Cleanup xmalloc.
[bpt/emacs.git] / src / fns.c
1 /* Random utility Lisp functions.
2 Copyright (C) 1985-1987, 1993-1995, 1997-2012
3 Free Software Foundation, Inc.
4
5 This file is part of GNU Emacs.
6
7 GNU Emacs is free software: you can redistribute it and/or modify
8 it under the terms of the GNU General Public License as published by
9 the Free Software Foundation, either version 3 of the License, or
10 (at your option) any later version.
11
12 GNU Emacs is distributed in the hope that it will be useful,
13 but WITHOUT ANY WARRANTY; without even the implied warranty of
14 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
15 GNU General Public License for more details.
16
17 You should have received a copy of the GNU General Public License
18 along with GNU Emacs. If not, see <http://www.gnu.org/licenses/>. */
19
20 #include <config.h>
21
22 #include <unistd.h>
23 #include <time.h>
24 #include <setjmp.h>
25
26 #include <intprops.h>
27
28 #include "lisp.h"
29 #include "commands.h"
30 #include "character.h"
31 #include "coding.h"
32 #include "buffer.h"
33 #include "keyboard.h"
34 #include "keymap.h"
35 #include "intervals.h"
36 #include "frame.h"
37 #include "window.h"
38 #include "blockinput.h"
39 #ifdef HAVE_MENUS
40 #if defined (HAVE_X_WINDOWS)
41 #include "xterm.h"
42 #endif
43 #endif /* HAVE_MENUS */
44
45 Lisp_Object Qstring_lessp;
46 static Lisp_Object Qprovide, Qrequire;
47 static Lisp_Object Qyes_or_no_p_history;
48 Lisp_Object Qcursor_in_echo_area;
49 static Lisp_Object Qwidget_type;
50 static Lisp_Object Qcodeset, Qdays, Qmonths, Qpaper;
51
52 static Lisp_Object Qmd5, Qsha1, Qsha224, Qsha256, Qsha384, Qsha512;
53
54 static int internal_equal (Lisp_Object , Lisp_Object, int, int);
55
56 #ifndef HAVE_UNISTD_H
57 extern long time ();
58 #endif
59 \f
60 DEFUN ("identity", Fidentity, Sidentity, 1, 1, 0,
61 doc: /* Return the argument unchanged. */)
62 (Lisp_Object arg)
63 {
64 return arg;
65 }
66
67 DEFUN ("random", Frandom, Srandom, 0, 1, 0,
68 doc: /* Return a pseudo-random number.
69 All integers representable in Lisp are equally likely.
70 On most systems, this is 29 bits' worth.
71 With positive integer LIMIT, return random number in interval [0,LIMIT).
72 With argument t, set the random number seed from the current time and pid.
73 Other values of LIMIT are ignored. */)
74 (Lisp_Object limit)
75 {
76 EMACS_INT val;
77 Lisp_Object lispy_val;
78
79 if (EQ (limit, Qt))
80 {
81 EMACS_TIME t;
82 EMACS_GET_TIME (t);
83 seed_random (getpid () ^ EMACS_SECS (t) ^ EMACS_NSECS (t));
84 }
85
86 if (NATNUMP (limit) && XFASTINT (limit) != 0)
87 {
88 /* Try to take our random number from the higher bits of VAL,
89 not the lower, since (says Gentzel) the low bits of `random'
90 are less random than the higher ones. We do this by using the
91 quotient rather than the remainder. At the high end of the RNG
92 it's possible to get a quotient larger than n; discarding
93 these values eliminates the bias that would otherwise appear
94 when using a large n. */
95 EMACS_INT denominator = (INTMASK + 1) / XFASTINT (limit);
96 do
97 val = get_random () / denominator;
98 while (val >= XFASTINT (limit));
99 }
100 else
101 val = get_random ();
102 XSETINT (lispy_val, val);
103 return lispy_val;
104 }
105 \f
106 /* Heuristic on how many iterations of a tight loop can be safely done
107 before it's time to do a QUIT. This must be a power of 2. */
108 enum { QUIT_COUNT_HEURISTIC = 1 << 16 };
109
110 /* Random data-structure functions */
111
112 DEFUN ("length", Flength, Slength, 1, 1, 0,
113 doc: /* Return the length of vector, list or string SEQUENCE.
114 A byte-code function object is also allowed.
115 If the string contains multibyte characters, this is not necessarily
116 the number of bytes in the string; it is the number of characters.
117 To get the number of bytes, use `string-bytes'. */)
118 (register Lisp_Object sequence)
119 {
120 register Lisp_Object val;
121
122 if (STRINGP (sequence))
123 XSETFASTINT (val, SCHARS (sequence));
124 else if (VECTORP (sequence))
125 XSETFASTINT (val, ASIZE (sequence));
126 else if (CHAR_TABLE_P (sequence))
127 XSETFASTINT (val, MAX_CHAR);
128 else if (BOOL_VECTOR_P (sequence))
129 XSETFASTINT (val, XBOOL_VECTOR (sequence)->size);
130 else if (COMPILEDP (sequence))
131 XSETFASTINT (val, ASIZE (sequence) & PSEUDOVECTOR_SIZE_MASK);
132 else if (CONSP (sequence))
133 {
134 EMACS_INT i = 0;
135
136 do
137 {
138 ++i;
139 if ((i & (QUIT_COUNT_HEURISTIC - 1)) == 0)
140 {
141 if (MOST_POSITIVE_FIXNUM < i)
142 error ("List too long");
143 QUIT;
144 }
145 sequence = XCDR (sequence);
146 }
147 while (CONSP (sequence));
148
149 CHECK_LIST_END (sequence, sequence);
150
151 val = make_number (i);
152 }
153 else if (NILP (sequence))
154 XSETFASTINT (val, 0);
155 else
156 wrong_type_argument (Qsequencep, sequence);
157
158 return val;
159 }
160
161 /* This does not check for quits. That is safe since it must terminate. */
162
163 DEFUN ("safe-length", Fsafe_length, Ssafe_length, 1, 1, 0,
164 doc: /* Return the length of a list, but avoid error or infinite loop.
165 This function never gets an error. If LIST is not really a list,
166 it returns 0. If LIST is circular, it returns a finite value
167 which is at least the number of distinct elements. */)
168 (Lisp_Object list)
169 {
170 Lisp_Object tail, halftail;
171 double hilen = 0;
172 uintmax_t lolen = 1;
173
174 if (! CONSP (list))
175 return make_number (0);
176
177 /* halftail is used to detect circular lists. */
178 for (tail = halftail = list; ; )
179 {
180 tail = XCDR (tail);
181 if (! CONSP (tail))
182 break;
183 if (EQ (tail, halftail))
184 break;
185 lolen++;
186 if ((lolen & 1) == 0)
187 {
188 halftail = XCDR (halftail);
189 if ((lolen & (QUIT_COUNT_HEURISTIC - 1)) == 0)
190 {
191 QUIT;
192 if (lolen == 0)
193 hilen += UINTMAX_MAX + 1.0;
194 }
195 }
196 }
197
198 /* If the length does not fit into a fixnum, return a float.
199 On all known practical machines this returns an upper bound on
200 the true length. */
201 return hilen ? make_float (hilen + lolen) : make_fixnum_or_float (lolen);
202 }
203
204 DEFUN ("string-bytes", Fstring_bytes, Sstring_bytes, 1, 1, 0,
205 doc: /* Return the number of bytes in STRING.
206 If STRING is multibyte, this may be greater than the length of STRING. */)
207 (Lisp_Object string)
208 {
209 CHECK_STRING (string);
210 return make_number (SBYTES (string));
211 }
212
213 DEFUN ("string-equal", Fstring_equal, Sstring_equal, 2, 2, 0,
214 doc: /* Return t if two strings have identical contents.
215 Case is significant, but text properties are ignored.
216 Symbols are also allowed; their print names are used instead. */)
217 (register Lisp_Object s1, Lisp_Object s2)
218 {
219 if (SYMBOLP (s1))
220 s1 = SYMBOL_NAME (s1);
221 if (SYMBOLP (s2))
222 s2 = SYMBOL_NAME (s2);
223 CHECK_STRING (s1);
224 CHECK_STRING (s2);
225
226 if (SCHARS (s1) != SCHARS (s2)
227 || SBYTES (s1) != SBYTES (s2)
228 || memcmp (SDATA (s1), SDATA (s2), SBYTES (s1)))
229 return Qnil;
230 return Qt;
231 }
232
233 DEFUN ("compare-strings", Fcompare_strings, Scompare_strings, 6, 7, 0,
234 doc: /* Compare the contents of two strings, converting to multibyte if needed.
235 In string STR1, skip the first START1 characters and stop at END1.
236 In string STR2, skip the first START2 characters and stop at END2.
237 END1 and END2 default to the full lengths of the respective strings.
238
239 Case is significant in this comparison if IGNORE-CASE is nil.
240 Unibyte strings are converted to multibyte for comparison.
241
242 The value is t if the strings (or specified portions) match.
243 If string STR1 is less, the value is a negative number N;
244 - 1 - N is the number of characters that match at the beginning.
