Mercurial > hg > xemacs-beta
comparison src/data.c @ 428:3ecd8885ac67 r21-2-22
Import from CVS: tag r21-2-22
author | cvs |
---|---|
date | Mon, 13 Aug 2007 11:28:15 +0200 |
parents | |
children | 080151679be2 |
comparison
equal
deleted
inserted
replaced
427:0a0253eac470 | 428:3ecd8885ac67 |
---|---|
1 /* Primitive operations on Lisp data types for XEmacs Lisp interpreter. | |
2 Copyright (C) 1985, 1986, 1988, 1992, 1993, 1994, 1995 | |
3 Free Software Foundation, Inc. | |
4 | |
5 This file is part of XEmacs. | |
6 | |
7 XEmacs is free software; you can redistribute it and/or modify it | |
8 under the terms of the GNU General Public License as published by the | |
9 Free Software Foundation; either version 2, or (at your option) any | |
10 later version. | |
11 | |
12 XEmacs is distributed in the hope that it will be useful, but WITHOUT | |
13 ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or | |
14 FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License | |
15 for more details. | |
16 | |
17 You should have received a copy of the GNU General Public License | |
18 along with XEmacs; see the file COPYING. If not, write to | |
19 the Free Software Foundation, Inc., 59 Temple Place - Suite 330, | |
20 Boston, MA 02111-1307, USA. */ | |
21 | |
22 /* Synched up with: Mule 2.0, FSF 19.30. Some of FSF's data.c is in | |
23 XEmacs' symbols.c. */ | |
24 | |
25 /* This file has been Mule-ized. */ | |
26 | |
27 #include <config.h> | |
28 #include "lisp.h" | |
29 | |
30 #include "buffer.h" | |
31 #include "bytecode.h" | |
32 #include "syssignal.h" | |
33 | |
34 #ifdef LISP_FLOAT_TYPE | |
35 /* Need to define a differentiating symbol -- see sysfloat.h */ | |
36 # define THIS_FILENAME data_c | |
37 # include "sysfloat.h" | |
38 #endif /* LISP_FLOAT_TYPE */ | |
39 | |
40 Lisp_Object Qnil, Qt, Qquote, Qlambda, Qunbound; | |
41 Lisp_Object Qerror_conditions, Qerror_message; | |
42 Lisp_Object Qerror, Qquit, Qwrong_type_argument, Qargs_out_of_range; | |
43 Lisp_Object Qvoid_variable, Qcyclic_variable_indirection; | |
44 Lisp_Object Qvoid_function, Qcyclic_function_indirection; | |
45 Lisp_Object Qsetting_constant, Qinvalid_read_syntax; | |
46 Lisp_Object Qmalformed_list, Qmalformed_property_list; | |
47 Lisp_Object Qcircular_list, Qcircular_property_list; | |
48 Lisp_Object Qinvalid_function, Qwrong_number_of_arguments, Qno_catch; | |
49 Lisp_Object Qio_error, Qend_of_file; | |
50 Lisp_Object Qarith_error, Qrange_error, Qdomain_error; | |
51 Lisp_Object Qsingularity_error, Qoverflow_error, Qunderflow_error; | |
52 Lisp_Object Qbeginning_of_buffer, Qend_of_buffer, Qbuffer_read_only; | |
53 Lisp_Object Qintegerp, Qnatnump, Qsymbolp; | |
54 Lisp_Object Qlistp, Qtrue_list_p, Qweak_listp; | |
55 Lisp_Object Qconsp, Qsubrp; | |
56 Lisp_Object Qcharacterp, Qstringp, Qarrayp, Qsequencep, Qvectorp; | |
57 Lisp_Object Qchar_or_string_p, Qmarkerp, Qinteger_or_marker_p, Qbufferp; | |
58 Lisp_Object Qinteger_or_char_p, Qinteger_char_or_marker_p; | |
59 Lisp_Object Qnumberp, Qnumber_char_or_marker_p; | |
60 Lisp_Object Qbit_vectorp, Qbitp, Qcdr; | |
61 | |
62 Lisp_Object Qfloatp; | |
63 | |
64 #ifdef DEBUG_XEMACS | |
65 | |
66 int debug_issue_ebola_notices; | |
67 | |
68 int debug_ebola_backtrace_length; | |
69 | |
70 int | |
71 eq_with_ebola_notice (Lisp_Object obj1, Lisp_Object obj2) | |
72 { | |
73 if (debug_issue_ebola_notices | |
74 && ((CHARP (obj1) && INTP (obj2)) || (CHARP (obj2) && INTP (obj1)))) | |
75 { | |
76 /* #### It would be really nice if this were a proper warning | |
77 instead of brain-dead print ro Qexternal_debugging_output. */ | |
78 write_c_string ("Comparison between integer and character is constant nil (", | |
79 Qexternal_debugging_output); | |
80 Fprinc (obj1, Qexternal_debugging_output); | |
81 write_c_string (" and ", Qexternal_debugging_output); | |
82 Fprinc (obj2, Qexternal_debugging_output); | |
83 write_c_string (")\n", Qexternal_debugging_output); | |
84 debug_short_backtrace (debug_ebola_backtrace_length); | |
85 } | |
86 return EQ (obj1, obj2); | |
87 } | |
88 | |
89 #endif /* DEBUG_XEMACS */ | |
90 | |
91 | |
92 | |
93 Lisp_Object | |
94 wrong_type_argument (Lisp_Object predicate, Lisp_Object value) | |
95 { | |
96 /* This function can GC */ | |
97 REGISTER Lisp_Object tem; | |
98 do | |
99 { | |
100 value = Fsignal (Qwrong_type_argument, list2 (predicate, value)); | |
101 tem = call1 (predicate, value); | |
102 } | |
103 while (NILP (tem)); | |
104 return value; | |
105 } | |
106 | |
107 DOESNT_RETURN | |
108 dead_wrong_type_argument (Lisp_Object predicate, Lisp_Object value) | |
109 { | |
110 signal_error (Qwrong_type_argument, list2 (predicate, value)); | |
111 } | |
112 | |
113 DEFUN ("wrong-type-argument", Fwrong_type_argument, 2, 2, 0, /* | |
114 Signal an error until the correct type value is given by the user. | |
115 This function loops, signalling a continuable `wrong-type-argument' error | |
116 with PREDICATE and VALUE as the data associated with the error and then | |
117 calling PREDICATE on the returned value, until the value gotten satisfies | |
118 PREDICATE. At that point, the gotten value is returned. | |
119 */ | |
120 (predicate, value)) | |
121 { | |
122 return wrong_type_argument (predicate, value); | |
123 } | |
124 | |
125 DOESNT_RETURN | |
126 c_write_error (Lisp_Object obj) | |
127 { | |
128 signal_simple_error ("Attempt to modify read-only object (c)", obj); | |
129 } | |
130 | |
131 DOESNT_RETURN | |
132 lisp_write_error (Lisp_Object obj) | |
133 { | |
134 signal_simple_error ("Attempt to modify read-only object (lisp)", obj); | |
135 } | |
136 | |
137 DOESNT_RETURN | |
138 args_out_of_range (Lisp_Object a1, Lisp_Object a2) | |
139 { | |
140 signal_error (Qargs_out_of_range, list2 (a1, a2)); | |
141 } | |
142 | |
143 DOESNT_RETURN | |
144 args_out_of_range_3 (Lisp_Object a1, Lisp_Object a2, Lisp_Object a3) | |
145 { | |
146 signal_error (Qargs_out_of_range, list3 (a1, a2, a3)); | |
147 } | |
148 | |
149 void | |
150 check_int_range (EMACS_INT val, EMACS_INT min, EMACS_INT max) | |
151 { | |
152 if (val < min || val > max) | |
153 args_out_of_range_3 (make_int (val), make_int (min), make_int (max)); | |
154 } | |
155 | |
156 /* On some machines, XINT needs a temporary location. | |
157 Here it is, in case it is needed. */ | |
158 | |
159 EMACS_INT sign_extend_temp; | |
160 | |
161 /* On a few machines, XINT can only be done by calling this. */ | |
162 /* XEmacs: only used by m/convex.h */ | |
163 EMACS_INT sign_extend_lisp_int (EMACS_INT num); | |
164 EMACS_INT | |
165 sign_extend_lisp_int (EMACS_INT num) | |
166 { | |
167 if (num & (1L << (VALBITS - 1))) | |
168 return num | ((-1L) << VALBITS); | |
169 else | |
170 return num & ((1L << VALBITS) - 1); | |
171 } | |
172 | |
173 | |
174 /* Data type predicates */ | |
175 | |
176 DEFUN ("eq", Feq, 2, 2, 0, /* | |
177 Return t if the two args are the same Lisp object. | |
178 */ | |
179 (obj1, obj2)) | |
180 { | |
181 return EQ_WITH_EBOLA_NOTICE (obj1, obj2) ? Qt : Qnil; | |
182 } | |
183 | |
184 DEFUN ("old-eq", Fold_eq, 2, 2, 0, /* | |
185 Return t if the two args are (in most cases) the same Lisp object. | |
186 | |
187 Special kludge: A character is considered `old-eq' to its equivalent integer | |
188 even though they are not the same object and are in fact of different | |
189 types. This is ABSOLUTELY AND UTTERLY HORRENDOUS but is necessary to | |
190 preserve byte-code compatibility with v19. This kludge is known as the | |
191 \"char-int confoundance disease\" and appears in a number of other | |
192 functions with `old-foo' equivalents. | |
193 | |
194 Do not use this function! | |
195 */ | |
196 (obj1, obj2)) | |
197 { | |
198 /* #### blasphemy */ | |
199 return HACKEQ_UNSAFE (obj1, obj2) ? Qt : Qnil; | |
200 } | |
201 | |
202 DEFUN ("null", Fnull, 1, 1, 0, /* | |
203 Return t if OBJECT is nil. | |
204 */ | |
205 (object)) | |
206 { | |
207 return NILP (object) ? Qt : Qnil; | |
208 } | |
209 | |
210 DEFUN ("consp", Fconsp, 1, 1, 0, /* | |
211 Return t if OBJECT is a cons cell. `nil' is not a cons cell. | |
212 */ | |
213 (object)) | |
214 { | |
215 return CONSP (object) ? Qt : Qnil; | |
216 } | |
217 | |
218 DEFUN ("atom", Fatom, 1, 1, 0, /* | |
219 Return t if OBJECT is not a cons cell. `nil' is not a cons cell. | |
220 */ | |
221 (object)) | |
222 { | |
223 return CONSP (object) ? Qnil : Qt; | |
224 } | |
225 | |
226 DEFUN ("listp", Flistp, 1, 1, 0, /* | |
227 Return t if OBJECT is a list. `nil' is a list. | |
228 */ | |
229 (object)) | |
230 { | |
231 return LISTP (object) ? Qt : Qnil; | |
232 } | |
233 | |
234 DEFUN ("nlistp", Fnlistp, 1, 1, 0, /* | |
235 Return t if OBJECT is not a list. `nil' is a list. | |
236 */ | |
237 (object)) | |
238 { | |
239 return LISTP (object) ? Qnil : Qt; | |
240 } | |
241 | |
242 DEFUN ("true-list-p", Ftrue_list_p, 1, 1, 0, /* | |
243 Return t if OBJECT is a non-dotted, i.e. nil-terminated, list. | |
244 */ | |
245 (object)) | |
246 { | |
247 return TRUE_LIST_P (object) ? Qt : Qnil; | |
248 } | |
249 | |
250 DEFUN ("symbolp", Fsymbolp, 1, 1, 0, /* | |
251 Return t if OBJECT is a symbol. | |
252 */ | |
253 (object)) | |
254 { | |
255 return SYMBOLP (object) ? Qt : Qnil; | |
