}). | |
573 | ||
574 | The @code{throw} need not appear lexically within the @code{catch} | |
575 | that it jumps to. It can equally well be called from another function | |
576 | called within the @code{catch}. As long as the @code{throw} takes place | |
577 | chronologically after entry to the @code{catch}, and chronologically | |
578 | before exit from it, it has access to that @code{catch}. This is why | |
579 | @code{throw} can be used in commands such as @code{exit-recursive-edit} | |
580 | that throw back to the editor command loop (@pxref{Recursive Editing}). | |
581 | ||
582 | @cindex CL note---only @code{throw} in Emacs | |
583 | @quotation | |
584 | @b{Common Lisp note:} Most other versions of Lisp, including Common Lisp, | |
585 | have several ways of transferring control nonsequentially: @code{return}, | |
586 | @code{return-from}, and @code{go}, for example. Emacs Lisp has only | |
587 | @code{throw}. | |
588 | @end quotation | |
589 | ||
590 | @defspec catch tag body@dots{} | |
591 | @cindex tag on run time stack | |
592 | @code{catch} establishes a return point for the @code{throw} function. | |
593 | The return point is distinguished from other such return points by | |
594 | @var{tag}, which may be any Lisp object except @code{nil}. The argument | |
595 | @var{tag} is evaluated normally before the return point is established. | |
596 | ||
597 | With the return point in effect, @code{catch} evaluates the forms of the | |
598 | @var{body} in textual order. If the forms execute normally (without | |
599 | error or nonlocal exit) the value of the last body form is returned from | |
600 | the @code{catch}. | |
601 | ||
602 | If a @code{throw} is executed during the execution of @var{body}, | |
603 | specifying the same value @var{tag}, the @code{catch} form exits | |
604 | immediately; the value it returns is whatever was specified as the | |
605 | second argument of @code{throw}. | |
606 | @end defspec | |
607 | ||
608 | @defun throw tag value | |
609 | The purpose of @code{throw} is to return from a return point previously | |
610 | established with @code{catch}. The argument @var{tag} is used to choose | |
611 | among the various existing return points; it must be @code{eq} to the value | |
612 | specified in the @code{catch}. If multiple return points match @var{tag}, | |
613 | the innermost one is used. | |
614 | ||
615 | The argument @var{value} is used as the value to return from that | |
616 | @code{catch}. | |
617 | ||
618 | @kindex no-catch | |
619 | If no return point is in effect with tag @var{tag}, then a @code{no-catch} | |
620 | error is signaled with data @code{(@var{tag} @var{value})}. | |
621 | @end defun | |
622 | ||
623 | @node Examples of Catch | |
624 | @subsection Examples of @code{catch} and @code{throw} | |
625 | ||
626 | One way to use @code{catch} and @code{throw} is to exit from a doubly | |
627 | nested loop. (In most languages, this would be done with a ``go to.'') | |
628 | Here we compute @code{(foo @var{i} @var{j})} for @var{i} and @var{j} | |
629 | varying from 0 to 9: | |
630 | ||
631 | @example | |
632 | @group | |
633 | (defun search-foo () | |
634 | (catch 'loop | |
635 | (let ((i 0)) | |
636 | (while (< i 10) | |
637 | (let ((j 0)) | |
638 | (while (< j 10) | |
639 | (if (foo i j) | |
640 | (throw 'loop (list i j))) | |
641 | (setq j (1+ j)))) | |
642 | (setq i (1+ i)))))) | |
643 | @end group | |
644 | @end example | |
645 | ||
646 | @noindent | |
647 | If @code{foo} ever returns non-@code{nil}, we stop immediately and return a | |
648 | list of @var{i} and @var{j}. If @code{foo} always returns @code{nil}, the | |
649 | @code{catch} returns normally, and the value is @code{nil}, since that | |
650 | is the result of the @code{while}. | |
651 | ||
652 | Here are two tricky examples, slightly different, showing two | |
653 | return points at once. First, two return points with the same tag, | |
654 | @code{hack}: | |
655 | ||
656 | @example | |
657 | @group | |
658 | (defun catch2 (tag) | |
659 | (catch tag | |
660 | (throw 'hack 'yes))) | |
661 | @result{} catch2 | |
662 | @end group | |
663 | ||
664 | @group | |
665 | (catch 'hack | |
666 | (print (catch2 'hack)) | |
667 | 'no) | |
668 | @print{} yes | |
669 | @result{} no | |
670 | @end group | |
671 | @end example | |
672 | ||
673 | @noindent | |
674 | Since both return points have tags that match the @code{throw}, it goes to | |
675 | the inner one, the one established in @code{catch2}. Therefore, | |
676 | @code{catch2} returns normally with value @code{yes}, and this value is | |