245 If string STR1 is greater, the value is a positive number N;
246 N - 1 is the number of characters that match at the beginning. */)
247 (Lisp_Object str1, Lisp_Object start1, Lisp_Object end1, Lisp_Object str2, Lisp_Object start2, Lisp_Object end2, Lisp_Object ignore_case)
248 {
249 register ptrdiff_t end1_char, end2_char;
250 register ptrdiff_t i1, i1_byte, i2, i2_byte;
251
252 CHECK_STRING (str1);
253 CHECK_STRING (str2);
254 if (NILP (start1))
255 start1 = make_number (0);
256 if (NILP (start2))
257 start2 = make_number (0);
258 CHECK_NATNUM (start1);
259 CHECK_NATNUM (start2);
260 if (! NILP (end1))
261 CHECK_NATNUM (end1);
262 if (! NILP (end2))
263 CHECK_NATNUM (end2);
264
265 end1_char = SCHARS (str1);
266 if (! NILP (end1) && end1_char > XINT (end1))
267 end1_char = XINT (end1);
268 if (end1_char < XINT (start1))
269 args_out_of_range (str1, start1);
270
271 end2_char = SCHARS (str2);
272 if (! NILP (end2) && end2_char > XINT (end2))
273 end2_char = XINT (end2);
274 if (end2_char < XINT (start2))
275 args_out_of_range (str2, start2);
276
277 i1 = XINT (start1);
278 i2 = XINT (start2);
279
280 i1_byte = string_char_to_byte (str1, i1);
281 i2_byte = string_char_to_byte (str2, i2);
282
283 while (i1 < end1_char && i2 < end2_char)
284 {
285 /* When we find a mismatch, we must compare the
286 characters, not just the bytes. */
287 int c1, c2;
288
289 if (STRING_MULTIBYTE (str1))
290 FETCH_STRING_CHAR_ADVANCE_NO_CHECK (c1, str1, i1, i1_byte);
291 else
292 {
293 c1 = SREF (str1, i1++);
294 MAKE_CHAR_MULTIBYTE (c1);
295 }
296
297 if (STRING_MULTIBYTE (str2))
298 FETCH_STRING_CHAR_ADVANCE_NO_CHECK (c2, str2, i2, i2_byte);
299 else
300 {
301 c2 = SREF (str2, i2++);
302 MAKE_CHAR_MULTIBYTE (c2);
303 }
304
305 if (c1 == c2)
306 continue;
307
308 if (! NILP (ignore_case))
309 {
310 Lisp_Object tem;
311
312 tem = Fupcase (make_number (c1));
313 c1 = XINT (tem);
314 tem = Fupcase (make_number (c2));
315 c2 = XINT (tem);
316 }
317
318 if (c1 == c2)
319 continue;
320
321 /* Note that I1 has already been incremented
322 past the character that we are comparing;
323 hence we don't add or subtract 1 here. */
324 if (c1 < c2)
325 return make_number (- i1 + XINT (start1));
326 else
327 return make_number (i1 - XINT (start1));
328 }
329
330 if (i1 < end1_char)
331 return make_number (i1 - XINT (start1) + 1);
332 if (i2 < end2_char)
333 return make_number (- i1 + XINT (start1) - 1);
334
335 return Qt;
336 }
337
338 DEFUN ("string-lessp", Fstring_lessp, Sstring_lessp, 2, 2, 0,
339 doc: /* Return t if first arg string is less than second in lexicographic order.
340 Case is significant.
341 Symbols are also allowed; their print names are used instead. */)
342 (register Lisp_Object s1, Lisp_Object s2)
343 {
344 register ptrdiff_t end;
345 register ptrdiff_t i1, i1_byte, i2, i2_byte;
346
347 if (SYMBOLP (s1))
348 s1 = SYMBOL_NAME (s1);
349 if (SYMBOLP (s2))
350 s2 = SYMBOL_NAME (s2);
351 CHECK_STRING (s1);
352 CHECK_STRING (s2);
353
354 i1 = i1_byte = i2 = i2_byte = 0;
355
356 end = SCHARS (s1);
357 if (end > SCHARS (s2))
358 end = SCHARS (s2);
359
360 while (i1 < end)
361 {
362 /* When we find a mismatch, we must compare the
363 characters, not just the bytes. */
364 int c1, c2;
365
366 FETCH_STRING_CHAR_ADVANCE (c1, s1, i1, i1_byte);
367 FETCH_STRING_CHAR_ADVANCE (c2, s2, i2, i2_byte);
368
369 if (c1 != c2)
370 return c1 < c2 ? Qt : Qnil;
371 }
372 return i1 < SCHARS (s2) ? Qt : Qnil;
373 }
374 \f
375 static Lisp_Object concat (ptrdiff_t nargs, Lisp_Object *args,
376 enum Lisp_Type target_type, int last_special);
377
378 /* ARGSUSED */
379 Lisp_Object
380 concat2 (Lisp_Object s1, Lisp_Object s2)
381 {
382 Lisp_Object args[2];
383 args[0] = s1;
384 args[1] = s2;
385 return concat (2, args, Lisp_String, 0);
386 }
387
388 /* ARGSUSED */
389 Lisp_Object
390 concat3 (Lisp_Object s1, Lisp_Object s2, Lisp_Object s3)
391 {
392 Lisp_Object args[3];
393 args[0] = s1;
394 args[1] = s2;
395 args[2] = s3;
396 return concat (3, args, Lisp_String, 0);
397 }
398
399 DEFUN ("append", Fappend, Sappend, 0, MANY, 0,
400 doc: /* Concatenate all the arguments and make the result a list.
401 The result is a list whose elements are the elements of all the arguments.
402 Each argument may be a list, vector or string.
403 The last argument is not copied, just used as the tail of the new list.
404 usage: (append &rest SEQUENCES) */)
405 (ptrdiff_t nargs, Lisp_Object *args)
406 {
407 return concat (nargs, args, Lisp_Cons, 1);
408 }
409
410 DEFUN ("concat", Fconcat, Sconcat, 0, MANY, 0,
411 doc: /* Concatenate all the arguments and make the result a string.
412 The result is a string whose elements are the elements of all the arguments.
413 Each argument may be a string or a list or vector of characters (integers).
414 usage: (concat &rest SEQUENCES) */)
415 (ptrdiff_t nargs, Lisp_Object *args)
416 {
417 return concat (nargs, args, Lisp_String, 0);
418 }
419
420 DEFUN ("vconcat", Fvconcat, Svconcat, 0, MANY, 0,
421 doc: /* Concatenate all the arguments and make the result a vector.
422 The result is a vector whose elements are the elements of all the arguments.
423 Each argument may be a list, vector or string.
424 usage: (vconcat &rest SEQUENCES) */)
425 (ptrdiff_t nargs, Lisp_Object *args)
426 {
427 return concat (nargs, args, Lisp_Vectorlike, 0);
428 }
429
430
431 DEFUN ("copy-sequence", Fcopy_sequence, Scopy_sequence, 1, 1, 0,
432 doc: /* Return a copy of a list, vector, string or char-table.
433 The elements of a list or vector are not copied; they are shared
434 with the original. */)
435 (Lisp_Object arg)
436 {
437 if (NILP (arg)) return arg;
438
439 if (CHAR_TABLE_P (arg))
440 {
441 return copy_char_table (arg);
442 }
443
444 if (BOOL_VECTOR_P (arg))
445 {
446 Lisp_Object val;
447 ptrdiff_t size_in_chars
448 = ((XBOOL_VECTOR (arg)->size + BOOL_VECTOR_BITS_PER_CHAR - 1)
449 / BOOL_VECTOR_BITS_PER_CHAR);
450
451 val = Fmake_bool_vector (Flength (arg), Qnil);
452 memcpy (XBOOL_VECTOR (val)->data, XBOOL_VECTOR (arg)->data,
453 size_in_chars);
454 return val;
455 }
456
457 if (!CONSP (arg) && !VECTORP (arg) && !STRINGP (arg))
458 wrong_type_argument (Qsequencep, arg);
459
460 return concat (1, &arg, CONSP (arg) ? Lisp_Cons : XTYPE (arg), 0);
461 }
462
463 /* This structure holds information of an argument of `concat' that is
464 a string and has text properties to be copied. */
465 struct textprop_rec
466 {
467 ptrdiff_t argnum; /* refer to ARGS (arguments of `concat') */
468 ptrdiff_t from; /* refer to ARGS[argnum] (argument string) */
469 ptrdiff_t to; /* refer to VAL (the target string) */
470 };
471
472 static Lisp_Object
473 concat (ptrdiff_t nargs, Lisp_Object *args,
474 enum Lisp_Type target_type, int last_special)
475 {
476 Lisp_Object val;
477 register Lisp_Object tail;
478 register Lisp_Object this;
479 ptrdiff_t toindex;
480 ptrdiff_t toindex_byte = 0;
481 register EMACS_INT result_len;
482 register EMACS_INT result_len_byte;
483 ptrdiff_t argnum;
484 Lisp_Object last_tail;
485 Lisp_Object prev;
486 int some_multibyte;
487 /* When we make a multibyte string, we can't copy text properties
488 while concatenating each string because the length of resulting
489 string can't be decided until we finish the whole concatenation.
490 So, we record strings that have text properties to be copied
491 here, and copy the text properties after the concatenation. */
492 struct textprop_rec *textprops = NULL;
493 /* Number of elements in textprops. */
494 ptrdiff_t num_textprops = 0;
495 USE_SAFE_ALLOCA;
496
497 tail = Qnil;
498
499 /* In append, the last arg isn't treated like the others */
500 if (last_special && nargs > 0)
501 {
502 nargs--;
503 last_tail = args[nargs];
504 }
505 else
506 last_tail = Qnil;
507
508 /* Check each argument. */
509 for (argnum = 0; argnum < nargs; argnum++)
510 {
511 this = args[argnum];
512 if (!(CONSP (this) || NILP (this) || VECTORP (this) || STRINGP (this)
513 || COMPILEDP (this) || BOOL_VECTOR_P (this)))
514 wrong_type_argument (Qsequencep, this);
515 }
516
517 /* Compute total length in chars of arguments in RESULT_LEN.