256 } | |
257 | |
258 DEFUN ("keywordp", Fkeywordp, 1, 1, 0, /* | |
259 Return t if OBJECT is a keyword. | |
260 */ | |
261 (object)) | |
262 { | |
263 return KEYWORDP (object) ? Qt : Qnil; | |
264 } | |
265 | |
266 DEFUN ("vectorp", Fvectorp, 1, 1, 0, /* | |
267 Return t if OBJECT is a vector. | |
268 */ | |
269 (object)) | |
270 { | |
271 return VECTORP (object) ? Qt : Qnil; | |
272 } | |
273 | |
274 DEFUN ("bit-vector-p", Fbit_vector_p, 1, 1, 0, /* | |
275 Return t if OBJECT is a bit vector. | |
276 */ | |
277 (object)) | |
278 { | |
279 return BIT_VECTORP (object) ? Qt : Qnil; | |
280 } | |
281 | |
282 DEFUN ("stringp", Fstringp, 1, 1, 0, /* | |
283 Return t if OBJECT is a string. | |
284 */ | |
285 (object)) | |
286 { | |
287 return STRINGP (object) ? Qt : Qnil; | |
288 } | |
289 | |
290 DEFUN ("arrayp", Farrayp, 1, 1, 0, /* | |
291 Return t if OBJECT is an array (string, vector, or bit vector). | |
292 */ | |
293 (object)) | |
294 { | |
295 return (VECTORP (object) || | |
296 STRINGP (object) || | |
297 BIT_VECTORP (object)) | |
298 ? Qt : Qnil; | |
299 } | |
300 | |
301 DEFUN ("sequencep", Fsequencep, 1, 1, 0, /* | |
302 Return t if OBJECT is a sequence (list or array). | |
303 */ | |
304 (object)) | |
305 { | |
306 return (LISTP (object) || | |
307 VECTORP (object) || | |
308 STRINGP (object) || | |
309 BIT_VECTORP (object)) | |
310 ? Qt : Qnil; | |
311 } | |
312 | |
313 DEFUN ("markerp", Fmarkerp, 1, 1, 0, /* | |
314 Return t if OBJECT is a marker (editor pointer). | |
315 */ | |
316 (object)) | |
317 { | |
318 return MARKERP (object) ? Qt : Qnil; | |
319 } | |
320 | |
321 DEFUN ("subrp", Fsubrp, 1, 1, 0, /* | |
322 Return t if OBJECT is a built-in function. | |
323 */ | |
324 (object)) | |
325 { | |
326 return SUBRP (object) ? Qt : Qnil; | |
327 } | |
328 | |
329 DEFUN ("subr-min-args", Fsubr_min_args, 1, 1, 0, /* | |
330 Return minimum number of args built-in function SUBR may be called with. | |
331 */ | |
332 (subr)) | |
333 { | |
334 CHECK_SUBR (subr); | |
335 return make_int (XSUBR (subr)->min_args); | |
336 } | |
337 | |
338 DEFUN ("subr-max-args", Fsubr_max_args, 1, 1, 0, /* | |
339 Return maximum number of args built-in function SUBR may be called with, | |
340 or nil if it takes an arbitrary number of arguments or is a special form. | |
341 */ | |
342 (subr)) | |
343 { | |
344 int nargs; | |
345 CHECK_SUBR (subr); | |
346 nargs = XSUBR (subr)->max_args; | |
347 if (nargs == MANY || nargs == UNEVALLED) | |
348 return Qnil; | |
349 else | |
350 return make_int (nargs); | |
351 } | |
352 | |
353 DEFUN ("subr-interactive", Fsubr_interactive, 1, 1, 0, /* | |
354 Return the interactive spec of the subr object, or nil. | |
355 If non-nil, the return value will be a list whose first element is | |
356 `interactive' and whose second element is the interactive spec. | |
357 */ | |
358 (subr)) | |
359 { | |
360 CONST char *prompt; | |
361 CHECK_SUBR (subr); | |
362 prompt = XSUBR (subr)->prompt; | |
363 return prompt ? list2 (Qinteractive, build_string (prompt)) : Qnil; | |
364 } | |
365 | |
366 | |
367 DEFUN ("characterp", Fcharacterp, 1, 1, 0, /* | |
368 Return t if OBJECT is a character. | |
369 Unlike in XEmacs v19 and FSF Emacs, a character is its own primitive type. | |
370 Any character can be converted into an equivalent integer using | |
371 `char-int'. To convert the other way, use `int-char'; however, | |
372 only some integers can be converted into characters. Such an integer | |
373 is called a `char-int'; see `char-int-p'. | |
374 | |
375 Some functions that work on integers (e.g. the comparison functions | |
376 <, <=, =, /=, etc. and the arithmetic functions +, -, *, etc.) | |
377 accept characters and implicitly convert them into integers. In | |
378 general, functions that work on characters also accept char-ints and | |
379 implicitly convert them into characters. WARNING: Neither of these | |
380 behaviors is very desirable, and they are maintained for backward | |
381 compatibility with old E-Lisp programs that confounded characters and | |
382 integers willy-nilly. These behaviors may change in the future; therefore, | |
383 do not rely on them. Instead, use the character-specific functions such | |
384 as `char='. | |
385 */ | |
386 (object)) | |
387 { | |
388 return CHARP (object) ? Qt : Qnil; | |
389 } | |
390 | |
391 DEFUN ("char-to-int", Fchar_to_int, 1, 1, 0, /* | |
392 Convert a character into an equivalent integer. | |
393 The resulting integer will always be non-negative. The integers in | |
394 the range 0 - 255 map to characters as follows: | |
395 | |
396 0 - 31 Control set 0 | |
397 32 - 127 ASCII | |
398 128 - 159 Control set 1 | |
399 160 - 255 Right half of ISO-8859-1 | |
400 | |
401 If support for Mule does not exist, these are the only valid character | |
402 values. When Mule support exists, the values assigned to other characters | |
403 may vary depending on the particular version of XEmacs, the order in which | |
404 character sets were loaded, etc., and you should not depend on them. | |
405 */ | |
406 (ch)) | |
407 { | |
408 CHECK_CHAR (ch); | |
409 return make_int (XCHAR (ch)); | |
410 } | |
411 | |
412 DEFUN ("int-to-char", Fint_to_char, 1, 1, 0, /* | |
413 Convert an integer into the equivalent character. | |
414 Not all integers correspond to valid characters; use `char-int-p' to | |
415 determine whether this is the case. If the integer cannot be converted, | |
416 nil is returned. | |
417 */ | |
418 (integer)) | |
419 { | |
420 CHECK_INT (integer); | |
421 if (CHAR_INTP (integer)) | |
422 return make_char (XINT (integer)); | |
423 else | |
424 return Qnil; | |
425 } | |
426 | |
427 DEFUN ("char-int-p", Fchar_int_p, 1, 1, 0, /* | |
428 Return t if OBJECT is an integer that can be converted into a character. | |
429 See `char-int'. | |
430 */ | |
431 (object)) | |
432 { | |
433 return CHAR_INTP (object) ? Qt : Qnil; | |
434 } | |
435 | |
436 DEFUN ("char-or-char-int-p", Fchar_or_char_int_p, 1, 1, 0, /* | |
437 Return t if OBJECT is a character or an integer that can be converted into one. | |
438 */ | |
439 (object)) | |
440 { | |
441 return CHAR_OR_CHAR_INTP (object) ? Qt : Qnil; | |
442 } | |
443 | |
444 DEFUN ("char-or-string-p", Fchar_or_string_p, 1, 1, 0, /* | |
445 Return t if OBJECT is a character (or a char-int) or a string. | |
446 It is semi-hateful that we allow a char-int here, as it goes against | |
447 the name of this function, but it makes the most sense considering the | |
448 other steps we take to maintain compatibility with the old character/integer | |
449 confoundedness in older versions of E-Lisp. | |
450 */ | |
451 (object)) | |
452 { | |
453 return CHAR_OR_CHAR_INTP (object) || STRINGP (object) ? Qt : Qnil; | |
454 } | |
455 | |
456 DEFUN ("integerp", Fintegerp, 1, 1, 0, /* | |
457 Return t if OBJECT is an integer. | |
458 */ | |
459 (object)) | |
460 { | |
461 return INTP (object) ? Qt : Qnil; | |
462 } | |
463 | |
464 DEFUN ("integer-or-marker-p", Finteger_or_marker_p, 1, 1, 0, /* | |
465 Return t if OBJECT is an integer or a marker (editor pointer). | |
466 */ | |
467 (object)) | |
468 { | |
469 return INTP (object) || MARKERP (object) ? Qt : Qnil; | |
470 } | |
471 | |
472 DEFUN ("integer-or-char-p", Finteger_or_char_p, 1, 1, 0, /* | |
473 Return t if OBJECT is an integer or a character. | |
474 */ | |
475 (object)) | |
476 { | |
477 return INTP (object) || CHARP (object) ? Qt : Qnil; | |
478 } | |
479 | |
480 DEFUN ("integer-char-or-marker-p", Finteger_char_or_marker_p, 1, 1, 0, /* | |
481 Return t if OBJECT is an integer, character or a marker (editor pointer). | |
482 */ | |
483 (object)) | |
484 { | |
485 return INTP (object) || CHARP (object) || MARKERP (object) ? Qt : Qnil; | |
486 } | |
487 | |
488 DEFUN ("natnump", Fnatnump, 1, 1, 0, /* | |
489 Return t if OBJECT is a nonnegative integer. | |
490 */ | |
491 (object)) | |
492 { | |
493 return NATNUMP (object) ? Qt : Qnil; | |
494 } | |
495 | |
496 DEFUN ("bitp", Fbitp, 1, 1, 0, /* | |
497 Return t if OBJECT is a bit (0 or 1). | |
498 */ | |
499 (object)) | |
500 { | |
501 return BITP (object) ? Qt : Qnil; | |
502 } | |
503 | |
504 DEFUN ("numberp", Fnumberp, 1, 1, 0, /* | |
505 Return t if OBJECT is a number (floating point or integer). | |
506 */ | |
507 (object)) | |
508 { | |
509 return INT_OR_FLOATP (object) ? Qt : Qnil; | |
510 } | |
511 | |
512 DEFUN ("number-or-marker-p", Fnumber_or_marker_p, 1, 1, 0, /* | |
513 Return t if OBJECT is a number or a marker. | |
514 */ | |
515 (object)) | |
516 { | |
517 return INT_OR_FLOATP (object) || MARKERP (object) ? Qt : Qnil; | |
518 } | |
519 | |
520 DEFUN ("number-char-or-marker-p", Fnumber_char_or_marker_p, 1, 1, 0, /* | |
521 Return t if OBJECT is a number, character or a marker. | |
522 */ | |
523 (object)) | |
524 { | |
525 return (INT_OR_FLOATP (object) || | |
526 CHARP (object) || | |
527 MARKERP (object)) | |
528 ? Qt : Qnil; | |
529 } | |
530 | |
531 #ifdef LISP_FLOAT_TYPE | |
532 DEFUN ("floatp", Ffloatp, 1, 1, 0, /* | |
533 Return t if OBJECT is a floating point number. | |
534 */ | |
535 (object)) | |
536 { | |
537 return FLOATP (object) ? Qt : Qnil; | |
538 } | |
539 #endif /* LISP_FLOAT_TYPE */ | |