677 | printed. Finally the second body form in the outer @code{catch}, which is | |
678 | @code{'no}, is evaluated and returned from the outer @code{catch}. | |
679 | ||
680 | Now let's change the argument given to @code{catch2}: | |
681 | ||
682 | @example | |
683 | @group | |
684 | (catch 'hack | |
685 | (print (catch2 'quux)) | |
686 | 'no) | |
687 | @result{} yes | |
688 | @end group | |
689 | @end example | |
690 | ||
691 | @noindent | |
692 | We still have two return points, but this time only the outer one has | |
693 | the tag @code{hack}; the inner one has the tag @code{quux} instead. | |
694 | Therefore, @code{throw} makes the outer @code{catch} return the value | |
695 | @code{yes}. The function @code{print} is never called, and the | |
696 | body-form @code{'no} is never evaluated. | |
697 | ||
698 | @node Errors | |
699 | @subsection Errors | |
700 | @cindex errors | |
701 | ||
702 | When Emacs Lisp attempts to evaluate a form that, for some reason, | |
703 | cannot be evaluated, it @dfn{signals} an @dfn{error}. | |
704 | ||
705 | When an error is signaled, Emacs's default reaction is to print an | |
706 | error message and terminate execution of the current command. This is | |
707 | the right thing to do in most cases, such as if you type @kbd{C-f} at | |
708 | the end of the buffer. | |
709 | ||
710 | In complicated programs, simple termination may not be what you want. | |
711 | For example, the program may have made temporary changes in data | |
712 | structures, or created temporary buffers that should be deleted before | |
713 | the program is finished. In such cases, you would use | |
714 | @code{unwind-protect} to establish @dfn{cleanup expressions} to be | |
715 | evaluated in case of error. (@xref{Cleanups}.) Occasionally, you may | |
716 | wish the program to continue execution despite an error in a subroutine. | |
717 | In these cases, you would use @code{condition-case} to establish | |
718 | @dfn{error handlers} to recover control in case of error. | |
719 | ||
720 | Resist the temptation to use error handling to transfer control from | |
721 | one part of the program to another; use @code{catch} and @code{throw} | |
722 | instead. @xref{Catch and Throw}. | |
723 | ||
724 | @menu | |
725 | * Signaling Errors:: How to report an error. | |
726 | * Processing of Errors:: What Emacs does when you report an error. | |
727 | * Handling Errors:: How you can trap errors and continue execution. | |
728 | * Error Symbols:: How errors are classified for trapping them. | |
729 | @end menu | |
730 | ||
731 | @node Signaling Errors | |
732 | @subsubsection How to Signal an Error | |
733 | @cindex signaling errors | |
734 | ||
735 | @dfn{Signaling} an error means beginning error processing. Error | |
736 | processing normally aborts all or part of the running program and | |
737 | returns to a point that is set up to handle the error | |
738 | (@pxref{Processing of Errors}). Here we describe how to signal an | |
739 | error. | |
740 | ||
741 | Most errors are signaled ``automatically'' within Lisp primitives | |
742 | which you call for other purposes, such as if you try to take the | |
743 | @sc{car} of an integer or move forward a character at the end of the | |
744 | buffer. You can also signal errors explicitly with the functions | |
745 | @code{error} and @code{signal}. | |
746 | ||
747 | Quitting, which happens when the user types @kbd{C-g}, is not | |
748 | considered an error, but it is handled almost like an error. | |
749 | @xref{Quitting}. | |
750 | ||
751 | Every error specifies an error message, one way or another. The | |
752 | message should state what is wrong (``File does not exist''), not how | |
753 | things ought to be (``File must exist''). The convention in Emacs | |
754 | Lisp is that error messages should start with a capital letter, but | |
755 | should not end with any sort of punctuation. | |
756 | ||
757 | @defun error format-string &rest args | |
758 | This function signals an error with an error message constructed by | |
759 | applying @code{format} (@pxref{Formatting Strings}) to | |
760 | @var{format-string} and @var{args}. | |
761 | ||
762 | These examples show typical uses of @code{error}: | |
763 | ||
764 | @example | |
765 | @group | |
766 | (error "That is an error -- try something else") | |
767 | @error{} That is an error -- try something else | |
768 | @end group | |
769 | ||
770 | @group | |
771 | (error "You have committed %d errors" 10) | |
772 | @error{} You have committed 10 errors | |
773 | @end group | |
774 | @end example | |