518 If desired output is a string, also compute length in bytes
519 in RESULT_LEN_BYTE, and determine in SOME_MULTIBYTE
520 whether the result should be a multibyte string. */
521 result_len_byte = 0;
522 result_len = 0;
523 some_multibyte = 0;
524 for (argnum = 0; argnum < nargs; argnum++)
525 {
526 EMACS_INT len;
527 this = args[argnum];
528 len = XFASTINT (Flength (this));
529 if (target_type == Lisp_String)
530 {
531 /* We must count the number of bytes needed in the string
532 as well as the number of characters. */
533 ptrdiff_t i;
534 Lisp_Object ch;
535 int c;
536 ptrdiff_t this_len_byte;
537
538 if (VECTORP (this) || COMPILEDP (this))
539 for (i = 0; i < len; i++)
540 {
541 ch = AREF (this, i);
542 CHECK_CHARACTER (ch);
543 c = XFASTINT (ch);
544 this_len_byte = CHAR_BYTES (c);
545 if (STRING_BYTES_BOUND - result_len_byte < this_len_byte)
546 string_overflow ();
547 result_len_byte += this_len_byte;
548 if (! ASCII_CHAR_P (c) && ! CHAR_BYTE8_P (c))
549 some_multibyte = 1;
550 }
551 else if (BOOL_VECTOR_P (this) && XBOOL_VECTOR (this)->size > 0)
552 wrong_type_argument (Qintegerp, Faref (this, make_number (0)));
553 else if (CONSP (this))
554 for (; CONSP (this); this = XCDR (this))
555 {
556 ch = XCAR (this);
557 CHECK_CHARACTER (ch);
558 c = XFASTINT (ch);
559 this_len_byte = CHAR_BYTES (c);
560 if (STRING_BYTES_BOUND - result_len_byte < this_len_byte)
561 string_overflow ();
562 result_len_byte += this_len_byte;
563 if (! ASCII_CHAR_P (c) && ! CHAR_BYTE8_P (c))
564 some_multibyte = 1;
565 }
566 else if (STRINGP (this))
567 {
568 if (STRING_MULTIBYTE (this))
569 {
570 some_multibyte = 1;
571 this_len_byte = SBYTES (this);
572 }
573 else
574 this_len_byte = count_size_as_multibyte (SDATA (this),
575 SCHARS (this));
576 if (STRING_BYTES_BOUND - result_len_byte < this_len_byte)
577 string_overflow ();
578 result_len_byte += this_len_byte;
579 }
580 }
581
582 result_len += len;
583 if (MOST_POSITIVE_FIXNUM < result_len)
584 memory_full (SIZE_MAX);
585 }
586
587 if (! some_multibyte)
588 result_len_byte = result_len;
589
590 /* Create the output object. */
591 if (target_type == Lisp_Cons)
592 val = Fmake_list (make_number (result_len), Qnil);
593 else if (target_type == Lisp_Vectorlike)
594 val = Fmake_vector (make_number (result_len), Qnil);
595 else if (some_multibyte)
596 val = make_uninit_multibyte_string (result_len, result_len_byte);
597 else
598 val = make_uninit_string (result_len);
599
600 /* In `append', if all but last arg are nil, return last arg. */
601 if (target_type == Lisp_Cons && EQ (val, Qnil))
602 return last_tail;
603
604 /* Copy the contents of the args into the result. */
605 if (CONSP (val))
606 tail = val, toindex = -1; /* -1 in toindex is flag we are making a list */
607 else
608 toindex = 0, toindex_byte = 0;
609
610 prev = Qnil;
611 if (STRINGP (val))
612 SAFE_NALLOCA (textprops, 1, nargs);
613
614 for (argnum = 0; argnum < nargs; argnum++)
615 {
616 Lisp_Object thislen;
617 ptrdiff_t thisleni = 0;
618 register ptrdiff_t thisindex = 0;
619 register ptrdiff_t thisindex_byte = 0;
620
621 this = args[argnum];
622 if (!CONSP (this))
623 thislen = Flength (this), thisleni = XINT (thislen);
624
625 /* Between strings of the same kind, copy fast. */
626 if (STRINGP (this) && STRINGP (val)
627 && STRING_MULTIBYTE (this) == some_multibyte)
628 {
629 ptrdiff_t thislen_byte = SBYTES (this);
630
631 memcpy (SDATA (val) + toindex_byte, SDATA (this), SBYTES (this));
632 if (! NULL_INTERVAL_P (STRING_INTERVALS (this)))
633 {
634 textprops[num_textprops].argnum = argnum;
635 textprops[num_textprops].from = 0;
636 textprops[num_textprops++].to = toindex;
637 }
638 toindex_byte += thislen_byte;
639 toindex += thisleni;
640 }
641 /* Copy a single-byte string to a multibyte string. */
642 else if (STRINGP (this) && STRINGP (val))
643 {
644 if (! NULL_INTERVAL_P (STRING_INTERVALS (this)))
645 {
646 textprops[num_textprops].argnum = argnum;
647 textprops[num_textprops].from = 0;
648 textprops[num_textprops++].to = toindex;
649 }
650 toindex_byte += copy_text (SDATA (this),
651 SDATA (val) + toindex_byte,
652 SCHARS (this), 0, 1);
653 toindex += thisleni;
654 }
655 else
656 /* Copy element by element. */
657 while (1)
658 {
659 register Lisp_Object elt;
660
661 /* Fetch next element of `this' arg into `elt', or break if
662 `this' is exhausted. */
663 if (NILP (this)) break;
664 if (CONSP (this))
665 elt = XCAR (this), this = XCDR (this);
666 else if (thisindex >= thisleni)
667 break;
668 else if (STRINGP (this))
669 {
670 int c;
671 if (STRING_MULTIBYTE (this))
672 FETCH_STRING_CHAR_ADVANCE_NO_CHECK (c, this,
673 thisindex,
674 thisindex_byte);
675 else
676 {
677 c = SREF (this, thisindex); thisindex++;
678 if (some_multibyte && !ASCII_CHAR_P (c))
679 c = BYTE8_TO_CHAR (c);
680 }
681 XSETFASTINT (elt, c);
682 }
683 else if (BOOL_VECTOR_P (this))
684 {
685 int byte;
686 byte = XBOOL_VECTOR (this)->data[thisindex / BOOL_VECTOR_BITS_PER_CHAR];
687 if (byte & (1 << (thisindex % BOOL_VECTOR_BITS_PER_CHAR)))
688 elt = Qt;
689 else
690 elt = Qnil;
691 thisindex++;
692 }
693 else
694 {
695 elt = AREF (this, thisindex);
696 thisindex++;
697 }
698
699 /* Store this element into the result. */
700 if (toindex < 0)
701 {
702 XSETCAR (tail, elt);
703 prev = tail;
704 tail = XCDR (tail);
705 }
706 else if (VECTORP (val))
707 {
708 ASET (val, toindex, elt);
709 toindex++;
710 }
711 else
712 {
713 int c;
714 CHECK_CHARACTER (elt);
715 c = XFASTINT (elt);
716 if (some_multibyte)
717 toindex_byte += CHAR_STRING (c, SDATA (val) + toindex_byte);
718 else
719 SSET (val, toindex_byte++, c);
720 toindex++;
721 }
722 }
723 }
724 if (!NILP (prev))
725 XSETCDR (prev, last_tail);
726
727 if (num_textprops > 0)
728 {
729 Lisp_Object props;
730 ptrdiff_t last_to_end = -1;
731
732 for (argnum = 0; argnum < num_textprops; argnum++)
733 {
734 this = args[textprops[argnum].argnum];
735 props = text_property_list (this,
736 make_number (0),
737 make_number (SCHARS (this)),
738 Qnil);
739 /* If successive arguments have properties, be sure that the
740 value of `composition' property be the copy. */
741 if (last_to_end == textprops[argnum].to)
742 make_composition_value_copy (props);
743 add_text_properties_from_list (val, props,
744 make_number (textprops[argnum].to));
745 last_to_end = textprops[argnum].to + SCHARS (this);
746 }
747 }
748
749 SAFE_FREE ();
750 return val;
751 }
752 \f
753 static Lisp_Object string_char_byte_cache_string;
754 static ptrdiff_t string_char_byte_cache_charpos;
755 static ptrdiff_t string_char_byte_cache_bytepos;
756
757 void
758 clear_string_char_byte_cache (void)
759 {
760 string_char_byte_cache_string = Qnil;
761 }
762
763 /* Return the byte index corresponding to CHAR_INDEX in STRING. */
764
765 ptrdiff_t
766 string_char_to_byte (Lisp_Object string, ptrdiff_t char_index)
767 {
768 ptrdiff_t i_byte;
769 ptrdiff_t best_below, best_below_byte;
770 ptrdiff_t best_above, best_above_byte;
771
772 best_below = best_below_byte = 0;
773 best_above = SCHARS (string);
774 best_above_byte = SBYTES (string);
775 if (best_above == best_above_byte)
776 return char_index;
777
778 if (EQ (string, string_char_byte_cache_string))
779 {
780 if (string_char_byte_cache_charpos < char_index)
781 {
782 best_below = string_char_byte_cache_charpos;
783 best_below_byte = string_char_byte_cache_bytepos;
784 }
785 else
786 {
787 best_above = string_char_byte_cache_charpos;
788 best_above_byte = string_char_byte_cache_bytepos;
789 }
790 }
791
792 if (char_index - best_below < best_above - char_index)
793 {
794 unsigned char *p = SDATA (string) + best_below_byte;
795
796 while (best_below < char_index)
797 {
798 p += BYTES_BY_CHAR_HEAD (*p);
799 best_below++;
800 }
801 i_byte = p - SDATA (string);
802 }
803 else
804 {
805 unsigned char *p = SDATA (string) + best_above_byte;
806
807 while (best_above > char_index)
808 {
809 p--;
810 while (!CHAR_HEAD_P (*p)) p--;
811 best_above--;
812 }
813 i_byte = p - SDATA (string);
814 }
815
816 string_char_byte_cache_bytepos = i_byte;
817 string_char_byte_cache_charpos = char_index;
818 string_char_byte_cache_string = string;
819
820 return i_byte;
821 }
822 \f
823 /* Return the character index corresponding to BYTE_INDEX in STRING. */
824
825 ptrdiff_t
826 string_byte_to_char (Lisp_Object string, ptrdiff_t byte_index)
827 {
828 ptrdiff_t i, i_byte;
829 ptrdiff_t best_below, best_below_byte;
830 ptrdiff_t best_above, best_above_byte;
831
832 best_below = best_below_byte = 0;
833 best_above = SCHARS (string);
834 best_above_byte = SBYTES (string);
835 if (best_above == best_above_byte)
836 return byte_index;
837
838 if (EQ (string, string_char_byte_cache_string))
839 {
840 if (string_char_byte_cache_bytepos < byte_index)
841 {
842 best_below = string_char_byte_cache_charpos;
843 best_below_byte = string_char_byte_cache_bytepos;
844 }
845 else
846 {
847 best_above = string_char_byte_cache_charpos;
848 best_above_byte = string_char_byte_cache_bytepos;
849 }
850 }
851
852 if (byte_index - best_below_byte < best_above_byte - byte_index)
853 {
854 unsigned char *p = SDATA (string) + best_below_byte;
855 unsigned char *pend = SDATA (string) + byte_index;
856
857 while (p < pend)
858 {
859 p += BYTES_BY_CHAR_HEAD (*p);
860 best_below++;
861 }
862 i = best_below;
863 i_byte = p - SDATA (string);
864 }
865 else
866 {
867 unsigned char *p = SDATA (string) + best_above_byte;
868 unsigned char *pbeg = SDATA (string) + byte_index;
869
870 while (p > pbeg)
871 {
872 p--;
873 while (!CHAR_HEAD_P (*p)) p--;
874 best_above--;
875 }
876 i = best_above;
877 i_byte = p - SDATA (string);
878 }
879
880 string_char_byte_cache_bytepos = i_byte;
881 string_char_byte_cache_charpos = i;
882 string_char_byte_cache_string = string;
883
884 return i;
885 }
886 \f
887 /* Convert STRING to a multibyte string. */
888
889 static Lisp_Object
890 string_make_multibyte (Lisp_Object string)
891 {
892 unsigned char *buf;
893 ptrdiff_t nbytes;
894 Lisp_Object ret;
895 USE_SAFE_ALLOCA;
896
897 if (STRING_MULTIBYTE (string))
898 return string;
899
900 nbytes = count_size_as_multibyte (SDATA (string),
901 SCHARS (string));
902 /* If all the chars are ASCII, they won't need any more bytes
903 once converted. In that case, we can return STRING itself. */
904 if (nbytes == SBYTES (string))
905 return string;
906
907 SAFE_ALLOCA (buf, unsigned char *, nbytes);
908 copy_text (SDATA (string), buf, SBYTES (string),
909 0, 1);
910
911 ret = make_multibyte_string ((char *) buf, SCHARS (string), nbytes);
912 SAFE_FREE ();
913
914 return ret;
915 }
916
917
918 /* Convert STRING (if unibyte) to a multibyte string without changing
919 the number of characters. Characters 0200 trough 0237 are
920 converted to eight-bit characters. */
921
922 Lisp_Object
923 string_to_multibyte (Lisp_Object string)
924 {
925 unsigned char *buf;
926 ptrdiff_t nbytes;
927 Lisp_Object ret;
928 USE_SAFE_ALLOCA;
929
930 if (STRING_MULTIBYTE (string))
931 return string;
932
933 nbytes = count_size_as_multibyte (SDATA (string), SBYTES (string));
934 /* If all the chars are ASCII, they won't need any more bytes once
935 converted. */
936 if (nbytes == SBYTES (string))
937 return make_multibyte_string (SSDATA (string), nbytes, nbytes);
938
939 SAFE_ALLOCA (buf, unsigned char *, nbytes);
940 memcpy (buf, SDATA (string), SBYTES (string));
941 str_to_multibyte (buf, nbytes, SBYTES (string));
942
943 ret = make_multibyte_string ((char *) buf, SCHARS (string), nbytes);
944 SAFE_FREE ();
945
946 return ret;
947 }
948
949
950 /* Convert STRING to a single-byte string. */
951
952 Lisp_Object
953 string_make_unibyte (Lisp_Object string)
954 {
955 ptrdiff_t nchars;
956 unsigned char *buf;
957 Lisp_Object ret;
958 USE_SAFE_ALLOCA;
959
960 if (! STRING_MULTIBYTE (string))
961 return string;
962
963 nchars = SCHARS (string);
964
965 SAFE_ALLOCA (buf, unsigned char *, nchars);
966 copy_text (SDATA (string), buf, SBYTES (string),
967 1, 0);
968
969 ret = make_unibyte_string ((char *) buf, nchars);
970 SAFE_FREE ();
971
972 return ret;
973 }
974
975 DEFUN ("string-make-multibyte", Fstring_make_multibyte, Sstring_make_multibyte,
976 1, 1, 0,
977 doc: /* Return the multibyte equivalent of STRING.