540 | |
541 DEFUN ("type-of", Ftype_of, 1, 1, 0, /* | |
542 Return a symbol representing the type of OBJECT. | |
543 */ | |
544 (object)) | |
545 { | |
546 switch (XTYPE (object)) | |
547 { | |
548 case Lisp_Type_Record: | |
549 return intern (XRECORD_LHEADER_IMPLEMENTATION (object)->name); | |
550 | |
551 case Lisp_Type_Char: return Qcharacter; | |
552 | |
553 default: return Qinteger; | |
554 } | |
555 } | |
556 | |
557 | |
558 /* Extract and set components of lists */ | |
559 | |
560 DEFUN ("car", Fcar, 1, 1, 0, /* | |
561 Return the car of LIST. If arg is nil, return nil. | |
562 Error if arg is not nil and not a cons cell. See also `car-safe'. | |
563 */ | |
564 (list)) | |
565 { | |
566 while (1) | |
567 { | |
568 if (CONSP (list)) | |
569 return XCAR (list); | |
570 else if (NILP (list)) | |
571 return Qnil; | |
572 else | |
573 list = wrong_type_argument (Qlistp, list); | |
574 } | |
575 } | |
576 | |
577 DEFUN ("car-safe", Fcar_safe, 1, 1, 0, /* | |
578 Return the car of OBJECT if it is a cons cell, or else nil. | |
579 */ | |
580 (object)) | |
581 { | |
582 return CONSP (object) ? XCAR (object) : Qnil; | |
583 } | |
584 | |
585 DEFUN ("cdr", Fcdr, 1, 1, 0, /* | |
586 Return the cdr of LIST. If arg is nil, return nil. | |
587 Error if arg is not nil and not a cons cell. See also `cdr-safe'. | |
588 */ | |
589 (list)) | |
590 { | |
591 while (1) | |
592 { | |
593 if (CONSP (list)) | |
594 return XCDR (list); | |
595 else if (NILP (list)) | |
596 return Qnil; | |
597 else | |
598 list = wrong_type_argument (Qlistp, list); | |
599 } | |
600 } | |
601 | |
602 DEFUN ("cdr-safe", Fcdr_safe, 1, 1, 0, /* | |
603 Return the cdr of OBJECT if it is a cons cell, else nil. | |
604 */ | |
605 (object)) | |
606 { | |
607 return CONSP (object) ? XCDR (object) : Qnil; | |
608 } | |
609 | |
610 DEFUN ("setcar", Fsetcar, 2, 2, 0, /* | |
611 Set the car of CONSCELL to be NEWCAR. Return NEWCAR. | |
612 */ | |
613 (conscell, newcar)) | |
614 { | |
615 if (!CONSP (conscell)) | |
616 conscell = wrong_type_argument (Qconsp, conscell); | |
617 | |
618 XCAR (conscell) = newcar; | |
619 return newcar; | |
620 } | |
621 | |
622 DEFUN ("setcdr", Fsetcdr, 2, 2, 0, /* | |
623 Set the cdr of CONSCELL to be NEWCDR. Return NEWCDR. | |
624 */ | |
625 (conscell, newcdr)) | |
626 { | |
627 if (!CONSP (conscell)) | |
628 conscell = wrong_type_argument (Qconsp, conscell); | |
629 | |
630 XCDR (conscell) = newcdr; | |
631 return newcdr; | |
632 } | |
633 | |
634 /* Find the function at the end of a chain of symbol function indirections. | |
635 | |
636 If OBJECT is a symbol, find the end of its function chain and | |
637 return the value found there. If OBJECT is not a symbol, just | |
638 return it. If there is a cycle in the function chain, signal a | |
639 cyclic-function-indirection error. | |
640 | |
641 This is like Findirect_function, except that it doesn't signal an | |
642 error if the chain ends up unbound. */ | |
643 Lisp_Object | |
644 indirect_function (Lisp_Object object, int errorp) | |
645 { | |
646 #define FUNCTION_INDIRECTION_SUSPICION_LENGTH 16 | |
647 Lisp_Object tortoise, hare; | |
648 int count; | |
649 | |
650 for (hare = tortoise = object, count = 0; | |
651 SYMBOLP (hare); | |
652 hare = XSYMBOL (hare)->function, count++) | |
653 { | |
654 if (count < FUNCTION_INDIRECTION_SUSPICION_LENGTH) continue; | |
655 | |
656 if (count & 1) | |
657 tortoise = XSYMBOL (tortoise)->function; | |
658 if (EQ (hare, tortoise)) | |
659 return Fsignal (Qcyclic_function_indirection, list1 (object)); | |
660 } | |
661 | |
662 if (errorp && UNBOUNDP (hare)) | |
663 signal_void_function_error (object); | |
664 | |
665 return hare; | |
666 } | |
667 | |
668 DEFUN ("indirect-function", Findirect_function, 1, 1, 0, /* | |
669 Return the function at the end of OBJECT's function chain. | |
670 If OBJECT is a symbol, follow all function indirections and return | |
671 the final function binding. | |
672 If OBJECT is not a symbol, just return it. | |
673 Signal a void-function error if the final symbol is unbound. | |
674 Signal a cyclic-function-indirection error if there is a loop in the | |
675 function chain of symbols. | |
676 */ | |
677 (object)) | |
678 { | |
679 return indirect_function (object, 1); | |
680 } | |
681 | |
682 /* Extract and set vector and string elements */ | |
683 | |
684 DEFUN ("aref", Faref, 2, 2, 0, /* | |
685 Return the element of ARRAY at index INDEX. | |
686 ARRAY may be a vector, bit vector, or string. INDEX starts at 0. | |
687 */ | |
688 (array, index_)) | |
689 { | |
690 EMACS_INT idx; | |
691 | |
692 retry: | |
693 | |
694 if (INTP (index_)) idx = XINT (index_); | |
695 else if (CHARP (index_)) idx = XCHAR (index_); /* yuck! */ | |
696 else | |
697 { | |
698 index_ = wrong_type_argument (Qinteger_or_char_p, index_); | |
699 goto retry; | |
700 } | |
701 | |
702 if (idx < 0) goto range_error; | |
703 | |
704 if (VECTORP (array)) | |
705 { | |
706 if (idx >= XVECTOR_LENGTH (array)) goto range_error; | |
707 return XVECTOR_DATA (array)[idx]; | |
708 } | |
709 else if (BIT_VECTORP (array)) | |
710 { | |
711 if (idx >= bit_vector_length (XBIT_VECTOR (array))) goto range_error; | |
712 return make_int (bit_vector_bit (XBIT_VECTOR (array), idx)); | |
713 } | |
714 else if (STRINGP (array)) | |
715 { | |
716 if (idx >= XSTRING_CHAR_LENGTH (array)) goto range_error; | |
717 return make_char (string_char (XSTRING (array), idx)); | |
718 } | |
719 #ifdef LOSING_BYTECODE | |
720 else if (COMPILED_FUNCTIONP (array)) | |
721 { | |
722 /* Weird, gross compatibility kludge */ | |
723 return Felt (array, index_); | |
724 } | |
725 #endif | |
726 else | |
727 { | |
728 check_losing_bytecode ("aref", array); | |
729 array = wrong_type_argument (Qarrayp, array); | |
730 goto retry; | |
731 } | |
732 | |
733 range_error: | |
734 args_out_of_range (array, index_); | |
735 return Qnil; /* not reached */ | |
736 } | |
737 | |
738 DEFUN ("aset", Faset, 3, 3, 0, /* | |
739 Store into the element of ARRAY at index INDEX the value NEWVAL. | |
740 ARRAY may be a vector, bit vector, or string. INDEX starts at 0. | |
741 */ | |
742 (array, index_, newval)) | |
743 { | |
744 EMACS_INT idx; | |
745 | |
746 retry: | |
747 | |
748 if (INTP (index_)) idx = XINT (index_); | |
749 else if (CHARP (index_)) idx = XCHAR (index_); /* yuck! */ | |
750 else | |
751 { | |
752 index_ = wrong_type_argument (Qinteger_or_char_p, index_); | |
753 goto retry; | |
754 } | |
755 | |
756 if (idx < 0) goto range_error; | |
757 | |
758 if (VECTORP (array)) | |
759 { | |
760 if (idx >= XVECTOR_LENGTH (array)) goto range_error; | |
761 XVECTOR_DATA (array)[idx] = newval; | |
762 } | |
763 else if (BIT_VECTORP (array)) | |
764 { | |
765 if (idx >= bit_vector_length (XBIT_VECTOR (array))) goto range_error; | |
766 CHECK_BIT (newval); | |
767 set_bit_vector_bit (XBIT_VECTOR (array), idx, !ZEROP (newval)); | |
768 } | |
769 else if (STRINGP (array)) | |
770 { | |
771 CHECK_CHAR_COERCE_INT (newval); | |
772 if (idx >= XSTRING_CHAR_LENGTH (array)) goto range_error; | |
773 set_string_char (XSTRING (array), idx, XCHAR (newval)); | |
774 bump_string_modiff (array); | |
775 } | |
776 else | |
777 { | |
778 array = wrong_type_argument (Qarrayp, array); | |
779 goto retry; | |
780 } | |
781 | |
782 return newval; | |
783 | |
784 range_error: | |
785 args_out_of_range (array, index_); | |
786 return Qnil; /* not reached */ | |
787 } | |
788 | |
789 | |
790 /**********************************************************************/ | |
791 /* Arithmetic functions */ | |
792 /**********************************************************************/ | |
793 typedef struct | |
794 { | |
795 int int_p; | |
796 union | |
797 { | |
798 EMACS_INT ival; | |
799 double dval; | |
800 } c; | |
801 } int_or_double; | |
802 | |
803 static void | |
804 number_char_or_marker_to_int_or_double (Lisp_Object obj, int_or_double *p) | |
805 { | |
806 retry: | |
807 p->int_p = 1; | |
808 if (INTP (obj)) p->c.ival = XINT (obj); | |
809 else if (CHARP (obj)) p->c.ival = XCHAR (obj); | |
810 else if (MARKERP (obj)) p->c.ival = marker_position (obj); | |
811 #ifdef LISP_FLOAT_TYPE | |
812 else if (FLOATP (obj)) p->c.dval = XFLOAT_DATA (obj), p->int_p = 0; | |
813 #endif | |
814 else | |
815 { | |
816 obj = wrong_type_argument (Qnumber_char_or_marker_p, obj); | |
817 goto retry; | |
818 } | |
819 } | |
820 | |
821 static double | |
822 number_char_or_marker_to_double (Lisp_Object obj) | |
823 { | |
824 retry: | |
825 if (INTP (obj)) return (double) XINT (obj); | |
826 else if (CHARP (obj)) return (double) XCHAR (obj); | |
827 else if (MARKERP (obj)) return (double) marker_position (obj); | |
828 #ifdef LISP_FLOAT_TYPE | |
829 else if (FLOATP (obj)) return XFLOAT_DATA (obj); | |
830 #endif | |
831 else | |
832 { | |
833 obj = wrong_type_argument (Qnumber_char_or_marker_p, obj); | |
834 goto retry; | |
835 } | |
836 } | |
837 | |
838 static EMACS_INT | |
839 integer_char_or_marker_to_int (Lisp_Object obj) | |
840 { | |
841 retry: | |
842 if (INTP (obj)) return XINT (obj); | |
843 else if (CHARP (obj)) return XCHAR (obj); | |
844 else if (MARKERP (obj)) return marker_position (obj); | |
845 else | |
846 { | |