775 | ||
776 | @code{error} works by calling @code{signal} with two arguments: the | |
777 | error symbol @code{error}, and a list containing the string returned by | |
778 | @code{format}. | |
779 | ||
780 | @strong{Warning:} If you want to use your own string as an error message | |
781 | verbatim, don't just write @code{(error @var{string})}. If @var{string} | |
782 | contains @samp{%}, it will be interpreted as a format specifier, with | |
783 | undesirable results. Instead, use @code{(error "%s" @var{string})}. | |
784 | @end defun | |
785 | ||
786 | @defun signal error-symbol data | |
787 | @anchor{Definition of signal} | |
788 | This function signals an error named by @var{error-symbol}. The | |
789 | argument @var{data} is a list of additional Lisp objects relevant to | |
790 | the circumstances of the error. | |
791 | ||
792 | The argument @var{error-symbol} must be an @dfn{error symbol}---a symbol | |
793 | bearing a property @code{error-conditions} whose value is a list of | |
794 | condition names. This is how Emacs Lisp classifies different sorts of | |
795 | errors. @xref{Error Symbols}, for a description of error symbols, | |
796 | error conditions and condition names. | |
797 | ||
798 | If the error is not handled, the two arguments are used in printing | |
799 | the error message. Normally, this error message is provided by the | |
800 | @code{error-message} property of @var{error-symbol}. If @var{data} is | |
801 | non-@code{nil}, this is followed by a colon and a comma separated list | |
802 | of the unevaluated elements of @var{data}. For @code{error}, the | |
803 | error message is the @sc{car} of @var{data} (that must be a string). | |
804 | Subcategories of @code{file-error} are handled specially. | |
805 | ||
806 | The number and significance of the objects in @var{data} depends on | |
59f80054 | 807 | @var{error-symbol}. For example, with a @code{wrong-type-argument} error, |
b8d4c8d0 GM |
808 | there should be two objects in the list: a predicate that describes the type |
809 | that was expected, and the object that failed to fit that type. | |
810 | ||
811 | Both @var{error-symbol} and @var{data} are available to any error | |
812 | handlers that handle the error: @code{condition-case} binds a local | |
813 | variable to a list of the form @code{(@var{error-symbol} .@: | |
814 | @var{data})} (@pxref{Handling Errors}). | |
815 | ||
816 | The function @code{signal} never returns (though in older Emacs versions | |
817 | it could sometimes return). | |
818 | ||
819 | @smallexample | |
820 | @group | |
821 | (signal 'wrong-number-of-arguments '(x y)) | |
822 | @error{} Wrong number of arguments: x, y | |
823 | @end group | |
824 | ||
825 | @group | |
826 | (signal 'no-such-error '("My unknown error condition")) | |
827 | @error{} peculiar error: "My unknown error condition" | |
828 | @end group | |
829 | @end smallexample | |
830 | @end defun | |
831 | ||
832 | @cindex CL note---no continuable errors | |
833 | @quotation | |
834 | @b{Common Lisp note:} Emacs Lisp has nothing like the Common Lisp | |
835 | concept of continuable errors. | |
836 | @end quotation | |
837 | ||
838 | @node Processing of Errors | |
839 | @subsubsection How Emacs Processes Errors | |
840 | ||
841 | When an error is signaled, @code{signal} searches for an active | |
842 | @dfn{handler} for the error. A handler is a sequence of Lisp | |
843 | expressions designated to be executed if an error happens in part of the | |
844 | Lisp program. If the error has an applicable handler, the handler is | |
845 | executed, and control resumes following the handler. The handler | |
846 | executes in the environment of the @code{condition-case} that | |
847 | established it; all functions called within that @code{condition-case} | |
848 | have already been exited, and the handler cannot return to them. | |
849 | ||
850 | If there is no applicable handler for the error, it terminates the | |
851 | current command and returns control to the editor command loop. (The | |
852 | command loop has an implicit handler for all kinds of errors.) The | |
853 | command loop's handler uses the error symbol and associated data to | |
854 | print an error message. You can use the variable | |
855 | @code{command-error-function} to control how this is done: | |
856 | ||
857 | @defvar command-error-function | |
858 | This variable, if non-@code{nil}, specifies a function to use to | |
859 | handle errors that return control to the Emacs command loop. The | |
860 | function should take three arguments: @var{data}, a list of the same | |