978 If STRING is unibyte and contains non-ASCII characters, the function
979 `unibyte-char-to-multibyte' is used to convert each unibyte character
980 to a multibyte character. In this case, the returned string is a
981 newly created string with no text properties. If STRING is multibyte
982 or entirely ASCII, it is returned unchanged. In particular, when
983 STRING is unibyte and entirely ASCII, the returned string is unibyte.
984 \(When the characters are all ASCII, Emacs primitives will treat the
985 string the same way whether it is unibyte or multibyte.) */)
986 (Lisp_Object string)
987 {
988 CHECK_STRING (string);
989
990 return string_make_multibyte (string);
991 }
992
993 DEFUN ("string-make-unibyte", Fstring_make_unibyte, Sstring_make_unibyte,
994 1, 1, 0,
995 doc: /* Return the unibyte equivalent of STRING.
996 Multibyte character codes are converted to unibyte according to
997 `nonascii-translation-table' or, if that is nil, `nonascii-insert-offset'.
998 If the lookup in the translation table fails, this function takes just
999 the low 8 bits of each character. */)
1000 (Lisp_Object string)
1001 {
1002 CHECK_STRING (string);
1003
1004 return string_make_unibyte (string);
1005 }
1006
1007 DEFUN ("string-as-unibyte", Fstring_as_unibyte, Sstring_as_unibyte,
1008 1, 1, 0,
1009 doc: /* Return a unibyte string with the same individual bytes as STRING.
1010 If STRING is unibyte, the result is STRING itself.
1011 Otherwise it is a newly created string, with no text properties.
1012 If STRING is multibyte and contains a character of charset
1013 `eight-bit', it is converted to the corresponding single byte. */)
1014 (Lisp_Object string)
1015 {
1016 CHECK_STRING (string);
1017
1018 if (STRING_MULTIBYTE (string))
1019 {
1020 ptrdiff_t bytes = SBYTES (string);
1021 unsigned char *str = xmalloc (bytes);
1022
1023 memcpy (str, SDATA (string), bytes);
1024 bytes = str_as_unibyte (str, bytes);
1025 string = make_unibyte_string ((char *) str, bytes);
1026 xfree (str);
1027 }
1028 return string;
1029 }
1030
1031 DEFUN ("string-as-multibyte", Fstring_as_multibyte, Sstring_as_multibyte,
1032 1, 1, 0,
1033 doc: /* Return a multibyte string with the same individual bytes as STRING.
1034 If STRING is multibyte, the result is STRING itself.
1035 Otherwise it is a newly created string, with no text properties.
1036
1037 If STRING is unibyte and contains an individual 8-bit byte (i.e. not
1038 part of a correct utf-8 sequence), it is converted to the corresponding
1039 multibyte character of charset `eight-bit'.
1040 See also `string-to-multibyte'.
1041
1042 Beware, this often doesn't really do what you think it does.
1043 It is similar to (decode-coding-string STRING 'utf-8-emacs).
1044 If you're not sure, whether to use `string-as-multibyte' or
1045 `string-to-multibyte', use `string-to-multibyte'. */)
1046 (Lisp_Object string)
1047 {
1048 CHECK_STRING (string);
1049
1050 if (! STRING_MULTIBYTE (string))
1051 {
1052 Lisp_Object new_string;
1053 ptrdiff_t nchars, nbytes;
1054
1055 parse_str_as_multibyte (SDATA (string),
1056 SBYTES (string),
1057 &nchars, &nbytes);
1058 new_string = make_uninit_multibyte_string (nchars, nbytes);
1059 memcpy (SDATA (new_string), SDATA (string), SBYTES (string));
1060 if (nbytes != SBYTES (string))
1061 str_as_multibyte (SDATA (new_string), nbytes,
1062 SBYTES (string), NULL);
1063 string = new_string;
1064 STRING_SET_INTERVALS (string, NULL_INTERVAL);
1065 }
1066 return string;
1067 }
1068
1069 DEFUN ("string-to-multibyte", Fstring_to_multibyte, Sstring_to_multibyte,
1070 1, 1, 0,
1071 doc: /* Return a multibyte string with the same individual chars as STRING.
1072 If STRING is multibyte, the result is STRING itself.
1073 Otherwise it is a newly created string, with no text properties.
1074
1075 If STRING is unibyte and contains an 8-bit byte, it is converted to
1076 the corresponding multibyte character of charset `eight-bit'.
1077
1078 This differs from `string-as-multibyte' by converting each byte of a correct
1079 utf-8 sequence to an eight-bit character, not just bytes that don't form a
1080 correct sequence. */)
1081 (Lisp_Object string)
1082 {
1083 CHECK_STRING (string);
1084
1085 return string_to_multibyte (string);
1086 }
1087
1088 DEFUN ("string-to-unibyte", Fstring_to_unibyte, Sstring_to_unibyte,
1089 1, 1, 0,
1090 doc: /* Return a unibyte string with the same individual chars as STRING.
1091 If STRING is unibyte, the result is STRING itself.
1092 Otherwise it is a newly created string, with no text properties,
1093 where each `eight-bit' character is converted to the corresponding byte.
1094 If STRING contains a non-ASCII, non-`eight-bit' character,
1095 an error is signaled. */)
1096 (Lisp_Object string)
1097 {
1098 CHECK_STRING (string);
1099
1100 if (STRING_MULTIBYTE (string))
1101 {
1102 ptrdiff_t chars = SCHARS (string);
1103 unsigned char *str = xmalloc (chars);
1104 ptrdiff_t converted = str_to_unibyte (SDATA (string), str, chars, 0);
1105
1106 if (converted < chars)
1107 error ("Can't convert the %"pD"dth character to unibyte", converted);
1108 string = make_unibyte_string ((char *) str, chars);
1109 xfree (str);
1110 }
1111 return string;
1112 }
1113
1114 \f
1115 DEFUN ("copy-alist", Fcopy_alist, Scopy_alist, 1, 1, 0,
1116 doc: /* Return a copy of ALIST.
1117 This is an alist which represents the same mapping from objects to objects,
1118 but does not share the alist structure with ALIST.
1119 The objects mapped (cars and cdrs of elements of the alist)
1120 are shared, however.
1121 Elements of ALIST that are not conses are also shared. */)
1122 (Lisp_Object alist)
1123 {
1124 register Lisp_Object tem;
1125
1126 CHECK_LIST (alist);
1127 if (NILP (alist))
1128 return alist;
1129 alist = concat (1, &alist, Lisp_Cons, 0);
1130 for (tem = alist; CONSP (tem); tem = XCDR (tem))
1131 {
1132 register Lisp_Object car;
1133 car = XCAR (tem);
1134
1135 if (CONSP (car))
1136 XSETCAR (tem, Fcons (XCAR (car), XCDR (car)));
1137 }
1138 return alist;
1139 }
1140
1141 DEFUN ("substring", Fsubstring, Ssubstring, 2, 3, 0,
1142 doc: /* Return a new string whose contents are a substring of STRING.
1143 The returned string consists of the characters between index FROM
1144 \(inclusive) and index TO (exclusive) of STRING. FROM and TO are
1145 zero-indexed: 0 means the first character of STRING. Negative values
1146 are counted from the end of STRING. If TO is nil, the substring runs
1147 to the end of STRING.