847 obj = wrong_type_argument (Qinteger_char_or_marker_p, obj); | |
848 goto retry; | |
849 } | |
850 } | |
851 | |
852 #define ARITHCOMPARE_MANY(op) \ | |
853 { \ | |
854 int_or_double iod1, iod2, *p = &iod1, *q = &iod2; \ | |
855 Lisp_Object *args_end = args + nargs; \ | |
856 \ | |
857 number_char_or_marker_to_int_or_double (*args++, p); \ | |
858 \ | |
859 while (args < args_end) \ | |
860 { \ | |
861 number_char_or_marker_to_int_or_double (*args++, q); \ | |
862 \ | |
863 if (!((p->int_p && q->int_p) ? \ | |
864 (p->c.ival op q->c.ival) : \ | |
865 ((p->int_p ? (double) p->c.ival : p->c.dval) op \ | |
866 (q->int_p ? (double) q->c.ival : q->c.dval)))) \ | |
867 return Qnil; \ | |
868 \ | |
869 { /* swap */ int_or_double *r = p; p = q; q = r; } \ | |
870 } \ | |
871 return Qt; \ | |
872 } | |
873 | |
874 DEFUN ("=", Feqlsign, 1, MANY, 0, /* | |
875 Return t if all the arguments are numerically equal. | |
876 The arguments may be numbers, characters or markers. | |
877 */ | |
878 (int nargs, Lisp_Object *args)) | |
879 { | |
880 ARITHCOMPARE_MANY (==) | |
881 } | |
882 | |
883 DEFUN ("<", Flss, 1, MANY, 0, /* | |
884 Return t if the sequence of arguments is monotonically increasing. | |
885 The arguments may be numbers, characters or markers. | |
886 */ | |
887 (int nargs, Lisp_Object *args)) | |
888 { | |
889 ARITHCOMPARE_MANY (<) | |
890 } | |
891 | |
892 DEFUN (">", Fgtr, 1, MANY, 0, /* | |
893 Return t if the sequence of arguments is monotonically decreasing. | |
894 The arguments may be numbers, characters or markers. | |
895 */ | |
896 (int nargs, Lisp_Object *args)) | |
897 { | |
898 ARITHCOMPARE_MANY (>) | |
899 } | |
900 | |
901 DEFUN ("<=", Fleq, 1, MANY, 0, /* | |
902 Return t if the sequence of arguments is monotonically nondecreasing. | |
903 The arguments may be numbers, characters or markers. | |
904 */ | |
905 (int nargs, Lisp_Object *args)) | |
906 { | |
907 ARITHCOMPARE_MANY (<=) | |
908 } | |
909 | |
910 DEFUN (">=", Fgeq, 1, MANY, 0, /* | |
911 Return t if the sequence of arguments is monotonically nonincreasing. | |
912 The arguments may be numbers, characters or markers. | |
913 */ | |
914 (int nargs, Lisp_Object *args)) | |
915 { | |
916 ARITHCOMPARE_MANY (>=) | |
917 } | |
918 | |
919 DEFUN ("/=", Fneq, 1, MANY, 0, /* | |
920 Return t if no two arguments are numerically equal. | |
921 The arguments may be numbers, characters or markers. | |
922 */ | |
923 (int nargs, Lisp_Object *args)) | |
924 { | |
925 Lisp_Object *args_end = args + nargs; | |
926 Lisp_Object *p, *q; | |
927 | |
928 /* Unlike all the other comparisons, this is an N*N algorithm. | |
929 We could use a hash table for nargs > 50 to make this linear. */ | |
930 for (p = args; p < args_end; p++) | |
931 { | |
932 int_or_double iod1, iod2; | |
933 number_char_or_marker_to_int_or_double (*p, &iod1); | |
934 | |
935 for (q = p + 1; q < args_end; q++) | |
936 { | |
937 number_char_or_marker_to_int_or_double (*q, &iod2); | |
938 | |
939 if (!((iod1.int_p && iod2.int_p) ? | |
940 (iod1.c.ival != iod2.c.ival) : | |
941 ((iod1.int_p ? (double) iod1.c.ival : iod1.c.dval) != | |
942 (iod2.int_p ? (double) iod2.c.ival : iod2.c.dval)))) | |
943 return Qnil; | |
944 } | |
945 } | |
946 return Qt; | |
947 } | |
948 | |
949 DEFUN ("zerop", Fzerop, 1, 1, 0, /* | |
950 Return t if NUMBER is zero. | |
951 */ | |
952 (number)) | |
953 { | |
954 retry: | |
955 if (INTP (number)) | |
956 return EQ (number, Qzero) ? Qt : Qnil; | |
957 #ifdef LISP_FLOAT_TYPE | |
958 else if (FLOATP (number)) | |
959 return XFLOAT_DATA (number) == 0.0 ? Qt : Qnil; | |
960 #endif /* LISP_FLOAT_TYPE */ | |
961 else | |
962 { | |
963 number = wrong_type_argument (Qnumberp, number); | |
964 goto retry; | |
965 } | |
966 } | |
967 | |
968 /* Convert between a 32-bit value and a cons of two 16-bit values. | |
969 This is used to pass 32-bit integers to and from the user. | |
970 Use time_to_lisp() and lisp_to_time() for time values. | |
971 | |
972 If you're thinking of using this to store a pointer into a Lisp Object | |
973 for internal purposes (such as when calling record_unwind_protect()), | |
974 try using make_opaque_ptr()/get_opaque_ptr() instead. */ | |
975 Lisp_Object | |
976 word_to_lisp (unsigned int item) | |
977 { | |
978 return Fcons (make_int (item >> 16), make_int (item & 0xffff)); | |
979 } | |
980 | |
981 unsigned int | |
982 lisp_to_word (Lisp_Object item) | |
983 { | |
984 if (INTP (item)) | |
985 return XINT (item); | |
986 else | |
987 { | |
988 Lisp_Object top = Fcar (item); | |
989 Lisp_Object bot = Fcdr (item); | |
990 CHECK_INT (top); | |
991 CHECK_INT (bot); | |
992 return (XINT (top) << 16) | (XINT (bot) & 0xffff); | |
993 } | |
994 } | |
995 | |
996 | |
997 DEFUN ("number-to-string", Fnumber_to_string, 1, 1, 0, /* | |
998 Convert NUM to a string by printing it in decimal. | |
999 Uses a minus sign if negative. | |
1000 NUM may be an integer or a floating point number. | |
1001 */ | |
1002 (num)) | |
1003 { | |
1004 char buffer[VALBITS]; | |
1005 | |
1006 CHECK_INT_OR_FLOAT (num); | |
1007 | |
1008 #ifdef LISP_FLOAT_TYPE | |
1009 if (FLOATP (num)) | |
1010 { | |
1011 char pigbuf[350]; /* see comments in float_to_string */ | |
1012 | |
1013 float_to_string (pigbuf, XFLOAT_DATA (num)); | |
1014 return build_string (pigbuf); | |
1015 } | |
1016 #endif /* LISP_FLOAT_TYPE */ | |
1017 | |
1018 long_to_string (buffer, XINT (num)); | |
1019 return build_string (buffer); | |
1020 } | |
1021 | |
1022 static int | |
1023 digit_to_number (int character, int base) | |
1024 { | |
1025 /* Assumes ASCII */ | |
1026 int digit = ((character >= '0' && character <= '9') ? character - '0' : | |
1027 (character >= 'a' && character <= 'z') ? character - 'a' + 10 : | |
1028 (character >= 'A' && character <= 'Z') ? character - 'A' + 10 : | |
1029 -1); | |
1030 | |
1031 return digit >= base ? -1 : digit; | |
1032 } | |
1033 | |
1034 DEFUN ("string-to-number", Fstring_to_number, 1, 2, 0, /* | |
1035 Convert STRING to a number by parsing it as a decimal number. | |
1036 This parses both integers and floating point numbers. | |
1037 It ignores leading spaces and tabs. | |
1038 | |
1039 If BASE, interpret STRING as a number in that base. If BASE isn't | |
1040 present, base 10 is used. BASE must be between 2 and 16 (inclusive). | |
1041 Floating point numbers always use base 10. | |
1042 */ | |
1043 (string, base)) | |
1044 { | |
1045 char *p; | |
1046 int b; | |
1047 | |
1048 CHECK_STRING (string); | |
1049 | |
1050 if (NILP (base)) | |
1051 b = 10; | |
1052 else | |
1053 { | |
1054 CHECK_INT (base); | |
1055 b = XINT (base); | |
1056 check_int_range (b, 2, 16); | |
1057 } | |
1058 | |
1059 p = (char *) XSTRING_DATA (string); | |
1060 | |
1061 /* Skip any whitespace at the front of the number. Some versions of | |
1062 atoi do this anyway, so we might as well make Emacs lisp consistent. */ | |
1063 while (*p == ' ' || *p == '\t') | |
1064 p++; | |
1065 | |
1066 #ifdef LISP_FLOAT_TYPE | |
1067 if (isfloat_string (p)) | |
1068 return make_float (atof (p)); | |
1069 #endif /* LISP_FLOAT_TYPE */ | |
1070 | |
1071 if (b == 10) | |
1072 { | |
1073 /* Use the system-provided functions for base 10. */ | |
1074 #if SIZEOF_EMACS_INT == SIZEOF_INT | |
1075 return make_int (atoi (p)); | |
1076 #elif SIZEOF_EMACS_INT == SIZEOF_LONG | |
1077 return make_int (atol (p)); | |
1078 #elif SIZEOF_EMACS_INT == SIZEOF_LONG_LONG | |
1079 return make_int (atoll (p)); | |
1080 #endif | |
1081 } | |
1082 else | |
1083 { | |
1084 int digit, negative = 1; | |
1085 EMACS_INT v = 0; | |
1086 | |
1087 if (*p == '-') | |
1088 { | |
1089 negative = -1; | |
1090 p++; | |
1091 } | |
1092 else if (*p == '+') | |
1093 p++; | |
1094 while (1) | |
1095 { | |
1096 digit = digit_to_number (*p++, b); | |
1097 if (digit < 0) | |
1098 break; | |
1099 v = v * b + digit; | |
1100 } | |
1101 return make_int (negative * v); | |
1102 } | |
1103 } | |
1104 | |
1105 | |
1106 DEFUN ("+", Fplus, 0, MANY, 0, /* | |
1107 Return sum of any number of arguments. | |
1108 The arguments should all be numbers, characters or markers. | |
1109 */ | |
1110 (int nargs, Lisp_Object *args)) | |
1111 { | |
1112 EMACS_INT iaccum = 0; | |
1113 Lisp_Object *args_end = args + nargs; | |
1114 | |
1115 while (args < args_end) | |
1116 { | |
1117 int_or_double iod; | |
1118 number_char_or_marker_to_int_or_double (*args++, &iod); | |
1119 if (iod.int_p) | |
1120 iaccum += iod.c.ival; | |
1121 else | |
1122 { | |
1123 double daccum = (double) iaccum + iod.c.dval; | |
1124 while (args < args_end) | |
1125 daccum += number_char_or_marker_to_double (*args++); | |
1126 return make_float (daccum); | |
1127 } | |
1128 } | |
1129 | |
1130 return make_int (iaccum); | |
1131 } | |
1132 | |
1133 DEFUN ("-", Fminus, 1, MANY, 0, /* | |
1134 Negate number or subtract numbers, characters or markers. | |
1135 With one arg, negates it. With more than one arg, | |
1136 subtracts all but the first from the first. | |
1137 */ | |
1138 (int nargs, Lisp_Object *args)) | |
1139 { | |
1140 EMACS_INT iaccum; | |
1141 double daccum; | |
1142 Lisp_Object *args_end = args + nargs; | |
1143 int_or_double iod; | |
1144 | |
1145 number_char_or_marker_to_int_or_double (*args++, &iod); | |