861 | form that @code{condition-case} would bind to its variable; | |
862 | @var{context}, a string describing the situation in which the error | |
863 | occurred, or (more often) @code{nil}; and @var{caller}, the Lisp | |
864 | function which called the primitive that signaled the error. | |
865 | @end defvar | |
866 | ||
867 | @cindex @code{debug-on-error} use | |
868 | An error that has no explicit handler may call the Lisp debugger. The | |
869 | debugger is enabled if the variable @code{debug-on-error} (@pxref{Error | |
870 | Debugging}) is non-@code{nil}. Unlike error handlers, the debugger runs | |
871 | in the environment of the error, so that you can examine values of | |
872 | variables precisely as they were at the time of the error. | |
873 | ||
874 | @node Handling Errors | |
875 | @subsubsection Writing Code to Handle Errors | |
876 | @cindex error handler | |
877 | @cindex handling errors | |
878 | ||
879 | The usual effect of signaling an error is to terminate the command | |
880 | that is running and return immediately to the Emacs editor command loop. | |
881 | You can arrange to trap errors occurring in a part of your program by | |
882 | establishing an error handler, with the special form | |
883 | @code{condition-case}. A simple example looks like this: | |
884 | ||
885 | @example | |
886 | @group | |
887 | (condition-case nil | |
888 | (delete-file filename) | |
889 | (error nil)) | |
890 | @end group | |
891 | @end example | |
892 | ||
893 | @noindent | |
894 | This deletes the file named @var{filename}, catching any error and | |
a33a1f2a EZ |
895 | returning @code{nil} if an error occurs@footnote{ |
896 | Actually, you should use @code{ignore-errors} in such a simple case; | |
897 | see below.}. | |
b8d4c8d0 GM |
898 | |
899 | The @code{condition-case} construct is often used to trap errors that | |
900 | are predictable, such as failure to open a file in a call to | |
901 | @code{insert-file-contents}. It is also used to trap errors that are | |
902 | totally unpredictable, such as when the program evaluates an expression | |
903 | read from the user. | |
904 | ||
905 | The second argument of @code{condition-case} is called the | |
906 | @dfn{protected form}. (In the example above, the protected form is a | |
907 | call to @code{delete-file}.) The error handlers go into effect when | |
908 | this form begins execution and are deactivated when this form returns. | |
909 | They remain in effect for all the intervening time. In particular, they | |
910 | are in effect during the execution of functions called by this form, in | |
911 | their subroutines, and so on. This is a good thing, since, strictly | |
912 | speaking, errors can be signaled only by Lisp primitives (including | |
913 | @code{signal} and @code{error}) called by the protected form, not by the | |
914 | protected form itself. | |
915 | ||
916 | The arguments after the protected form are handlers. Each handler | |
917 | lists one or more @dfn{condition names} (which are symbols) to specify | |
918 | which errors it will handle. The error symbol specified when an error | |
919 | is signaled also defines a list of condition names. A handler applies | |
920 | to an error if they have any condition names in common. In the example | |
921 | above, there is one handler, and it specifies one condition name, | |
922 | @code{error}, which covers all errors. | |
923 | ||
924 | The search for an applicable handler checks all the established handlers | |
925 | starting with the most recently established one. Thus, if two nested | |
926 | @code{condition-case} forms offer to handle the same error, the inner of | |
927 | the two gets to handle it. | |
928 | ||
929 | If an error is handled by some @code{condition-case} form, this | |
930 | ordinarily prevents the debugger from being run, even if | |
931 | @code{debug-on-error} says this error should invoke the debugger. | |
932 | ||
933 | If you want to be able to debug errors that are caught by a | |
934 | @code{condition-case}, set the variable @code{debug-on-signal} to a | |
935 | non-@code{nil} value. You can also specify that a particular handler | |
936 | should let the debugger run first, by writing @code{debug} among the | |
937 | conditions, like this: | |
938 | ||
939 | @example | |
940 | @group | |
941 | (condition-case nil | |
942 | (delete-file filename) | |
943 | ((debug error) nil)) | |
944 | @end group | |
945 | @end example | |
946 | ||
947 | @noindent | |
948 | The effect of @code{debug} here is only to prevent | |