1148
1149 The STRING argument may also be a vector. In that case, the return
1150 value is a new vector that contains the elements between index FROM
1151 \(inclusive) and index TO (exclusive) of that vector argument. */)
1152 (Lisp_Object string, register Lisp_Object from, Lisp_Object to)
1153 {
1154 Lisp_Object res;
1155 ptrdiff_t size;
1156 EMACS_INT from_char, to_char;
1157
1158 CHECK_VECTOR_OR_STRING (string);
1159 CHECK_NUMBER (from);
1160
1161 if (STRINGP (string))
1162 size = SCHARS (string);
1163 else
1164 size = ASIZE (string);
1165
1166 if (NILP (to))
1167 to_char = size;
1168 else
1169 {
1170 CHECK_NUMBER (to);
1171
1172 to_char = XINT (to);
1173 if (to_char < 0)
1174 to_char += size;
1175 }
1176
1177 from_char = XINT (from);
1178 if (from_char < 0)
1179 from_char += size;
1180 if (!(0 <= from_char && from_char <= to_char && to_char <= size))
1181 args_out_of_range_3 (string, make_number (from_char),
1182 make_number (to_char));
1183
1184 if (STRINGP (string))
1185 {
1186 ptrdiff_t to_byte =
1187 (NILP (to) ? SBYTES (string) : string_char_to_byte (string, to_char));
1188 ptrdiff_t from_byte = string_char_to_byte (string, from_char);
1189 res = make_specified_string (SSDATA (string) + from_byte,
1190 to_char - from_char, to_byte - from_byte,
1191 STRING_MULTIBYTE (string));
1192 copy_text_properties (make_number (from_char), make_number (to_char),
1193 string, make_number (0), res, Qnil);
1194 }
1195 else
1196 res = Fvector (to_char - from_char, &AREF (string, from_char));
1197
1198 return res;
1199 }
1200
1201
1202 DEFUN ("substring-no-properties", Fsubstring_no_properties, Ssubstring_no_properties, 1, 3, 0,
1203 doc: /* Return a substring of STRING, without text properties.
1204 It starts at index FROM and ends before TO.
1205 TO may be nil or omitted; then the substring runs to the end of STRING.
1206 If FROM is nil or omitted, the substring starts at the beginning of STRING.
1207 If FROM or TO is negative, it counts from the end.
1208
1209 With one argument, just copy STRING without its properties. */)
1210 (Lisp_Object string, register Lisp_Object from, Lisp_Object to)
1211 {
1212 ptrdiff_t size;
1213 EMACS_INT from_char, to_char;
1214 ptrdiff_t from_byte, to_byte;
1215
1216 CHECK_STRING (string);
1217
1218 size = SCHARS (string);
1219
1220 if (NILP (from))
1221 from_char = 0;
1222 else
1223 {
1224 CHECK_NUMBER (from);
1225 from_char = XINT (from);
1226 if (from_char < 0)
1227 from_char += size;
1228 }
1229
1230 if (NILP (to))
1231 to_char = size;
1232 else
1233 {
1234 CHECK_NUMBER (to);
1235 to_char = XINT (to);
1236 if (to_char < 0)
1237 to_char += size;
1238 }
1239
1240 if (!(0 <= from_char && from_char <= to_char && to_char <= size))
1241 args_out_of_range_3 (string, make_number (from_char),
1242 make_number (to_char));
1243
1244 from_byte = NILP (from) ? 0 : string_char_to_byte (string, from_char);
1245 to_byte =
1246 NILP (to) ? SBYTES (string) : string_char_to_byte (string, to_char);
1247 return make_specified_string (SSDATA (string) + from_byte,
1248 to_char - from_char, to_byte - from_byte,
1249 STRING_MULTIBYTE (string));
1250 }
1251
1252 /* Extract a substring of STRING, giving start and end positions
1253 both in characters and in bytes. */
1254
1255 Lisp_Object
1256 substring_both (Lisp_Object string, ptrdiff_t from, ptrdiff_t from_byte,
1257 ptrdiff_t to, ptrdiff_t to_byte)
1258 {
1259 Lisp_Object res;
1260 ptrdiff_t size;
1261
1262 CHECK_VECTOR_OR_STRING (string);
1263
1264 size = STRINGP (string) ? SCHARS (string) : ASIZE (string);
1265
1266 if (!(0 <= from && from <= to && to <= size))
1267 args_out_of_range_3 (string, make_number (from), make_number (to));
1268
1269 if (STRINGP (string))
1270 {
1271 res = make_specified_string (SSDATA (string) + from_byte,
1272 to - from, to_byte - from_byte,
1273 STRING_MULTIBYTE (string));
1274 copy_text_properties (make_number (from), make_number (to),
1275 string, make_number (0), res, Qnil);
1276 }
1277 else
1278 res = Fvector (to - from, &AREF (string, from));
1279
1280 return res;
1281 }
1282 \f
1283 DEFUN ("nthcdr", Fnthcdr, Snthcdr, 2, 2, 0,
1284 doc: /* Take cdr N times on LIST, return the result. */)
1285 (Lisp_Object n, Lisp_Object list)
1286 {
1287 EMACS_INT i, num;
1288 CHECK_NUMBER (n);
1289 num = XINT (n);
1290 for (i = 0; i < num && !NILP (list); i++)
1291 {
1292 QUIT;
1293 CHECK_LIST_CONS (list, list);
1294 list = XCDR (list);
1295 }
1296 return list;
1297 }
1298
1299 DEFUN ("nth", Fnth, Snth, 2, 2, 0,
1300 doc: /* Return the Nth element of LIST.
1301 N counts from zero. If LIST is not that long, nil is returned. */)
1302 (Lisp_Object n, Lisp_Object list)
1303 {
1304 return Fcar (Fnthcdr (n, list));
1305 }
1306
1307 DEFUN ("elt", Felt, Selt, 2, 2, 0,
1308 doc: /* Return element of SEQUENCE at index N. */)
1309 (register Lisp_Object sequence, Lisp_Object n)
1310 {
1311 CHECK_NUMBER (n);
1312 if (CONSP (sequence) || NILP (sequence))
1313 return Fcar (Fnthcdr (n, sequence));
1314
1315 /* Faref signals a "not array" error, so check here. */
1316 CHECK_ARRAY (sequence, Qsequencep);
1317 return Faref (sequence, n);
1318 }
1319
1320 DEFUN ("member", Fmember, Smember, 2, 2, 0,
1321 doc: /* Return non-nil if ELT is an element of LIST. Comparison done with `equal'.
1322 The value is actually the tail of LIST whose car is ELT. */)
1323 (register Lisp_Object elt, Lisp_Object list)
1324 {
1325 register Lisp_Object tail;
1326 for (tail = list; CONSP (tail); tail = XCDR (tail))
1327 {
1328 register Lisp_Object tem;
1329 CHECK_LIST_CONS (tail, list);
1330 tem = XCAR (tail);
1331 if (! NILP (Fequal (elt, tem)))
1332 return tail;
1333 QUIT;
1334 }
1335 return Qnil;
1336 }
1337
1338 DEFUN ("memq", Fmemq, Smemq, 2, 2, 0,
1339 doc: /* Return non-nil if ELT is an element of LIST. Comparison done with `eq'.
1340 The value is actually the tail of LIST whose car is ELT. */)
1341 (register Lisp_Object elt, Lisp_Object list)
1342 {
1343 while (1)
1344 {
1345 if (!CONSP (list) || EQ (XCAR (list), elt))
1346 break;
1347
1348 list = XCDR (list);
1349 if (!CONSP (list) || EQ (XCAR (list), elt))
1350 break;
1351
1352 list = XCDR (list);
1353 if (!CONSP (list) || EQ (XCAR (list), elt))
1354 break;
1355
1356 list = XCDR (list);
1357 QUIT;
1358 }
1359
1360 CHECK_LIST (list);
1361 return list;
1362 }
1363
1364 DEFUN ("memql", Fmemql, Smemql, 2, 2, 0,
1365 doc: /* Return non-nil if ELT is an element of LIST. Comparison done with `eql'.
1366 The value is actually the tail of LIST whose car is ELT. */)
1367 (register Lisp_Object elt, Lisp_Object list)
1368 {
1369 register Lisp_Object tail;
1370
1371 if (!FLOATP (elt))
1372 return Fmemq (elt, list);
1373
1374 for (tail = list; CONSP (tail); tail = XCDR (tail))
1375 {
1376 register Lisp_Object tem;
1377 CHECK_LIST_CONS (tail, list);
1378 tem = XCAR (tail);
1379 if (FLOATP (tem) && internal_equal (elt, tem, 0, 0))
1380 return tail;
1381 QUIT;
1382 }
1383 return Qnil;
1384 }
1385
1386 DEFUN ("assq", Fassq, Sassq, 2, 2, 0,
1387 doc: /* Return non-nil if KEY is `eq' to the car of an element of LIST.
1388 The value is actually the first element of LIST whose car is KEY.
1389 Elements of LIST that are not conses are ignored. */)
1390 (Lisp_Object key, Lisp_Object list)
1391 {
1392 while (1)
1393 {
1394 if (!CONSP (list)
1395 || (CONSP (XCAR (list))
1396 && EQ (XCAR (XCAR (list)), key)))
1397 break;
1398
1399 list = XCDR (list);
1400 if (!CONSP (list)
1401 || (CONSP (XCAR (list))
1402 && EQ (XCAR (XCAR (list)), key)))
1403 break;
1404
1405 list = XCDR (list);
1406 if (!CONSP (list)
1407 || (CONSP (XCAR (list))
1408 && EQ (XCAR (XCAR (list)), key)))
1409 break;
1410
1411 list = XCDR (list);
1412 QUIT;
1413 }
1414
1415 return CAR (list);
1416 }
1417
1418 /* Like Fassq but never report an error and do not allow quits.
1419 Use only on lists known never to be circular. */
1420
1421 Lisp_Object
1422 assq_no_quit (Lisp_Object key, Lisp_Object list)
1423 {
1424 while (CONSP (list)
1425 && (!CONSP (XCAR (list))
1426 || !EQ (XCAR (XCAR (list)), key)))
1427 list = XCDR (list);
1428
1429 return CAR_SAFE (list);
1430 }
1431
1432 DEFUN ("assoc", Fassoc, Sassoc, 2, 2, 0,
1433 doc: /* Return non-nil if KEY is `equal' to the car of an element of LIST.
1434 The value is actually the first element of LIST whose car equals KEY. */)
1435 (Lisp_Object key, Lisp_Object list)
1436 {
1437 Lisp_Object car;
1438
1439 while (1)
1440 {
1441 if (!CONSP (list)
1442 || (CONSP (XCAR (list))
1443 && (car = XCAR (XCAR (list)),
1444 EQ (car, key) || !NILP (Fequal (car, key)))))
1445 break;
1446
1447 list = XCDR (list);
1448 if (!CONSP (list)
1449 || (CONSP (XCAR (list))
1450 && (car = XCAR (XCAR (list)),
1451 EQ (car, key) || !NILP (Fequal (car, key)))))
1452 break;
1453
1454 list = XCDR (list);
1455 if (!CONSP (list)
1456 || (CONSP (XCAR (list))
1457 && (car = XCAR (XCAR (list)),
1458 EQ (car, key) || !NILP (Fequal (car, key)))))
1459 break;
1460
1461 list = XCDR (list);
1462 QUIT;
1463 }
1464
1465 return CAR (list);
1466 }
1467
1468 /* Like Fassoc but never report an error and do not allow quits.