1146 if (iod.int_p) | |
1147 iaccum = nargs > 1 ? iod.c.ival : - iod.c.ival; | |
1148 else | |
1149 { | |
1150 daccum = nargs > 1 ? iod.c.dval : - iod.c.dval; | |
1151 goto do_float; | |
1152 } | |
1153 | |
1154 while (args < args_end) | |
1155 { | |
1156 number_char_or_marker_to_int_or_double (*args++, &iod); | |
1157 if (iod.int_p) | |
1158 iaccum -= iod.c.ival; | |
1159 else | |
1160 { | |
1161 daccum = (double) iaccum - iod.c.dval; | |
1162 goto do_float; | |
1163 } | |
1164 } | |
1165 | |
1166 return make_int (iaccum); | |
1167 | |
1168 do_float: | |
1169 for (; args < args_end; args++) | |
1170 daccum -= number_char_or_marker_to_double (*args); | |
1171 return make_float (daccum); | |
1172 } | |
1173 | |
1174 DEFUN ("*", Ftimes, 0, MANY, 0, /* | |
1175 Return product of any number of arguments. | |
1176 The arguments should all be numbers, characters or markers. | |
1177 */ | |
1178 (int nargs, Lisp_Object *args)) | |
1179 { | |
1180 EMACS_INT iaccum = 1; | |
1181 Lisp_Object *args_end = args + nargs; | |
1182 | |
1183 while (args < args_end) | |
1184 { | |
1185 int_or_double iod; | |
1186 number_char_or_marker_to_int_or_double (*args++, &iod); | |
1187 if (iod.int_p) | |
1188 iaccum *= iod.c.ival; | |
1189 else | |
1190 { | |
1191 double daccum = (double) iaccum * iod.c.dval; | |
1192 while (args < args_end) | |
1193 daccum *= number_char_or_marker_to_double (*args++); | |
1194 return make_float (daccum); | |
1195 } | |
1196 } | |
1197 | |
1198 return make_int (iaccum); | |
1199 } | |
1200 | |
1201 DEFUN ("/", Fquo, 1, MANY, 0, /* | |
1202 Return first argument divided by all the remaining arguments. | |
1203 The arguments must be numbers, characters or markers. | |
1204 With one argument, reciprocates the argument. | |
1205 */ | |
1206 (int nargs, Lisp_Object *args)) | |
1207 { | |
1208 EMACS_INT iaccum; | |
1209 double daccum; | |
1210 Lisp_Object *args_end = args + nargs; | |
1211 int_or_double iod; | |
1212 | |
1213 if (nargs == 1) | |
1214 iaccum = 1; | |
1215 else | |
1216 { | |
1217 number_char_or_marker_to_int_or_double (*args++, &iod); | |
1218 if (iod.int_p) | |
1219 iaccum = iod.c.ival; | |
1220 else | |
1221 { | |
1222 daccum = iod.c.dval; | |
1223 goto divide_floats; | |
1224 } | |
1225 } | |
1226 | |
1227 while (args < args_end) | |
1228 { | |
1229 number_char_or_marker_to_int_or_double (*args++, &iod); | |
1230 if (iod.int_p) | |
1231 { | |
1232 if (iod.c.ival == 0) goto divide_by_zero; | |
1233 iaccum /= iod.c.ival; | |
1234 } | |
1235 else | |
1236 { | |
1237 if (iod.c.dval == 0) goto divide_by_zero; | |
1238 daccum = (double) iaccum / iod.c.dval; | |
1239 goto divide_floats; | |
1240 } | |
1241 } | |
1242 | |
1243 return make_int (iaccum); | |
1244 | |
1245 divide_floats: | |
1246 for (; args < args_end; args++) | |
1247 { | |
1248 double dval = number_char_or_marker_to_double (*args); | |
1249 if (dval == 0) goto divide_by_zero; | |
1250 daccum /= dval; | |
1251 } | |
1252 return make_float (daccum); | |
1253 | |
1254 divide_by_zero: | |
1255 Fsignal (Qarith_error, Qnil); | |
1256 return Qnil; /* not reached */ | |
1257 } | |
1258 | |
1259 DEFUN ("max", Fmax, 1, MANY, 0, /* | |
1260 Return largest of all the arguments. | |
1261 All arguments must be numbers, characters or markers. | |
1262 The value is always a number; markers and characters are converted | |
1263 to numbers. | |
1264 */ | |
1265 (int nargs, Lisp_Object *args)) | |
1266 { | |
1267 EMACS_INT imax; | |
1268 double dmax; | |
1269 Lisp_Object *args_end = args + nargs; | |
1270 int_or_double iod; | |
1271 | |
1272 number_char_or_marker_to_int_or_double (*args++, &iod); | |
1273 if (iod.int_p) | |
1274 imax = iod.c.ival; | |
1275 else | |
1276 { | |
1277 dmax = iod.c.dval; | |
1278 goto max_floats; | |
1279 } | |
1280 | |
1281 while (args < args_end) | |
1282 { | |
1283 number_char_or_marker_to_int_or_double (*args++, &iod); | |
1284 if (iod.int_p) | |
1285 { | |
1286 if (imax < iod.c.ival) imax = iod.c.ival; | |
1287 } | |
1288 else | |
1289 { | |
1290 dmax = (double) imax; | |
1291 if (dmax < iod.c.dval) dmax = iod.c.dval; | |
1292 goto max_floats; | |
1293 } | |
1294 } | |
1295 | |
1296 return make_int (imax); | |
1297 | |
1298 max_floats: | |
1299 while (args < args_end) | |
1300 { | |
1301 double dval = number_char_or_marker_to_double (*args++); | |
1302 if (dmax < dval) dmax = dval; | |
1303 } | |
1304 return make_float (dmax); | |
1305 } | |
1306 | |
1307 DEFUN ("min", Fmin, 1, MANY, 0, /* | |
1308 Return smallest of all the arguments. | |
1309 All arguments must be numbers, characters or markers. | |
1310 The value is always a number; markers and characters are converted | |
1311 to numbers. | |
1312 */ | |
1313 (int nargs, Lisp_Object *args)) | |
1314 { | |
1315 EMACS_INT imin; | |
1316 double dmin; | |
1317 Lisp_Object *args_end = args + nargs; | |
1318 int_or_double iod; | |
1319 | |
1320 number_char_or_marker_to_int_or_double (*args++, &iod); | |
1321 if (iod.int_p) | |
1322 imin = iod.c.ival; | |
1323 else | |
1324 { | |
1325 dmin = iod.c.dval; | |
1326 goto min_floats; | |
1327 } | |
1328 | |
1329 while (args < args_end) | |
1330 { | |
1331 number_char_or_marker_to_int_or_double (*args++, &iod); | |
1332 if (iod.int_p) | |
1333 { | |
1334 if (imin > iod.c.ival) imin = iod.c.ival; | |
1335 } | |
1336 else | |
1337 { | |
1338 dmin = (double) imin; | |
1339 if (dmin > iod.c.dval) dmin = iod.c.dval; | |
1340 goto min_floats; | |
1341 } | |
1342 } | |
1343 | |
1344 return make_int (imin); | |
1345 | |
1346 min_floats: | |
1347 while (args < args_end) | |
1348 { | |
1349 double dval = number_char_or_marker_to_double (*args++); | |
1350 if (dmin > dval) dmin = dval; | |
1351 } | |
1352 return make_float (dmin); | |
1353 } | |
1354 | |
1355 DEFUN ("logand", Flogand, 0, MANY, 0, /* | |
1356 Return bitwise-and of all the arguments. | |
1357 Arguments may be integers, or markers or characters converted to integers. | |
1358 */ | |
1359 (int nargs, Lisp_Object *args)) | |
1360 { | |
1361 EMACS_INT bits = ~0; | |
1362 Lisp_Object *args_end = args + nargs; | |
1363 | |
1364 while (args < args_end) | |
1365 bits &= integer_char_or_marker_to_int (*args++); | |
1366 | |
1367 return make_int (bits); | |
1368 } | |
1369 | |
1370 DEFUN ("logior", Flogior, 0, MANY, 0, /* | |
1371 Return bitwise-or of all the arguments. | |
1372 Arguments may be integers, or markers or characters converted to integers. | |
1373 */ | |
1374 (int nargs, Lisp_Object *args)) | |
1375 { | |
1376 EMACS_INT bits = 0; | |
1377 Lisp_Object *args_end = args + nargs; | |
1378 | |
1379 while (args < args_end) | |
1380 bits |= integer_char_or_marker_to_int (*args++); | |
1381 | |
1382 return make_int (bits); | |
1383 } | |
1384 | |
1385 DEFUN ("logxor", Flogxor, 0, MANY, 0, /* | |
1386 Return bitwise-exclusive-or of all the arguments. | |
1387 Arguments may be integers, or markers or characters converted to integers. | |
1388 */ | |
1389 (int nargs, Lisp_Object *args)) | |
1390 { | |
1391 EMACS_INT bits = 0; | |
1392 Lisp_Object *args_end = args + nargs; | |
1393 | |
1394 while (args < args_end) | |
1395 bits ^= integer_char_or_marker_to_int (*args++); | |
1396 | |
1397 return make_int (bits); | |
1398 } | |
1399 | |
1400 DEFUN ("lognot", Flognot, 1, 1, 0, /* | |
1401 Return the bitwise complement of NUMBER. | |
1402 NUMBER may be an integer, marker or character converted to integer. | |
1403 */ | |
1404 (number)) | |
1405 { | |
1406 return make_int (~ integer_char_or_marker_to_int (number)); | |
1407 } | |
1408 | |
1409 DEFUN ("%", Frem, 2, 2, 0, /* | |
1410 Return remainder of first arg divided by second. | |
1411 Both must be integers, characters or markers. | |
1412 */ | |
1413 (num1, num2)) | |
1414 { | |
1415 EMACS_INT ival1 = integer_char_or_marker_to_int (num1); | |
1416 EMACS_INT ival2 = integer_char_or_marker_to_int (num2); | |
1417 | |
1418 if (ival2 == 0) | |
1419 Fsignal (Qarith_error, Qnil); | |
1420 | |
1421 return make_int (ival1 % ival2); | |
1422 } | |
1423 | |
1424 /* Note, ANSI *requires* the presence of the fmod() library routine. | |
1425 If your system doesn't have it, complain to your vendor, because | |
1426 that is a bug. */ | |
1427 | |
1428 #ifndef HAVE_FMOD | |
1429 double | |
1430 fmod (double f1, double f2) | |
1431 { | |
1432 if (f2 < 0.0) | |
1433 f2 = -f2; | |
1434 return f1 - f2 * floor (f1/f2); | |
1435 } | |
1436 #endif /* ! HAVE_FMOD */ | |
1437 | |
1438 | |
1439 DEFUN ("mod", Fmod, 2, 2, 0, /* | |
1440 Return X modulo Y. | |
1441 The result falls between zero (inclusive) and Y (exclusive). | |
1442 Both X and Y must be numbers, characters or markers. | |
1443 If either argument is a float, a float will be returned. | |
1444 */ | |
1445 (x, y)) | |
1446 { | |
1447 int_or_double iod1, iod2; | |
1448 number_char_or_marker_to_int_or_double (x, &iod1); | |
1449 number_char_or_marker_to_int_or_double (y, &iod2); | |
1450 | |
1451 #ifdef LISP_FLOAT_TYPE | |
1452 if (!iod1.int_p || !iod2.int_p) | |
1453 { | |
1454 double dval1 = iod1.int_p ? (double) iod1.c.ival : iod1.c.dval; | |
1455 double dval2 = iod2.int_p ? (double) iod2.c.ival : iod2.c.dval; | |
1456 if (dval2 == 0) goto divide_by_zero; | |