949 | @code{condition-case} from suppressing the call to the debugger. Any | |
950 | given error will invoke the debugger only if @code{debug-on-error} and | |
951 | the other usual filtering mechanisms say it should. @xref{Error Debugging}. | |
952 | ||
953 | Once Emacs decides that a certain handler handles the error, it | |
954 | returns control to that handler. To do so, Emacs unbinds all variable | |
955 | bindings made by binding constructs that are being exited, and | |
956 | executes the cleanups of all @code{unwind-protect} forms that are | |
957 | being exited. Once control arrives at the handler, the body of the | |
958 | handler executes normally. | |
959 | ||
960 | After execution of the handler body, execution returns from the | |
961 | @code{condition-case} form. Because the protected form is exited | |
962 | completely before execution of the handler, the handler cannot resume | |
963 | execution at the point of the error, nor can it examine variable | |
964 | bindings that were made within the protected form. All it can do is | |
965 | clean up and proceed. | |
966 | ||
967 | Error signaling and handling have some resemblance to @code{throw} and | |
968 | @code{catch} (@pxref{Catch and Throw}), but they are entirely separate | |
969 | facilities. An error cannot be caught by a @code{catch}, and a | |
970 | @code{throw} cannot be handled by an error handler (though using | |
971 | @code{throw} when there is no suitable @code{catch} signals an error | |
972 | that can be handled). | |
973 | ||
974 | @defspec condition-case var protected-form handlers@dots{} | |
975 | This special form establishes the error handlers @var{handlers} around | |
976 | the execution of @var{protected-form}. If @var{protected-form} executes | |
977 | without error, the value it returns becomes the value of the | |
978 | @code{condition-case} form; in this case, the @code{condition-case} has | |
979 | no effect. The @code{condition-case} form makes a difference when an | |
980 | error occurs during @var{protected-form}. | |
981 | ||
982 | Each of the @var{handlers} is a list of the form @code{(@var{conditions} | |
983 | @var{body}@dots{})}. Here @var{conditions} is an error condition name | |
984 | to be handled, or a list of condition names (which can include @code{debug} | |
985 | to allow the debugger to run before the handler); @var{body} is one or more | |
986 | Lisp expressions to be executed when this handler handles an error. | |
987 | Here are examples of handlers: | |
988 | ||
989 | @smallexample | |
990 | @group | |
991 | (error nil) | |
992 | ||
993 | (arith-error (message "Division by zero")) | |
994 | ||
995 | ((arith-error file-error) | |
996 | (message | |
997 | "Either division by zero or failure to open a file")) | |
998 | @end group | |
999 | @end smallexample | |
1000 | ||
1001 | Each error that occurs has an @dfn{error symbol} that describes what | |
1002 | kind of error it is. The @code{error-conditions} property of this | |
1003 | symbol is a list of condition names (@pxref{Error Symbols}). Emacs | |
1004 | searches all the active @code{condition-case} forms for a handler that | |
1005 | specifies one or more of these condition names; the innermost matching | |
1006 | @code{condition-case} handles the error. Within this | |
1007 | @code{condition-case}, the first applicable handler handles the error. | |
1008 | ||
1009 | After executing the body of the handler, the @code{condition-case} | |
1010 | returns normally, using the value of the last form in the handler body | |
1011 | as the overall value. | |
1012 | ||
1013 | @cindex error description | |
1014 | The argument @var{var} is a variable. @code{condition-case} does not | |
1015 | bind this variable when executing the @var{protected-form}, only when it | |
1016 | handles an error. At that time, it binds @var{var} locally to an | |
1017 | @dfn{error description}, which is a list giving the particulars of the | |
1018 | error. The error description has the form @code{(@var{error-symbol} | |
1019 | . @var{data})}. The handler can refer to this list to decide what to | |
1020 | do. For example, if the error is for failure opening a file, the file | |
1021 | name is the second element of @var{data}---the third element of the | |
1022 | error description. | |
1023 | ||
1024 | If @var{var} is @code{nil}, that means no variable is bound. Then the | |
1025 | error symbol and associated data are not available to the handler. | |
7a1831cf EZ |
1026 | |
1027 | @cindex rethrow a signal | |