1469 Use only on lists known never to be circular. */
1470
1471 Lisp_Object
1472 assoc_no_quit (Lisp_Object key, Lisp_Object list)
1473 {
1474 while (CONSP (list)
1475 && (!CONSP (XCAR (list))
1476 || (!EQ (XCAR (XCAR (list)), key)
1477 && NILP (Fequal (XCAR (XCAR (list)), key)))))
1478 list = XCDR (list);
1479
1480 return CONSP (list) ? XCAR (list) : Qnil;
1481 }
1482
1483 DEFUN ("rassq", Frassq, Srassq, 2, 2, 0,
1484 doc: /* Return non-nil if KEY is `eq' to the cdr of an element of LIST.
1485 The value is actually the first element of LIST whose cdr is KEY. */)
1486 (register Lisp_Object key, Lisp_Object list)
1487 {
1488 while (1)
1489 {
1490 if (!CONSP (list)
1491 || (CONSP (XCAR (list))
1492 && EQ (XCDR (XCAR (list)), key)))
1493 break;
1494
1495 list = XCDR (list);
1496 if (!CONSP (list)
1497 || (CONSP (XCAR (list))
1498 && EQ (XCDR (XCAR (list)), key)))
1499 break;
1500
1501 list = XCDR (list);
1502 if (!CONSP (list)
1503 || (CONSP (XCAR (list))
1504 && EQ (XCDR (XCAR (list)), key)))
1505 break;
1506
1507 list = XCDR (list);
1508 QUIT;
1509 }
1510
1511 return CAR (list);
1512 }
1513
1514 DEFUN ("rassoc", Frassoc, Srassoc, 2, 2, 0,
1515 doc: /* Return non-nil if KEY is `equal' to the cdr of an element of LIST.
1516 The value is actually the first element of LIST whose cdr equals KEY. */)
1517 (Lisp_Object key, Lisp_Object list)
1518 {
1519 Lisp_Object cdr;
1520
1521 while (1)
1522 {
1523 if (!CONSP (list)
1524 || (CONSP (XCAR (list))
1525 && (cdr = XCDR (XCAR (list)),
1526 EQ (cdr, key) || !NILP (Fequal (cdr, key)))))
1527 break;
1528
1529 list = XCDR (list);
1530 if (!CONSP (list)
1531 || (CONSP (XCAR (list))
1532 && (cdr = XCDR (XCAR (list)),
1533 EQ (cdr, key) || !NILP (Fequal (cdr, key)))))
1534 break;
1535
1536 list = XCDR (list);
1537 if (!CONSP (list)
1538 || (CONSP (XCAR (list))
1539 && (cdr = XCDR (XCAR (list)),
1540 EQ (cdr, key) || !NILP (Fequal (cdr, key)))))
1541 break;
1542
1543 list = XCDR (list);
1544 QUIT;
1545 }
1546
1547 return CAR (list);
1548 }
1549 \f
1550 DEFUN ("delq", Fdelq, Sdelq, 2, 2, 0,
1551 doc: /* Delete by side effect any occurrences of ELT as a member of LIST.
1552 The modified LIST is returned. Comparison is done with `eq'.
1553 If the first member of LIST is ELT, there is no way to remove it by side effect;
1554 therefore, write `(setq foo (delq element foo))'
1555 to be sure of changing the value of `foo'. */)
1556 (register Lisp_Object elt, Lisp_Object list)
1557 {
1558 register Lisp_Object tail, prev;
1559 register Lisp_Object tem;
1560
1561 tail = list;
1562 prev = Qnil;
1563 while (!NILP (tail))
1564 {
1565 CHECK_LIST_CONS (tail, list);
1566 tem = XCAR (tail);
1567 if (EQ (elt, tem))
1568 {
1569 if (NILP (prev))
1570 list = XCDR (tail);
1571 else
1572 Fsetcdr (prev, XCDR (tail));
1573 }
1574 else
1575 prev = tail;
1576 tail = XCDR (tail);
1577 QUIT;
1578 }
1579 return list;
1580 }
1581
1582 DEFUN ("delete", Fdelete, Sdelete, 2, 2, 0,
1583 doc: /* Delete by side effect any occurrences of ELT as a member of SEQ.
1584 SEQ must be a list, a vector, or a string.
1585 The modified SEQ is returned. Comparison is done with `equal'.
1586 If SEQ is not a list, or the first member of SEQ is ELT, deleting it
1587 is not a side effect; it is simply using a different sequence.
1588 Therefore, write `(setq foo (delete element foo))'
1589 to be sure of changing the value of `foo'. */)
1590 (Lisp_Object elt, Lisp_Object seq)
1591 {
1592 if (VECTORP (seq))
1593 {
1594 ptrdiff_t i, n;
1595
1596 for (i = n = 0; i < ASIZE (seq); ++i)
1597 if (NILP (Fequal (AREF (seq, i), elt)))
1598 ++n;
1599
1600 if (n != ASIZE (seq))
1601 {
1602 struct Lisp_Vector *p = allocate_vector (n);
1603
1604 for (i = n = 0; i < ASIZE (seq); ++i)
1605 if (NILP (Fequal (AREF (seq, i), elt)))
1606 p->contents[n++] = AREF (seq, i);
1607
1608 XSETVECTOR (seq, p);
1609 }
1610 }
1611 else if (STRINGP (seq))
1612 {
1613 ptrdiff_t i, ibyte, nchars, nbytes, cbytes;
1614 int c;
1615
1616 for (i = nchars = nbytes = ibyte = 0;
1617 i < SCHARS (seq);
1618 ++i, ibyte += cbytes)
1619 {
1620 if (STRING_MULTIBYTE (seq))
1621 {
1622 c = STRING_CHAR (SDATA (seq) + ibyte);
1623 cbytes = CHAR_BYTES (c);
1624 }
1625 else
1626 {
1627 c = SREF (seq, i);
1628 cbytes = 1;
1629 }
1630
1631 if (!INTEGERP (elt) || c != XINT (elt))
1632 {
1633 ++nchars;
1634 nbytes += cbytes;
1635 }
1636 }
1637
1638 if (nchars != SCHARS (seq))
1639 {
1640 Lisp_Object tem;
1641
1642 tem = make_uninit_multibyte_string (nchars, nbytes);
1643 if (!STRING_MULTIBYTE (seq))
1644 STRING_SET_UNIBYTE (tem);
1645
1646 for (i = nchars = nbytes = ibyte = 0;
1647 i < SCHARS (seq);
1648 ++i, ibyte += cbytes)
1649 {
1650 if (STRING_MULTIBYTE (seq))
1651 {
1652 c = STRING_CHAR (SDATA (seq) + ibyte);
1653 cbytes = CHAR_BYTES (c);
1654 }
1655 else
1656 {
1657 c = SREF (seq, i);
1658 cbytes = 1;
1659 }
1660
1661 if (!INTEGERP (elt) || c != XINT (elt))
1662 {
1663 unsigned char *from = SDATA (seq) + ibyte;
1664 unsigned char *to = SDATA (tem) + nbytes;
1665 ptrdiff_t n;
1666
1667 ++nchars;
1668 nbytes += cbytes;
1669
1670 for (n = cbytes; n--; )
1671 *to++ = *from++;
1672 }
1673 }
1674
1675 seq = tem;
1676 }
1677 }
1678 else
1679 {
1680 Lisp_Object tail, prev;
1681
1682 for (tail = seq, prev = Qnil; CONSP (tail); tail = XCDR (tail))
1683 {
1684 CHECK_LIST_CONS (tail, seq);
1685
1686 if (!NILP (Fequal (elt, XCAR (tail))))
1687 {
1688 if (NILP (prev))
1689 seq = XCDR (tail);
1690 else
1691 Fsetcdr (prev, XCDR (tail));
1692 }
1693 else
1694 prev = tail;
1695 QUIT;
1696 }
1697 }
1698
1699 return seq;
1700 }
1701
1702 DEFUN ("nreverse", Fnreverse, Snreverse, 1, 1, 0,
1703 doc: /* Reverse LIST by modifying cdr pointers.
1704 Return the reversed list. */)
1705 (Lisp_Object list)
1706 {
1707 register Lisp_Object prev, tail, next;
1708
1709 if (NILP (list)) return list;
1710 prev = Qnil;
1711 tail = list;
1712 while (!NILP (tail))
1713 {
1714 QUIT;
1715 CHECK_LIST_CONS (tail, list);
1716 next = XCDR (tail);
1717 Fsetcdr (tail, prev);
1718 prev = tail;
1719 tail = next;
1720 }
1721 return prev;
1722 }
1723
1724 DEFUN ("reverse", Freverse, Sreverse, 1, 1, 0,
1725 doc: /* Reverse LIST, copying. Return the reversed list.
1726 See also the function `nreverse', which is used more often. */)
1727 (Lisp_Object list)
1728 {
1729 Lisp_Object new;
1730
1731 for (new = Qnil; CONSP (list); list = XCDR (list))
1732 {
1733 QUIT;
1734 new = Fcons (XCAR (list), new);
1735 }
1736 CHECK_LIST_END (list, list);
1737 return new;
1738 }
1739 \f
1740 Lisp_Object merge (Lisp_Object org_l1, Lisp_Object org_l2, Lisp_Object pred);
1741
1742 DEFUN ("sort", Fsort, Ssort, 2, 2, 0,
1743 doc: /* Sort LIST, stably, comparing elements using PREDICATE.
1744 Returns the sorted list. LIST is modified by side effects.
1745 PREDICATE is called with two elements of LIST, and should return non-nil
1746 if the first element should sort before the second. */)
1747 (Lisp_Object list, Lisp_Object predicate)
1748 {
1749 Lisp_Object front, back;
1750 register Lisp_Object len, tem;
1751 struct gcpro gcpro1, gcpro2;
1752 EMACS_INT length;
1753
1754 front = list;
1755 len = Flength (list);
1756 length = XINT (len);
1757 if (length < 2)
1758 return list;
1759
1760 XSETINT (len, (length / 2) - 1);
1761 tem = Fnthcdr (len, list);
1762 back = Fcdr (tem);
1763 Fsetcdr (tem, Qnil);
1764
1765 GCPRO2 (front, back);
1766 front = Fsort (front, predicate);
1767 back = Fsort (back, predicate);
1768 UNGCPRO;
1769 return merge (front, back, predicate);
1770 }
1771
1772 Lisp_Object
1773 merge (Lisp_Object org_l1, Lisp_Object org_l2, Lisp_Object pred)
1774 {
1775 Lisp_Object value;
1776 register Lisp_Object tail;
1777 Lisp_Object tem;
1778 register Lisp_Object l1, l2;
1779 struct gcpro gcpro1, gcpro2, gcpro3, gcpro4;
1780
1781 l1 = org_l1;
1782 l2 = org_l2;
1783 tail = Qnil;
1784 value = Qnil;
1785
1786 /* It is sufficient to protect org_l1 and org_l2.