1457 dval1 = fmod (dval1, dval2); | |
1458 | |
1459 /* If the "remainder" comes out with the wrong sign, fix it. */ | |
1460 if (dval2 < 0 ? dval1 > 0 : dval1 < 0) | |
1461 dval1 += dval2; | |
1462 | |
1463 return make_float (dval1); | |
1464 } | |
1465 #endif /* LISP_FLOAT_TYPE */ | |
1466 { | |
1467 EMACS_INT ival; | |
1468 if (iod2.c.ival == 0) goto divide_by_zero; | |
1469 | |
1470 ival = iod1.c.ival % iod2.c.ival; | |
1471 | |
1472 /* If the "remainder" comes out with the wrong sign, fix it. */ | |
1473 if (iod2.c.ival < 0 ? ival > 0 : ival < 0) | |
1474 ival += iod2.c.ival; | |
1475 | |
1476 return make_int (ival); | |
1477 } | |
1478 | |
1479 divide_by_zero: | |
1480 Fsignal (Qarith_error, Qnil); | |
1481 return Qnil; /* not reached */ | |
1482 } | |
1483 | |
1484 DEFUN ("ash", Fash, 2, 2, 0, /* | |
1485 Return VALUE with its bits shifted left by COUNT. | |
1486 If COUNT is negative, shifting is actually to the right. | |
1487 In this case, the sign bit is duplicated. | |
1488 */ | |
1489 (value, count)) | |
1490 { | |
1491 CHECK_INT_COERCE_CHAR (value); | |
1492 CONCHECK_INT (count); | |
1493 | |
1494 return make_int (XINT (count) > 0 ? | |
1495 XINT (value) << XINT (count) : | |
1496 XINT (value) >> -XINT (count)); | |
1497 } | |
1498 | |
1499 DEFUN ("lsh", Flsh, 2, 2, 0, /* | |
1500 Return VALUE with its bits shifted left by COUNT. | |
1501 If COUNT is negative, shifting is actually to the right. | |
1502 In this case, zeros are shifted in on the left. | |
1503 */ | |
1504 (value, count)) | |
1505 { | |
1506 CHECK_INT_COERCE_CHAR (value); | |
1507 CONCHECK_INT (count); | |
1508 | |
1509 return make_int (XINT (count) > 0 ? | |
1510 XUINT (value) << XINT (count) : | |
1511 XUINT (value) >> -XINT (count)); | |
1512 } | |
1513 | |
1514 DEFUN ("1+", Fadd1, 1, 1, 0, /* | |
1515 Return NUMBER plus one. NUMBER may be a number, character or marker. | |
1516 Markers and characters are converted to integers. | |
1517 */ | |
1518 (number)) | |
1519 { | |
1520 retry: | |
1521 | |
1522 if (INTP (number)) return make_int (XINT (number) + 1); | |
1523 if (CHARP (number)) return make_int (XCHAR (number) + 1); | |
1524 if (MARKERP (number)) return make_int (marker_position (number) + 1); | |
1525 #ifdef LISP_FLOAT_TYPE | |
1526 if (FLOATP (number)) return make_float (XFLOAT_DATA (number) + 1.0); | |
1527 #endif /* LISP_FLOAT_TYPE */ | |
1528 | |
1529 number = wrong_type_argument (Qnumber_char_or_marker_p, number); | |
1530 goto retry; | |
1531 } | |
1532 | |
1533 DEFUN ("1-", Fsub1, 1, 1, 0, /* | |
1534 Return NUMBER minus one. NUMBER may be a number, character or marker. | |
1535 Markers and characters are converted to integers. | |
1536 */ | |
1537 (number)) | |
1538 { | |
1539 retry: | |
1540 | |
1541 if (INTP (number)) return make_int (XINT (number) - 1); | |
1542 if (CHARP (number)) return make_int (XCHAR (number) - 1); | |
1543 if (MARKERP (number)) return make_int (marker_position (number) - 1); | |
1544 #ifdef LISP_FLOAT_TYPE | |
1545 if (FLOATP (number)) return make_float (XFLOAT_DATA (number) - 1.0); | |
1546 #endif /* LISP_FLOAT_TYPE */ | |
1547 | |
1548 number = wrong_type_argument (Qnumber_char_or_marker_p, number); | |
1549 goto retry; | |
1550 } | |
1551 | |
1552 | |
1553 /************************************************************************/ | |
1554 /* weak lists */ | |
1555 /************************************************************************/ | |
1556 | |
1557 /* A weak list is like a normal list except that elements automatically | |
1558 disappear when no longer in use, i.e. when no longer GC-protected. | |
1559 The basic idea is that we don't mark the elements during GC, but | |
1560 wait for them to be marked elsewhere. If they're not marked, we | |
1561 remove them. This is analogous to weak hash tables; see the explanation | |
1562 there for more info. */ | |
1563 | |
1564 static Lisp_Object Vall_weak_lists; /* Gemarke es nicht!!! */ | |
1565 | |
1566 static Lisp_Object encode_weak_list_type (enum weak_list_type type); | |
1567 | |
1568 static Lisp_Object | |
1569 mark_weak_list (Lisp_Object obj) | |
1570 { | |
1571 return Qnil; /* nichts ist gemarkt */ | |
1572 } | |
1573 | |
1574 static void | |
1575 print_weak_list (Lisp_Object obj, Lisp_Object printcharfun, int escapeflag) | |
1576 { | |
1577 if (print_readably) | |
1578 error ("printing unreadable object #<weak-list>"); | |
1579 | |
1580 write_c_string ("#<weak-list ", printcharfun); | |
1581 print_internal (encode_weak_list_type (XWEAK_LIST (obj)->type), | |
1582 printcharfun, 0); | |
1583 write_c_string (" ", printcharfun); | |
1584 print_internal (XWEAK_LIST (obj)->list, printcharfun, escapeflag); | |
1585 write_c_string (">", printcharfun); | |
1586 } | |
1587 | |
1588 static int | |
1589 weak_list_equal (Lisp_Object obj1, Lisp_Object obj2, int depth) | |
1590 { | |
1591 struct weak_list *w1 = XWEAK_LIST (obj1); | |
1592 struct weak_list *w2 = XWEAK_LIST (obj2); | |
1593 | |
1594 return ((w1->type == w2->type) && | |
1595 internal_equal (w1->list, w2->list, depth + 1)); | |
1596 } | |
1597 | |
1598 static unsigned long | |
1599 weak_list_hash (Lisp_Object obj, int depth) | |
1600 { | |
1601 struct weak_list *w = XWEAK_LIST (obj); | |
1602 | |
1603 return HASH2 ((unsigned long) w->type, | |
1604 internal_hash (w->list, depth + 1)); | |
1605 } | |
1606 | |
1607 Lisp_Object | |
1608 make_weak_list (enum weak_list_type type) | |
1609 { | |
1610 Lisp_Object result; | |
1611 struct weak_list *wl = | |
1612 alloc_lcrecord_type (struct weak_list, &lrecord_weak_list); | |
1613 | |
1614 wl->list = Qnil; | |
1615 wl->type = type; | |
1616 XSETWEAK_LIST (result, wl); | |
1617 wl->next_weak = Vall_weak_lists; | |
1618 Vall_weak_lists = result; | |
1619 return result; | |
1620 } | |
1621 | |
1622 static const struct lrecord_description weak_list_description[] = { | |
1623 { XD_LISP_OBJECT, offsetof(struct weak_list, list), 1 }, | |
1624 { XD_LO_LINK, offsetof(struct weak_list, next_weak) }, | |
1625 { XD_END } | |
1626 }; | |
1627 | |
1628 DEFINE_LRECORD_IMPLEMENTATION ("weak-list", weak_list, | |
1629 mark_weak_list, print_weak_list, | |
1630 0, weak_list_equal, weak_list_hash, | |
1631 weak_list_description, | |
1632 struct weak_list); | |
1633 /* | |
1634 -- we do not mark the list elements (either the elements themselves | |
1635 or the cons cells that hold them) in the normal marking phase. | |
1636 -- at the end of marking, we go through all weak lists that are | |
1637 marked, and mark the cons cells that hold all marked | |
1638 objects, and possibly parts of the objects themselves. | |
1639 (See alloc.c, "after-mark".) | |
1640 -- after that, we prune away all the cons cells that are not marked. | |
1641 | |
1642 WARNING WARNING WARNING WARNING WARNING: | |
1643 | |
1644 The code in the following two functions is *unbelievably* tricky. | |
1645 Don't mess with it. You'll be sorry. | |
1646 | |
1647 Linked lists just majorly suck, d'ya know? | |
1648 */ | |
1649 | |
1650 int | |
1651 finish_marking_weak_lists (void) | |
1652 { | |
1653 Lisp_Object rest; | |
1654 int did_mark = 0; | |
1655 | |
1656 for (rest = Vall_weak_lists; | |
1657 !NILP (rest); | |
1658 rest = XWEAK_LIST (rest)->next_weak) | |
1659 { | |
1660 Lisp_Object rest2; | |
1661 enum weak_list_type type = XWEAK_LIST (rest)->type; | |
1662 | |
1663 if (! marked_p (rest)) | |
1664 /* The weak list is probably garbage. Ignore it. */ | |
1665 continue; | |
1666 | |
1667 for (rest2 = XWEAK_LIST (rest)->list; | |
1668 /* We need to be trickier since we're inside of GC; | |
1669 use CONSP instead of !NILP in case of user-visible | |
1670 imperfect lists */ | |
1671 CONSP (rest2); | |
1672 rest2 = XCDR (rest2)) | |
1673 { | |
1674 Lisp_Object elem; | |
1675 /* If the element is "marked" (meaning depends on the type | |
1676 of weak list), we need to mark the cons containing the | |
1677 element, and maybe the element itself (if only some part | |
1678 was already marked). */ | |
1679 int need_to_mark_cons = 0; | |
1680 int need_to_mark_elem = 0; | |
1681 | |
1682 /* If a cons is already marked, then its car is already marked | |
1683 (either because of an external pointer or because of | |
1684 a previous call to this function), and likewise for all | |
1685 the rest of the elements in the list, so we can stop now. */ | |
1686 if (marked_p (rest2)) | |
1687 break; | |
1688 | |
1689 elem = XCAR (rest2); | |
1690 | |
1691 switch (type) | |
1692 { | |
1693 case WEAK_LIST_SIMPLE: | |
1694 if (marked_p (elem)) | |
1695 need_to_mark_cons = 1; | |
1696 break; | |
1697 | |
1698 case WEAK_LIST_ASSOC: | |
1699 if (!CONSP (elem)) | |
1700 { | |
1701 /* just leave bogus elements there */ | |
1702 need_to_mark_cons = 1; | |
1703 need_to_mark_elem = 1; | |
1704 } | |
1705 else if (marked_p (XCAR (elem)) && | |
1706 marked_p (XCDR (elem))) | |
1707 { | |
1708 need_to_mark_cons = 1; | |
1709 /* We still need to mark elem, because it's | |
1710 probably not marked. */ | |
1711 need_to_mark_elem = 1; | |
1712 } | |
1713 break; | |
1714 | |
1715 case WEAK_LIST_KEY_ASSOC: | |
1716 if (!CONSP (elem)) | |
1717 { | |
1718 /* just leave bogus elements there */ | |
1719 need_to_mark_cons = 1; | |