1028 | Sometimes it is necessary to re-throw a signal caught by | |
1029 | @code{condition-case}, for some outer-level handler to catch. Here's | |
1030 | how to do that: | |
1031 | ||
1032 | @smallexample | |
1033 | (signal (car err) (cdr err)) | |
1034 | @end smallexample | |
1035 | ||
1036 | @noindent | |
1037 | where @code{err} is the error description variable, the first argument | |
1038 | to @code{condition-case} whose error condition you want to re-throw. | |
1039 | @xref{Definition of signal}. | |
b8d4c8d0 GM |
1040 | @end defspec |
1041 | ||
ee301a7a | 1042 | @defun error-message-string error-descriptor |
b8d4c8d0 GM |
1043 | This function returns the error message string for a given error |
1044 | descriptor. It is useful if you want to handle an error by printing the | |
1045 | usual error message for that error. @xref{Definition of signal}. | |
1046 | @end defun | |
1047 | ||
1048 | @cindex @code{arith-error} example | |
1049 | Here is an example of using @code{condition-case} to handle the error | |
1050 | that results from dividing by zero. The handler displays the error | |
1051 | message (but without a beep), then returns a very large number. | |
1052 | ||
1053 | @smallexample | |
1054 | @group | |
1055 | (defun safe-divide (dividend divisor) | |
1056 | (condition-case err | |
1057 | ;; @r{Protected form.} | |
1058 | (/ dividend divisor) | |
1059 | @end group | |
1060 | @group | |
1061 | ;; @r{The handler.} | |
1062 | (arith-error ; @r{Condition.} | |
1063 | ;; @r{Display the usual message for this error.} | |
1064 | (message "%s" (error-message-string err)) | |
1065 | 1000000))) | |
1066 | @result{} safe-divide | |
1067 | @end group | |
1068 | ||
1069 | @group | |
1070 | (safe-divide 5 0) | |
1071 | @print{} Arithmetic error: (arith-error) | |
1072 | @result{} 1000000 | |
1073 | @end group | |
1074 | @end smallexample | |
1075 | ||
1076 | @noindent | |
1077 | The handler specifies condition name @code{arith-error} so that it will handle only division-by-zero errors. Other kinds of errors will not be handled, at least not by this @code{condition-case}. Thus, | |
1078 | ||
1079 | @smallexample | |
1080 | @group | |
1081 | (safe-divide nil 3) | |
1082 | @error{} Wrong type argument: number-or-marker-p, nil | |
1083 | @end group | |
1084 | @end smallexample | |
1085 | ||
1086 | Here is a @code{condition-case} that catches all kinds of errors, | |
1087 | including those signaled with @code{error}: | |
1088 | ||
1089 | @smallexample | |
1090 | @group | |
1091 | (setq baz 34) | |
1092 | @result{} 34 | |
1093 | @end group | |
1094 | ||
1095 | @group | |
1096 | (condition-case err | |
1097 | (if (eq baz 35) | |
1098 | t | |
1099 | ;; @r{This is a call to the function @code{error}.} | |
1100 | (error "Rats! The variable %s was %s, not 35" 'baz baz)) | |
1101 | ;; @r{This is the handler; it is not a form.} | |
1102 | (error (princ (format "The error was: %s" err)) | |
1103 | 2)) | |
1104 | @print{} The error was: (error "Rats! The variable baz was 34, not 35") | |
1105 | @result{} 2 | |
1106 | @end group | |
1107 | @end smallexample | |
1108 | ||
a33a1f2a EZ |
1109 | @defmac ignore-errors body@dots{} |
1110 | This construct executes @var{body}, ignoring any errors that occur | |
1111 | during its execution. If the execution is without error, | |
1112 | @code{ignore-errors} returns the value of the last form in @var{body}; | |
1113 | otherwise, it returns @code{nil}. | |
1114 | ||
1115 | Here's the example at the beginning of this subsection rewritten using | |
1116 | @code{ignore-errors}: | |
1117 | ||
1118 | @smallexample | |
1119 | @group | |
1120 | (ignore-errors | |
1121 | (delete-file filename)) | |
1122 | @end group | |
1123 | @end smallexample | |
1124 | @end defmac | |
1125 | ||
7a1831cf | 1126 | |
b8d4c8d0 GM |
1127 | @node Error Symbols |
1128 | @subsubsection Error Symbols and Condition Names | |
1129 | @cindex error symbol | |
1130 | @cindex error name | |
1131 | @cindex condition name | |
1132 | @cindex user-defined error | |
1133 | @kindex error-conditions | |
1134 | ||
1135 | When you signal an error, you specify an @dfn{error symbol} to specify | |
1136 | the kind of error you have in mind. Each error has one and only one | |
1137 | error symbol to categorize it. This is the finest classification of | |
1138 | errors defined by the Emacs Lisp language. | |
1139 | ||
1140 | These narrow classifications are grouped into a hierarchy of wider | |
1141 | classes called @dfn{error conditions}, identified by @dfn{condition | |
1142 | names}. The narrowest such classes belong to the error symbols | |