1787 When l1 and l2 are updated, we copy the new values
1788 back into the org_ vars. */
1789 GCPRO4 (org_l1, org_l2, pred, value);
1790
1791 while (1)
1792 {
1793 if (NILP (l1))
1794 {
1795 UNGCPRO;
1796 if (NILP (tail))
1797 return l2;
1798 Fsetcdr (tail, l2);
1799 return value;
1800 }
1801 if (NILP (l2))
1802 {
1803 UNGCPRO;
1804 if (NILP (tail))
1805 return l1;
1806 Fsetcdr (tail, l1);
1807 return value;
1808 }
1809 tem = call2 (pred, Fcar (l2), Fcar (l1));
1810 if (NILP (tem))
1811 {
1812 tem = l1;
1813 l1 = Fcdr (l1);
1814 org_l1 = l1;
1815 }
1816 else
1817 {
1818 tem = l2;
1819 l2 = Fcdr (l2);
1820 org_l2 = l2;
1821 }
1822 if (NILP (tail))
1823 value = tem;
1824 else
1825 Fsetcdr (tail, tem);
1826 tail = tem;
1827 }
1828 }
1829
1830 \f
1831 /* This does not check for quits. That is safe since it must terminate. */
1832
1833 DEFUN ("plist-get", Fplist_get, Splist_get, 2, 2, 0,
1834 doc: /* Extract a value from a property list.
1835 PLIST is a property list, which is a list of the form
1836 \(PROP1 VALUE1 PROP2 VALUE2...). This function returns the value
1837 corresponding to the given PROP, or nil if PROP is not one of the
1838 properties on the list. This function never signals an error. */)
1839 (Lisp_Object plist, Lisp_Object prop)
1840 {
1841 Lisp_Object tail, halftail;
1842
1843 /* halftail is used to detect circular lists. */
1844 tail = halftail = plist;
1845 while (CONSP (tail) && CONSP (XCDR (tail)))
1846 {
1847 if (EQ (prop, XCAR (tail)))
1848 return XCAR (XCDR (tail));
1849
1850 tail = XCDR (XCDR (tail));
1851 halftail = XCDR (halftail);
1852 if (EQ (tail, halftail))
1853 break;
1854
1855 #if 0 /* Unsafe version. */
1856 /* This function can be called asynchronously
1857 (setup_coding_system). Don't QUIT in that case. */
1858 if (!interrupt_input_blocked)
1859 QUIT;
1860 #endif
1861 }
1862
1863 return Qnil;
1864 }
1865
1866 DEFUN ("get", Fget, Sget, 2, 2, 0,
1867 doc: /* Return the value of SYMBOL's PROPNAME property.
1868 This is the last value stored with `(put SYMBOL PROPNAME VALUE)'. */)
1869 (Lisp_Object symbol, Lisp_Object propname)
1870 {
1871 CHECK_SYMBOL (symbol);
1872 return Fplist_get (XSYMBOL (symbol)->plist, propname);
1873 }
1874
1875 DEFUN ("plist-put", Fplist_put, Splist_put, 3, 3, 0,
1876 doc: /* Change value in PLIST of PROP to VAL.
1877 PLIST is a property list, which is a list of the form
1878 \(PROP1 VALUE1 PROP2 VALUE2 ...). PROP is a symbol and VAL is any object.
1879 If PROP is already a property on the list, its value is set to VAL,
1880 otherwise the new PROP VAL pair is added. The new plist is returned;
1881 use `(setq x (plist-put x prop val))' to be sure to use the new value.
1882 The PLIST is modified by side effects. */)
1883 (Lisp_Object plist, register Lisp_Object prop, Lisp_Object val)
1884 {
1885 register Lisp_Object tail, prev;
1886 Lisp_Object newcell;
1887 prev = Qnil;
1888 for (tail = plist; CONSP (tail) && CONSP (XCDR (tail));
1889 tail = XCDR (XCDR (tail)))
1890 {
1891 if (EQ (prop, XCAR (tail)))
1892 {
1893 Fsetcar (XCDR (tail), val);
1894 return plist;
1895 }
1896
1897 prev = tail;
1898 QUIT;
1899 }
1900 newcell = Fcons (prop, Fcons (val, NILP (prev) ? plist : XCDR (XCDR (prev))));
1901 if (NILP (prev))
1902 return newcell;
1903 else
1904 Fsetcdr (XCDR (prev), newcell);
1905 return plist;
1906 }
1907
1908 DEFUN ("put", Fput, Sput, 3, 3, 0,
1909 doc: /* Store SYMBOL's PROPNAME property with value VALUE.
1910 It can be retrieved with `(get SYMBOL PROPNAME)'. */)
1911 (Lisp_Object symbol, Lisp_Object propname, Lisp_Object value)
1912 {
1913 CHECK_SYMBOL (symbol);
1914 XSYMBOL (symbol)->plist
1915 = Fplist_put (XSYMBOL (symbol)->plist, propname, value);
1916 return value;
1917 }
1918 \f
1919 DEFUN ("lax-plist-get", Flax_plist_get, Slax_plist_get, 2, 2, 0,
1920 doc: /* Extract a value from a property list, comparing with `equal'.
1921 PLIST is a property list, which is a list of the form
1922 \(PROP1 VALUE1 PROP2 VALUE2...). This function returns the value
1923 corresponding to the given PROP, or nil if PROP is not
1924 one of the properties on the list. */)
1925 (Lisp_Object plist, Lisp_Object prop)
1926 {
1927 Lisp_Object tail;
1928
1929 for (tail = plist;
1930 CONSP (tail) && CONSP (XCDR (tail));
1931 tail = XCDR (XCDR (tail)))
1932 {
1933 if (! NILP (Fequal (prop, XCAR (tail))))
1934 return XCAR (XCDR (tail));
1935
1936 QUIT;
1937 }
1938
1939 CHECK_LIST_END (tail, prop);
1940
1941 return Qnil;
1942 }
1943
1944 DEFUN ("lax-plist-put", Flax_plist_put, Slax_plist_put, 3, 3, 0,
1945 doc: /* Change value in PLIST of PROP to VAL, comparing with `equal'.
1946 PLIST is a property list, which is a list of the form
1947 \(PROP1 VALUE1 PROP2 VALUE2 ...). PROP and VAL are any objects.
1948 If PROP is already a property on the list, its value is set to VAL,
1949 otherwise the new PROP VAL pair is added. The new plist is returned;
1950 use `(setq x (lax-plist-put x prop val))' to be sure to use the new value.
1951 The PLIST is modified by side effects. */)
1952 (Lisp_Object plist, register Lisp_Object prop, Lisp_Object val)
1953 {
1954 register Lisp_Object tail, prev;
1955 Lisp_Object newcell;
1956 prev = Qnil;
1957 for (tail = plist; CONSP (tail) && CONSP (XCDR (tail));
1958 tail = XCDR (XCDR (tail)))
1959 {
1960 if (! NILP (Fequal (prop, XCAR (tail))))
1961 {
1962 Fsetcar (XCDR (tail), val);
1963 return plist;
1964 }
1965
1966 prev = tail;
1967 QUIT;
1968 }
1969 newcell = Fcons (prop, Fcons (val, Qnil));
1970 if (NILP (prev))
1971 return newcell;
1972 else
1973 Fsetcdr (XCDR (prev), newcell);
1974 return plist;
1975 }
1976 \f
1977 DEFUN ("eql", Feql, Seql, 2, 2, 0,
1978 doc: /* Return t if the two args are the same Lisp object.
1979 Floating-point numbers of equal value are `eql', but they may not be `eq'. */)
1980 (Lisp_Object obj1, Lisp_Object obj2)
1981 {
1982 if (FLOATP (obj1))
1983 return internal_equal (obj1, obj2, 0, 0) ? Qt : Qnil;
1984 else
1985 return EQ (obj1, obj2) ? Qt : Qnil;
1986 }
1987
1988 DEFUN ("equal", Fequal, Sequal, 2, 2, 0,
1989 doc: /* Return t if two Lisp objects have similar structure and contents.
1990 They must have the same data type.
1991 Conses are compared by comparing the cars and the cdrs.
1992 Vectors and strings are compared element by element.
1993 Numbers are compared by value, but integers cannot equal floats.
1994 (Use `=' if you want integers and floats to be able to be equal.)
1995 Symbols must match exactly. */)
1996 (register Lisp_Object o1, Lisp_Object o2)
1997 {
1998 return internal_equal (o1, o2, 0, 0) ? Qt : Qnil;
1999 }
2000
2001 DEFUN ("equal-including-properties", Fequal_including_properties, Sequal_including_properties, 2, 2, 0,
2002 doc: /* Return t if two Lisp objects have similar structure and contents.
2003 This is like `equal' except that it compares the text properties
2004 of strings. (`equal' ignores text properties.) */)
2005 (register Lisp_Object o1, Lisp_Object o2)
2006 {
2007 return internal_equal (o1, o2, 0, 1) ? Qt : Qnil;
2008 }
2009
2010 /* DEPTH is current depth of recursion. Signal an error if it
2011 gets too deep.