1720 need_to_mark_elem = 1; | |
1721 } | |
1722 else if (marked_p (XCAR (elem))) | |
1723 { | |
1724 need_to_mark_cons = 1; | |
1725 /* We still need to mark elem and XCDR (elem); | |
1726 marking elem does both */ | |
1727 need_to_mark_elem = 1; | |
1728 } | |
1729 break; | |
1730 | |
1731 case WEAK_LIST_VALUE_ASSOC: | |
1732 if (!CONSP (elem)) | |
1733 { | |
1734 /* just leave bogus elements there */ | |
1735 need_to_mark_cons = 1; | |
1736 need_to_mark_elem = 1; | |
1737 } | |
1738 else if (marked_p (XCDR (elem))) | |
1739 { | |
1740 need_to_mark_cons = 1; | |
1741 /* We still need to mark elem and XCAR (elem); | |
1742 marking elem does both */ | |
1743 need_to_mark_elem = 1; | |
1744 } | |
1745 break; | |
1746 | |
1747 default: | |
1748 abort (); | |
1749 } | |
1750 | |
1751 if (need_to_mark_elem && ! marked_p (elem)) | |
1752 { | |
1753 mark_object (elem); | |
1754 did_mark = 1; | |
1755 } | |
1756 | |
1757 /* We also need to mark the cons that holds the elem or | |
1758 assoc-pair. We do *not* want to call (mark_object) here | |
1759 because that will mark the entire list; we just want to | |
1760 mark the cons itself. | |
1761 */ | |
1762 if (need_to_mark_cons) | |
1763 { | |
1764 Lisp_Cons *c = XCONS (rest2); | |
1765 if (!CONS_MARKED_P (c)) | |
1766 { | |
1767 MARK_CONS (c); | |
1768 did_mark = 1; | |
1769 } | |
1770 } | |
1771 } | |
1772 | |
1773 /* In case of imperfect list, need to mark the final cons | |
1774 because we're not removing it */ | |
1775 if (!NILP (rest2) && ! marked_p (rest2)) | |
1776 { | |
1777 mark_object (rest2); | |
1778 did_mark = 1; | |
1779 } | |
1780 } | |
1781 | |
1782 return did_mark; | |
1783 } | |
1784 | |
1785 void | |
1786 prune_weak_lists (void) | |
1787 { | |
1788 Lisp_Object rest, prev = Qnil; | |
1789 | |
1790 for (rest = Vall_weak_lists; | |
1791 !NILP (rest); | |
1792 rest = XWEAK_LIST (rest)->next_weak) | |
1793 { | |
1794 if (! (marked_p (rest))) | |
1795 { | |
1796 /* This weak list itself is garbage. Remove it from the list. */ | |
1797 if (NILP (prev)) | |
1798 Vall_weak_lists = XWEAK_LIST (rest)->next_weak; | |
1799 else | |
1800 XWEAK_LIST (prev)->next_weak = | |
1801 XWEAK_LIST (rest)->next_weak; | |
1802 } | |
1803 else | |
1804 { | |
1805 Lisp_Object rest2, prev2 = Qnil; | |
1806 Lisp_Object tortoise; | |
1807 int go_tortoise = 0; | |
1808 | |
1809 for (rest2 = XWEAK_LIST (rest)->list, tortoise = rest2; | |
1810 /* We need to be trickier since we're inside of GC; | |
1811 use CONSP instead of !NILP in case of user-visible | |
1812 imperfect lists */ | |
1813 CONSP (rest2);) | |
1814 { | |
1815 /* It suffices to check the cons for marking, | |
1816 regardless of the type of weak list: | |
1817 | |
1818 -- if the cons is pointed to somewhere else, | |
1819 then it should stay around and will be marked. | |
1820 -- otherwise, if it should stay around, it will | |
1821 have been marked in finish_marking_weak_lists(). | |
1822 -- otherwise, it's not marked and should disappear. | |
1823 */ | |
1824 if (! marked_p (rest2)) | |
1825 { | |
1826 /* bye bye :-( */ | |
1827 if (NILP (prev2)) | |
1828 XWEAK_LIST (rest)->list = XCDR (rest2); | |
1829 else | |
1830 XCDR (prev2) = XCDR (rest2); | |
1831 rest2 = XCDR (rest2); | |
1832 /* Ouch. Circularity checking is even trickier | |
1833 than I thought. When we cut out a link | |
1834 like this, we can't advance the turtle or | |
1835 it'll catch up to us. Imagine that we're | |
1836 standing on floor tiles and moving forward -- | |
1837 what we just did here is as if the floor | |
1838 tile under us just disappeared and all the | |
1839 ones ahead of us slid one tile towards us. | |
1840 In other words, we didn't move at all; | |
1841 if the tortoise was one step behind us | |
1842 previously, it still is, and therefore | |
1843 it must not move. */ | |
1844 } | |
1845 else | |
1846 { | |
1847 prev2 = rest2; | |
1848 | |
1849 /* Implementing circularity checking is trickier here | |
1850 than in other places because we have to guarantee | |
1851 that we've processed all elements before exiting | |
1852 due to a circularity. (In most places, an error | |
1853 is issued upon encountering a circularity, so it | |
1854 doesn't really matter if all elements are processed.) | |
1855 The idea is that we process along with the hare | |
1856 rather than the tortoise. If at any point in | |
1857 our forward process we encounter the tortoise, | |
1858 we must have already visited the spot, so we exit. | |
1859 (If we process with the tortoise, we can fail to | |
1860 process cases where a cons points to itself, or | |
1861 where cons A points to cons B, which points to | |
1862 cons A.) */ | |
1863 | |
1864 rest2 = XCDR (rest2); | |
1865 if (go_tortoise) | |
1866 tortoise = XCDR (tortoise); | |
1867 go_tortoise = !go_tortoise; | |
1868 if (EQ (rest2, tortoise)) | |
1869 break; | |
1870 } | |
1871 } | |
1872 | |
1873 prev = rest; | |
1874 } | |
1875 } | |
1876 } | |
1877 | |
1878 static enum weak_list_type | |
1879 decode_weak_list_type (Lisp_Object symbol) | |
1880 { | |
1881 CHECK_SYMBOL (symbol); | |
1882 if (EQ (symbol, Qsimple)) return WEAK_LIST_SIMPLE; | |
1883 if (EQ (symbol, Qassoc)) return WEAK_LIST_ASSOC; | |
1884 if (EQ (symbol, Qold_assoc)) return WEAK_LIST_ASSOC; /* EBOLA ALERT! */ | |
1885 if (EQ (symbol, Qkey_assoc)) return WEAK_LIST_KEY_ASSOC; | |
1886 if (EQ (symbol, Qvalue_assoc)) return WEAK_LIST_VALUE_ASSOC; | |
1887 | |
1888 signal_simple_error ("Invalid weak list type", symbol); | |
1889 return WEAK_LIST_SIMPLE; /* not reached */ | |
1890 } | |
1891 | |
1892 static Lisp_Object | |
1893 encode_weak_list_type (enum weak_list_type type) | |
1894 { | |
1895 switch (type) | |
1896 { | |
1897 case WEAK_LIST_SIMPLE: return Qsimple; | |
1898 case WEAK_LIST_ASSOC: return Qassoc; | |
1899 case WEAK_LIST_KEY_ASSOC: return Qkey_assoc; | |
1900 case WEAK_LIST_VALUE_ASSOC: return Qvalue_assoc; | |
1901 default: | |
1902 abort (); | |
1903 } | |
1904 | |
1905 return Qnil; /* not reached */ | |
1906 } | |
1907 | |
1908 DEFUN ("weak-list-p", Fweak_list_p, 1, 1, 0, /* | |
1909 Return non-nil if OBJECT is a weak list. | |
1910 */ | |
1911 (object)) | |
1912 { | |
1913 return WEAK_LISTP (object) ? Qt : Qnil; | |
1914 } | |
1915 | |
1916 DEFUN ("make-weak-list", Fmake_weak_list, 0, 1, 0, /* | |
1917 Return a new weak list object of type TYPE. | |
1918 A weak list object is an object that contains a list. This list behaves | |
1919 like any other list except that its elements do not count towards | |
1920 garbage collection -- if the only pointer to an object in inside a weak | |
1921 list (other than pointers in similar objects such as weak hash tables), | |
1922 the object is garbage collected and automatically removed from the list. | |
1923 This is used internally, for example, to manage the list holding the | |
1924 children of an extent -- an extent that is unused but has a parent will | |
1925 still be reclaimed, and will automatically be removed from its parent's | |
1926 list of children. | |
1927 | |
1928 Optional argument TYPE specifies the type of the weak list, and defaults | |
1929 to `simple'. Recognized types are | |
1930 | |
1931 `simple' Objects in the list disappear if not pointed to. | |
1932 `assoc' Objects in the list disappear if they are conses | |
1933 and either the car or the cdr of the cons is not | |
1934 pointed to. | |
1935 `key-assoc' Objects in the list disappear if they are conses | |
1936 and the car is not pointed to. | |
1937 `value-assoc' Objects in the list disappear if they are conses | |
1938 and the cdr is not pointed to. | |
1939 */ | |
1940 (type)) | |
1941 { | |
1942 if (NILP (type)) | |
1943 type = Qsimple; | |
1944 | |
1945 return make_weak_list (decode_weak_list_type (type)); | |
1946 } | |
1947 | |
1948 DEFUN ("weak-list-type", Fweak_list_type, 1, 1, 0, /* | |
1949 Return the type of the given weak-list object. | |
1950 */ | |
1951 (weak)) | |
1952 { | |
1953 CHECK_WEAK_LIST (weak); | |
1954 return encode_weak_list_type (XWEAK_LIST (weak)->type); | |
1955 } | |
1956 | |
1957 DEFUN ("weak-list-list", Fweak_list_list, 1, 1, 0, /* | |
1958 Return the list contained in a weak-list object. | |
1959 */ | |
1960 (weak)) | |
1961 { | |
1962 CHECK_WEAK_LIST (weak); | |
1963 return XWEAK_LIST_LIST (weak); | |
1964 } | |
1965 | |
1966 DEFUN ("set-weak-list-list", Fset_weak_list_list, 2, 2, 0, /* | |
1967 Change the list contained in a weak-list object. | |
1968 */ | |
1969 (weak, new_list)) | |
1970 { | |
1971 CHECK_WEAK_LIST (weak); | |
1972 XWEAK_LIST_LIST (weak) = new_list; | |
1973 return new_list; | |
1974 } | |
1975 | |
1976 | |
1977 /************************************************************************/ | |
1978 /* initialization */ | |
1979 /************************************************************************/ | |
1980 | |
1981 static SIGTYPE | |
1982 arith_error (int signo) | |
1983 { | |
1984 EMACS_REESTABLISH_SIGNAL (signo, arith_error); | |
1985 EMACS_UNBLOCK_SIGNAL (signo); | |
1986 signal_error (Qarith_error, Qnil); | |
1987 } | |
1988 | |
1989 void | |
1990 init_data_very_early (void) | |
1991 { | |
1992 /* Don't do this if just dumping out. | |