1143 | themselves: each error symbol is also a condition name. There are also | |
1144 | condition names for more extensive classes, up to the condition name | |
1145 | @code{error} which takes in all kinds of errors (but not @code{quit}). | |
1146 | Thus, each error has one or more condition names: @code{error}, the | |
1147 | error symbol if that is distinct from @code{error}, and perhaps some | |
1148 | intermediate classifications. | |
1149 | ||
1150 | In order for a symbol to be an error symbol, it must have an | |
1151 | @code{error-conditions} property which gives a list of condition names. | |
1152 | This list defines the conditions that this kind of error belongs to. | |
1153 | (The error symbol itself, and the symbol @code{error}, should always be | |
1154 | members of this list.) Thus, the hierarchy of condition names is | |
1155 | defined by the @code{error-conditions} properties of the error symbols. | |
1156 | Because quitting is not considered an error, the value of the | |
1157 | @code{error-conditions} property of @code{quit} is just @code{(quit)}. | |
1158 | ||
1159 | @cindex peculiar error | |
1160 | In addition to the @code{error-conditions} list, the error symbol | |
1161 | should have an @code{error-message} property whose value is a string to | |
1162 | be printed when that error is signaled but not handled. If the | |
1163 | error symbol has no @code{error-message} property or if the | |
1164 | @code{error-message} property exists, but is not a string, the error | |
1165 | message @samp{peculiar error} is used. @xref{Definition of signal}. | |
1166 | ||
1167 | Here is how we define a new error symbol, @code{new-error}: | |
1168 | ||
1169 | @example | |
1170 | @group | |
1171 | (put 'new-error | |
1172 | 'error-conditions | |
1173 | '(error my-own-errors new-error)) | |
1174 | @result{} (error my-own-errors new-error) | |
1175 | @end group | |
1176 | @group | |
1177 | (put 'new-error 'error-message "A new error") | |
1178 | @result{} "A new error" | |
1179 | @end group | |
1180 | @end example | |
1181 | ||
1182 | @noindent | |
1183 | This error has three condition names: @code{new-error}, the narrowest | |
1184 | classification; @code{my-own-errors}, which we imagine is a wider | |
1185 | classification; and @code{error}, which is the widest of all. | |
1186 | ||
1187 | The error string should start with a capital letter but it should | |
1188 | not end with a period. This is for consistency with the rest of Emacs. | |
1189 | ||
1190 | Naturally, Emacs will never signal @code{new-error} on its own; only | |
1191 | an explicit call to @code{signal} (@pxref{Definition of signal}) in | |
1192 | your code can do this: | |
1193 | ||
1194 | @example | |
1195 | @group | |
1196 | (signal 'new-error '(x y)) | |
1197 | @error{} A new error: x, y | |
1198 | @end group | |
1199 | @end example | |
1200 | ||
1201 | This error can be handled through any of the three condition names. | |
1202 | This example handles @code{new-error} and any other errors in the class | |
1203 | @code{my-own-errors}: | |
1204 | ||
1205 | @example | |
1206 | @group | |
1207 | (condition-case foo | |
1208 | (bar nil t) | |
1209 | (my-own-errors nil)) | |
1210 | @end group | |
1211 | @end example | |
1212 | ||
1213 | The significant way that errors are classified is by their condition | |
1214 | names---the names used to match errors with handlers. An error symbol | |
1215 | serves only as a convenient way to specify the intended error message | |
1216 | and list of condition names. It would be cumbersome to give | |
1217 | @code{signal} a list of condition names rather than one error symbol. | |
1218 | ||
1219 | By contrast, using only error symbols without condition names would | |
1220 | seriously decrease the power of @code{condition-case}. Condition names | |
1221 | make it possible to categorize errors at various levels of generality | |
1222 | when you write an error handler. Using error symbols alone would | |
1223 | eliminate all but the narrowest level of classification. | |
1224 | ||
1225 | @xref{Standard Errors}, for a list of all the standard error symbols | |
1226 | and their conditions. | |
1227 | ||
1228 | @node Cleanups | |
1229 | @subsection Cleaning Up from Nonlocal Exits | |
1230 | ||
1231 | The @code{unwind-protect} construct is essential whenever you | |
1232 | temporarily put a data structure in an inconsistent state; it permits | |
1233 | you to make the data consistent again in the event of an error or | |
1234 | throw. (Another more specific cleanup construct that is used only for | |