1993 We don't want to call `signal' in this case | |
1994 so that we don't have trouble with dumping | |
1995 signal-delivering routines in an inconsistent state. */ | |
1996 #ifndef CANNOT_DUMP | |
1997 if (!initialized) | |
1998 return; | |
1999 #endif /* CANNOT_DUMP */ | |
2000 signal (SIGFPE, arith_error); | |
2001 #ifdef uts | |
2002 signal (SIGEMT, arith_error); | |
2003 #endif /* uts */ | |
2004 } | |
2005 | |
2006 void | |
2007 init_errors_once_early (void) | |
2008 { | |
2009 defsymbol (&Qerror_conditions, "error-conditions"); | |
2010 defsymbol (&Qerror_message, "error-message"); | |
2011 | |
2012 /* We declare the errors here because some other deferrors depend | |
2013 on some of the errors below. */ | |
2014 | |
2015 /* ERROR is used as a signaler for random errors for which nothing | |
2016 else is right */ | |
2017 | |
2018 deferror (&Qerror, "error", "error", Qnil); | |
2019 deferror (&Qquit, "quit", "Quit", Qnil); | |
2020 | |
2021 deferror (&Qwrong_type_argument, "wrong-type-argument", | |
2022 "Wrong type argument", Qerror); | |
2023 deferror (&Qargs_out_of_range, "args-out-of-range", "Args out of range", | |
2024 Qerror); | |
2025 deferror (&Qvoid_function, "void-function", | |
2026 "Symbol's function definition is void", Qerror); | |
2027 deferror (&Qcyclic_function_indirection, "cyclic-function-indirection", | |
2028 "Symbol's chain of function indirections contains a loop", Qerror); | |
2029 deferror (&Qvoid_variable, "void-variable", | |
2030 "Symbol's value as variable is void", Qerror); | |
2031 deferror (&Qcyclic_variable_indirection, "cyclic-variable-indirection", | |
2032 "Symbol's chain of variable indirections contains a loop", Qerror); | |
2033 deferror (&Qsetting_constant, "setting-constant", | |
2034 "Attempt to set a constant symbol", Qerror); | |
2035 deferror (&Qinvalid_read_syntax, "invalid-read-syntax", | |
2036 "Invalid read syntax", Qerror); | |
2037 | |
2038 /* Generated by list traversal macros */ | |
2039 deferror (&Qmalformed_list, "malformed-list", | |
2040 "Malformed list", Qerror); | |
2041 deferror (&Qmalformed_property_list, "malformed-property-list", | |
2042 "Malformed property list", Qmalformed_list); | |
2043 deferror (&Qcircular_list, "circular-list", | |
2044 "Circular list", Qerror); | |
2045 deferror (&Qcircular_property_list, "circular-property-list", | |
2046 "Circular property list", Qcircular_list); | |
2047 | |
2048 deferror (&Qinvalid_function, "invalid-function", "Invalid function", | |
2049 Qerror); | |
2050 deferror (&Qwrong_number_of_arguments, "wrong-number-of-arguments", | |
2051 "Wrong number of arguments", Qerror); | |
2052 deferror (&Qno_catch, "no-catch", "No catch for tag", | |
2053 Qerror); | |
2054 deferror (&Qbeginning_of_buffer, "beginning-of-buffer", | |
2055 "Beginning of buffer", Qerror); | |
2056 deferror (&Qend_of_buffer, "end-of-buffer", "End of buffer", Qerror); | |
2057 deferror (&Qbuffer_read_only, "buffer-read-only", "Buffer is read-only", | |
2058 Qerror); | |
2059 | |
2060 deferror (&Qio_error, "io-error", "IO Error", Qerror); | |
2061 deferror (&Qend_of_file, "end-of-file", "End of stream", Qio_error); | |
2062 | |
2063 deferror (&Qarith_error, "arith-error", "Arithmetic error", Qerror); | |
2064 deferror (&Qrange_error, "range-error", "Arithmetic range error", | |
2065 Qarith_error); | |
2066 deferror (&Qdomain_error, "domain-error", "Arithmetic domain error", | |
2067 Qarith_error); | |
2068 deferror (&Qsingularity_error, "singularity-error", | |
2069 "Arithmetic singularity error", Qdomain_error); | |
2070 deferror (&Qoverflow_error, "overflow-error", | |
2071 "Arithmetic overflow error", Qdomain_error); | |
2072 deferror (&Qunderflow_error, "underflow-error", | |
2073 "Arithmetic underflow error", Qdomain_error); | |
2074 } | |
2075 | |
2076 void | |
2077 syms_of_data (void) | |
2078 { | |
2079 defsymbol (&Qquote, "quote"); | |
2080 defsymbol (&Qlambda, "lambda"); | |
2081 defsymbol (&Qlistp, "listp"); | |
2082 defsymbol (&Qtrue_list_p, "true-list-p"); | |
2083 defsymbol (&Qconsp, "consp"); | |
2084 defsymbol (&Qsubrp, "subrp"); | |
2085 defsymbol (&Qsymbolp, "symbolp"); | |
2086 defsymbol (&Qintegerp, "integerp"); | |
2087 defsymbol (&Qcharacterp, "characterp"); | |
2088 defsymbol (&Qnatnump, "natnump"); | |
2089 defsymbol (&Qstringp, "stringp"); | |
2090 defsymbol (&Qarrayp, "arrayp"); | |
2091 defsymbol (&Qsequencep, "sequencep"); | |
2092 defsymbol (&Qbufferp, "bufferp"); | |
2093 defsymbol (&Qbitp, "bitp"); | |
2094 defsymbol (&Qbit_vectorp, "bit-vector-p"); | |
2095 defsymbol (&Qvectorp, "vectorp"); | |
2096 defsymbol (&Qchar_or_string_p, "char-or-string-p"); | |
2097 defsymbol (&Qmarkerp, "markerp"); | |
2098 defsymbol (&Qinteger_or_marker_p, "integer-or-marker-p"); | |
2099 defsymbol (&Qinteger_or_char_p, "integer-or-char-p"); | |
2100 defsymbol (&Qinteger_char_or_marker_p, "integer-char-or-marker-p"); | |
2101 defsymbol (&Qnumberp, "numberp"); | |
2102 defsymbol (&Qnumber_char_or_marker_p, "number-char-or-marker-p"); | |
2103 defsymbol (&Qcdr, "cdr"); | |
2104 defsymbol (&Qweak_listp, "weak-list-p"); | |
2105 | |
2106 #ifdef LISP_FLOAT_TYPE | |
2107 defsymbol (&Qfloatp, "floatp"); | |
2108 #endif /* LISP_FLOAT_TYPE */ | |
2109 | |
2110 DEFSUBR (Fwrong_type_argument); | |
2111 | |
2112 DEFSUBR (Feq); | |
2113 DEFSUBR (Fold_eq); | |
2114 DEFSUBR (Fnull); | |
2115 Ffset (intern ("not"), intern ("null")); | |
2116 DEFSUBR (Flistp); | |
2117 DEFSUBR (Fnlistp); | |
2118 DEFSUBR (Ftrue_list_p); | |
2119 DEFSUBR (Fconsp); | |
2120 DEFSUBR (Fatom); | |
2121 DEFSUBR (Fchar_or_string_p); | |
2122 DEFSUBR (Fcharacterp); | |
2123 DEFSUBR (Fchar_int_p); | |
2124 DEFSUBR (Fchar_to_int); | |
2125 DEFSUBR (Fint_to_char); | |
2126 DEFSUBR (Fchar_or_char_int_p); | |
2127 DEFSUBR (Fintegerp); | |
2128 DEFSUBR (Finteger_or_marker_p); | |
2129 DEFSUBR (Finteger_or_char_p); | |
2130 DEFSUBR (Finteger_char_or_marker_p); | |
2131 DEFSUBR (Fnumberp); | |
2132 DEFSUBR (Fnumber_or_marker_p); | |
2133 DEFSUBR (Fnumber_char_or_marker_p); | |
2134 #ifdef LISP_FLOAT_TYPE | |
2135 DEFSUBR (Ffloatp); | |
2136 #endif /* LISP_FLOAT_TYPE */ | |
2137 DEFSUBR (Fnatnump); | |
2138 DEFSUBR (Fsymbolp); | |
2139 DEFSUBR (Fkeywordp); | |
2140 DEFSUBR (Fstringp); | |
2141 DEFSUBR (Fvectorp); | |
2142 DEFSUBR (Fbitp); | |
2143 DEFSUBR (Fbit_vector_p); | |
2144 DEFSUBR (Farrayp); | |
2145 DEFSUBR (Fsequencep); | |
2146 DEFSUBR (Fmarkerp); | |
2147 DEFSUBR (Fsubrp); | |
2148 DEFSUBR (Fsubr_min_args); | |
2149 DEFSUBR (Fsubr_max_args); | |
2150 DEFSUBR (Fsubr_interactive); | |
2151 DEFSUBR (Ftype_of); | |
2152 DEFSUBR (Fcar); | |
2153 DEFSUBR (Fcdr); | |
2154 DEFSUBR (Fcar_safe); | |
2155 DEFSUBR (Fcdr_safe); | |
2156 DEFSUBR (Fsetcar); | |
2157 DEFSUBR (Fsetcdr); | |
2158 DEFSUBR (Findirect_function); | |
2159 DEFSUBR (Faref); | |
2160 DEFSUBR (Faset); | |
2161 | |
2162 DEFSUBR (Fnumber_to_string); | |
2163 DEFSUBR (Fstring_to_number); | |
2164 DEFSUBR (Feqlsign); | |
2165 DEFSUBR (Flss); | |
2166 DEFSUBR (Fgtr); | |
2167 DEFSUBR (Fleq); | |
2168 DEFSUBR (Fgeq); | |
2169 DEFSUBR (Fneq); | |
2170 DEFSUBR (Fzerop); | |
2171 DEFSUBR (Fplus); | |
2172 DEFSUBR (Fminus); | |
2173 DEFSUBR (Ftimes); | |
2174 DEFSUBR (Fquo); | |
2175 DEFSUBR (Frem); | |
2176 DEFSUBR (Fmod); | |
2177 DEFSUBR (Fmax); | |
2178 DEFSUBR (Fmin); | |
2179 DEFSUBR (Flogand); | |
2180 DEFSUBR (Flogior); | |
2181 DEFSUBR (Flogxor); | |
2182 DEFSUBR (Flsh); | |
2183 DEFSUBR (Fash); | |
2184 DEFSUBR (Fadd1); | |
2185 DEFSUBR (Fsub1); | |
2186 DEFSUBR (Flognot); | |
2187 | |
2188 DEFSUBR (Fweak_list_p); | |
2189 DEFSUBR (Fmake_weak_list); | |
2190 DEFSUBR (Fweak_list_type); | |
2191 DEFSUBR (Fweak_list_list); | |
2192 DEFSUBR (Fset_weak_list_list); | |
2193 } | |
2194 | |
2195 void | |
2196 vars_of_data (void) | |
2197 { | |
2198 /* This must not be staticpro'd */ | |
2199 Vall_weak_lists = Qnil; | |
2200 pdump_wire_list (&Vall_weak_lists); | |
2201 | |
2202 #ifdef DEBUG_XEMACS | |
2203 DEFVAR_BOOL ("debug-issue-ebola-notices", &debug_issue_ebola_notices /* | |
2204 If non-zero, note when your code may be suffering from char-int confoundance. | |
2205 That is to say, if XEmacs encounters a usage of `eq', `memq', `equal', | |
2206 etc. where an int and a char with the same value are being compared, | |
2207 it will issue a notice on stderr to this effect, along with a backtrace. | |
2208 In such situations, the result would be different in XEmacs 19 versus | |
2209 XEmacs 20, and you probably don't want this. | |
2210 | |
2211 Note that in order to see these notices, you have to byte compile your | |
2212 code under XEmacs 20 -- any code byte-compiled under XEmacs 19 will | |
2213 have its chars and ints all confounded in the byte code, making it | |
2214 impossible to accurately determine Ebola infection. | |
2215 */ ); | |
2216 | |
2217 debug_issue_ebola_notices = 0; | |
2218 | |
2219 DEFVAR_INT ("debug-ebola-backtrace-length", | |
2220 &debug_ebola_backtrace_length /* | |
2221 Length (in stack frames) of short backtrace printed out in Ebola notices. | |
2222 See `debug-issue-ebola-notices'. | |
2223 */ ); | |
2224 debug_ebola_backtrace_length = 32; | |
2225 | |
2226 #endif /* DEBUG_XEMACS */ | |
2227 } |