1235 | changes in buffer contents is the atomic change group; @ref{Atomic | |
1236 | Changes}.) | |
1237 | ||
1238 | @defspec unwind-protect body-form cleanup-forms@dots{} | |
1239 | @cindex cleanup forms | |
1240 | @cindex protected forms | |
1241 | @cindex error cleanup | |
1242 | @cindex unwinding | |
1243 | @code{unwind-protect} executes @var{body-form} with a guarantee that | |
1244 | the @var{cleanup-forms} will be evaluated if control leaves | |
1245 | @var{body-form}, no matter how that happens. @var{body-form} may | |
1246 | complete normally, or execute a @code{throw} out of the | |
1247 | @code{unwind-protect}, or cause an error; in all cases, the | |
1248 | @var{cleanup-forms} will be evaluated. | |
1249 | ||
1250 | If @var{body-form} finishes normally, @code{unwind-protect} returns the | |
1251 | value of @var{body-form}, after it evaluates the @var{cleanup-forms}. | |
1252 | If @var{body-form} does not finish, @code{unwind-protect} does not | |
1253 | return any value in the normal sense. | |
1254 | ||
1255 | Only @var{body-form} is protected by the @code{unwind-protect}. If any | |
1256 | of the @var{cleanup-forms} themselves exits nonlocally (via a | |
1257 | @code{throw} or an error), @code{unwind-protect} is @emph{not} | |
1258 | guaranteed to evaluate the rest of them. If the failure of one of the | |
1259 | @var{cleanup-forms} has the potential to cause trouble, then protect | |
1260 | it with another @code{unwind-protect} around that form. | |
1261 | ||
1262 | The number of currently active @code{unwind-protect} forms counts, | |
1263 | together with the number of local variable bindings, against the limit | |
1264 | @code{max-specpdl-size} (@pxref{Definition of max-specpdl-size,, Local | |
1265 | Variables}). | |
1266 | @end defspec | |
1267 | ||
1268 | For example, here we make an invisible buffer for temporary use, and | |
1269 | make sure to kill it before finishing: | |
1270 | ||
1271 | @smallexample | |
1272 | @group | |
c57008f6 SM |
1273 | (let ((buffer (get-buffer-create " *temp*"))) |
1274 | (with-current-buffer buffer | |
b8d4c8d0 GM |
1275 | (unwind-protect |
1276 | @var{body-form} | |
1277 | (kill-buffer buffer)))) | |
1278 | @end group | |
1279 | @end smallexample | |
1280 | ||
1281 | @noindent | |
1282 | You might think that we could just as well write @code{(kill-buffer | |
1283 | (current-buffer))} and dispense with the variable @code{buffer}. | |
1284 | However, the way shown above is safer, if @var{body-form} happens to | |
1285 | get an error after switching to a different buffer! (Alternatively, | |
c57008f6 | 1286 | you could write a @code{save-current-buffer} around @var{body-form}, |
b8d4c8d0 GM |
1287 | to ensure that the temporary buffer becomes current again in time to |
1288 | kill it.) | |
1289 | ||
1290 | Emacs includes a standard macro called @code{with-temp-buffer} which | |
1291 | expands into more or less the code shown above (@pxref{Definition of | |
1292 | with-temp-buffer,, Current Buffer}). Several of the macros defined in | |
1293 | this manual use @code{unwind-protect} in this way. | |
1294 | ||
1295 | @findex ftp-login | |
1296 | Here is an actual example derived from an FTP package. It creates a | |
1297 | process (@pxref{Processes}) to try to establish a connection to a remote | |
1298 | machine. As the function @code{ftp-login} is highly susceptible to | |
1299 | numerous problems that the writer of the function cannot anticipate, it | |
1300 | is protected with a form that guarantees deletion of the process in the | |
1301 | event of failure. Otherwise, Emacs might fill up with useless | |
1302 | subprocesses. | |
1303 | ||
1304 | @smallexample | |
1305 | @group | |
1306 | (let ((win nil)) | |
1307 | (unwind-protect | |
1308 | (progn | |
1309 | (setq process (ftp-setup-buffer host file)) | |
1310 | (if (setq win (ftp-login process host user password)) | |
1311 | (message "Logged in") | |
1312 | (error "Ftp login failed"))) | |
1313 | (or win (and process (delete-process process))))) | |
1314 | @end group | |
1315 | @end smallexample | |
1316 | ||
1317 | This example has a small bug: if the user types @kbd{C-g} to | |
1318 | quit, and the quit happens immediately after the function | |
1319 | @code{ftp-setup-buffer} returns but before the variable @code{process} is | |
1320 | set, the process will not be killed. There is no easy way to fix this bug, | |
1321 | but at least it is very unlikely. | |
1322 | ||
1323 | @ignore | |
1324 | arch-tag: 8abc30d4-4d3a-47f9-b908-e9e971c18c6d | |
1325 | @end ignore |