2012 PROPS, if non-nil, means compare string text properties too. */
2013
2014 static int
2015 internal_equal (register Lisp_Object o1, register Lisp_Object o2, int depth, int props)
2016 {
2017 if (depth > 200)
2018 error ("Stack overflow in equal");
2019
2020 tail_recurse:
2021 QUIT;
2022 if (EQ (o1, o2))
2023 return 1;
2024 if (XTYPE (o1) != XTYPE (o2))
2025 return 0;
2026
2027 switch (XTYPE (o1))
2028 {
2029 case Lisp_Float:
2030 {
2031 double d1, d2;
2032
2033 d1 = extract_float (o1);
2034 d2 = extract_float (o2);
2035 /* If d is a NaN, then d != d. Two NaNs should be `equal' even
2036 though they are not =. */
2037 return d1 == d2 || (d1 != d1 && d2 != d2);
2038 }
2039
2040 case Lisp_Cons:
2041 if (!internal_equal (XCAR (o1), XCAR (o2), depth + 1, props))
2042 return 0;
2043 o1 = XCDR (o1);
2044 o2 = XCDR (o2);
2045 goto tail_recurse;
2046
2047 case Lisp_Misc:
2048 if (XMISCTYPE (o1) != XMISCTYPE (o2))
2049 return 0;
2050 if (OVERLAYP (o1))
2051 {
2052 if (!internal_equal (OVERLAY_START (o1), OVERLAY_START (o2),
2053 depth + 1, props)
2054 || !internal_equal (OVERLAY_END (o1), OVERLAY_END (o2),
2055 depth + 1, props))
2056 return 0;
2057 o1 = XOVERLAY (o1)->plist;
2058 o2 = XOVERLAY (o2)->plist;
2059 goto tail_recurse;
2060 }
2061 if (MARKERP (o1))
2062 {
2063 return (XMARKER (o1)->buffer == XMARKER (o2)->buffer
2064 && (XMARKER (o1)->buffer == 0
2065 || XMARKER (o1)->bytepos == XMARKER (o2)->bytepos));
2066 }
2067 break;
2068
2069 case Lisp_Vectorlike:
2070 {
2071 register int i;
2072 ptrdiff_t size = ASIZE (o1);
2073 /* Pseudovectors have the type encoded in the size field, so this test
2074 actually checks that the objects have the same type as well as the
2075 same size. */
2076 if (ASIZE (o2) != size)
2077 return 0;
2078 /* Boolvectors are compared much like strings. */
2079 if (BOOL_VECTOR_P (o1))
2080 {
2081 if (XBOOL_VECTOR (o1)->size != XBOOL_VECTOR (o2)->size)
2082 return 0;
2083 if (memcmp (XBOOL_VECTOR (o1)->data, XBOOL_VECTOR (o2)->data,
2084 ((XBOOL_VECTOR (o1)->size
2085 + BOOL_VECTOR_BITS_PER_CHAR - 1)
2086 / BOOL_VECTOR_BITS_PER_CHAR)))
2087 return 0;
2088 return 1;
2089 }
2090 if (WINDOW_CONFIGURATIONP (o1))
2091 return compare_window_configurations (o1, o2, 0);
2092
2093 /* Aside from them, only true vectors, char-tables, compiled
2094 functions, and fonts (font-spec, font-entity, font-object)
2095 are sensible to compare, so eliminate the others now. */
2096 if (size & PSEUDOVECTOR_FLAG)
2097 {
2098 if (!(size & ((PVEC_COMPILED | PVEC_CHAR_TABLE
2099 | PVEC_SUB_CHAR_TABLE | PVEC_FONT)
2100 << PSEUDOVECTOR_SIZE_BITS)))
2101 return 0;
2102 size &= PSEUDOVECTOR_SIZE_MASK;
2103 }
2104 for (i = 0; i < size; i++)
2105 {
2106 Lisp_Object v1, v2;
2107 v1 = AREF (o1, i);
2108 v2 = AREF (o2, i);
2109 if (!internal_equal (v1, v2, depth + 1, props))
2110 return 0;
2111 }
2112 return 1;
2113 }
2114 break;
2115
2116 case Lisp_String:
2117 if (SCHARS (o1) != SCHARS (o2))
2118 return 0;
2119 if (SBYTES (o1) != SBYTES (o2))
2120 return 0;
2121 if (memcmp (SDATA (o1), SDATA (o2), SBYTES (o1)))
2122 return 0;
2123 if (props && !compare_string_intervals (o1, o2))
2124 return 0;
2125 return 1;
2126
2127 default:
2128 break;
2129 }
2130
2131 return 0;
2132 }
2133 \f
2134
2135 DEFUN ("fillarray", Ffillarray, Sfillarray, 2, 2, 0,
2136 doc: /* Store each element of ARRAY with ITEM.
2137 ARRAY is a vector, string, char-table, or bool-vector. */)
2138 (Lisp_Object array, Lisp_Object item)
2139 {
2140 register ptrdiff_t size, idx;
2141
2142 if (VECTORP (array))
2143 {
2144 register Lisp_Object *p = XVECTOR (array)->contents;
2145 size = ASIZE (array);
2146 for (idx = 0; idx < size; idx++)
2147 p[idx] = item;
2148 }
2149 else if (CHAR_TABLE_P (array))
2150 {
2151 int i;
2152
2153 for (i = 0; i < (1 << CHARTAB_SIZE_BITS_0); i++)
2154 XCHAR_TABLE (array)->contents[i] = item;
2155 XCHAR_TABLE (array)->defalt = item;
2156 }
2157 else if (STRINGP (array))
2158 {
2159 register unsigned char *p = SDATA (array);
2160 int charval;
2161 CHECK_CHARACTER (item);
2162 charval = XFASTINT (item);
2163 size = SCHARS (array);
2164 if (STRING_MULTIBYTE (array))
2165 {
2166 unsigned char str[MAX_MULTIBYTE_LENGTH];
2167 int len = CHAR_STRING (charval, str);
2168 ptrdiff_t size_byte = SBYTES (array);
2169
2170 if (INT_MULTIPLY_OVERFLOW (SCHARS (array), len)
2171 || SCHARS (array) * len != size_byte)
2172 error ("Attempt to change byte length of a string");
2173 for (idx = 0; idx < size_byte; idx++)
2174 *p++ = str[idx % len];
2175 }
2176 else
2177 for (idx = 0; idx < size; idx++)
2178 p[idx] = charval;
2179 }
2180 else if (BOOL_VECTOR_P (array))
2181 {
2182 register unsigned char *p = XBOOL_VECTOR (array)->data;
2183 size =
2184 ((XBOOL_VECTOR (array)->size + BOOL_VECTOR_BITS_PER_CHAR - 1)
2185 / BOOL_VECTOR_BITS_PER_CHAR);
2186
2187 if (size)
2188 {
2189 memset (p, ! NILP (item) ? -1 : 0, size);
2190
2191 /* Clear any extraneous bits in the last byte. */
2192 p[size - 1] &= (1 << (size % BOOL_VECTOR_BITS_PER_CHAR)) - 1;
2193 }
2194 }
2195 else
2196 wrong_type_argument (Qarrayp, array);
2197 return array;
2198 }
2199
2200 DEFUN ("clear-string", Fclear_string, Sclear_string,
2201 1, 1, 0,
2202 doc: /* Clear the contents of STRING.
2203 This makes STRING unibyte and may change its length. */)
2204 (Lisp_Object string)
2205 {
2206 ptrdiff_t len;
2207 CHECK_STRING (string);
2208 len = SBYTES (string);
2209 memset (SDATA (string), 0, len);
2210 STRING_SET_CHARS (string, len);
2211 STRING_SET_UNIBYTE (string);
2212 return Qnil;
2213 }
2214 \f
2215 /* ARGSUSED */
2216 Lisp_Object
2217 nconc2 (Lisp_Object s1, Lisp_Object s2)
2218 {
2219 Lisp_Object args[2];
2220 args[0] = s1;
2221 args[1] = s2;
2222 return Fnconc (2, args);
2223 }
2224
2225 DEFUN ("nconc", Fnconc, Snconc, 0, MANY, 0,
2226 doc: /* Concatenate any number of lists by altering them.
2227 Only the last argument is not altered, and need not be a list.
2228 usage: (nconc &rest LISTS) */)
2229 (ptrdiff_t nargs, Lisp_Object *args)
2230 {
2231 ptrdiff_t argnum;
2232 register Lisp_Object tail, tem, val;
2233
2234 val = tail = Qnil;
2235
2236 for (argnum = 0; argnum < nargs; argnum++)
2237 {
2238 tem = args[argnum];
2239 if (NILP (tem)) continue;
2240
2241 if (NILP (val))
2242 val = tem;
2243
2244 if (argnum + 1 == nargs) break;
2245
2246 CHECK_LIST_CONS (tem, tem);
2247
2248 while (CONSP (tem))
2249 {
2250 tail = tem;
2251 tem = XCDR (tail);
2252 QUIT;
2253 }
2254
2255 tem = args[argnum + 1];
2256 Fsetcdr (tail, tem);
2257 if (NILP (tem))
2258 args[argnum + 1] = tail;
2259 }
2260
2261 return val;
2262 }
2263 \f
2264 /* This is the guts of all mapping functions.
2265 Apply FN to each element of SEQ, one by one,
2266 storing the results into elements of VALS, a C vector of Lisp_Objects.
2267 LENI is the length of VALS, which should also be the length of SEQ. */
2268
2269 static void
2270 mapcar1 (EMACS_INT leni, Lisp_Object *vals, Lisp_Object fn, Lisp_Object seq)
2271 {
2272 register Lisp_Object tail;
2273 Lisp_Object dummy;
2274 register EMACS_INT i;
2275 struct gcpro gcpro1, gcpro2, gcpro3;
2276
2277 if (vals)
2278 {
2279 /* Don't let vals contain any garbage when GC happens. */
2280 for (i = 0; i < leni; i++)
2281 vals[i] = Qnil;
2282
2283 GCPRO3 (dummy, fn, seq);
2284 gcpro1.var = vals;
2285 gcpro1.nvars = leni;
2286 }
2287 else
2288 GCPRO2 (fn, seq);
2289 /* We need not explicitly protect `tail' because it is used only on lists, and
2290 1) lists are not relocated and 2) the list is marked via `seq' so will not
2291 be freed */
2292
2293 if (VECTORP (seq) || COMPILEDP (seq))
2294 {
2295 for (i = 0; i < leni; i++)
2296 {
2297 dummy = call1 (fn, AREF (seq, i));
2298 if (vals)
2299 vals[i] = dummy;
2300 }
2301 }
2302 else if (BOOL_VECTOR_P (seq))
2303 {
2304 for (i = 0; i < leni; i++)
2305 {
2306 unsigned char byte;
2307 byte = XBOOL_VECTOR (seq)->data[i / BOOL_VECTOR_BITS_PER_CHAR];
2308 dummy = (byte & (1 << (i % BOOL_VECTOR_BITS_PER_CHAR))) ? Qt : Qnil;
2309 dummy = call1 (fn, dummy);
2310 if (vals)
2311 vals[i] = dummy;
2312 }
2313 }
2314 else if (STRINGP (seq))
2315 {
2316 ptrdiff_t i_byte;
2317
2318 for (i = 0, i_byte = 0; i < leni;)
2319 {
2320 int c;
2321 ptrdiff_t i_before = i;
2322
2323 FETCH_STRING_CHAR_ADVANCE (c, seq, i, i_byte);
2324 XSETFASTINT (dummy, c);
2325 dummy = call1 (fn, dummy);
2326 if (vals)
2327 vals[i_before] = dummy;
2328 }
2329 }
2330 else /* Must be a list, since Flength did not get an error */
2331 {
2332 tail = seq;
2333 for (i = 0; i < leni && CONSP (tail); i++)
2334 {
2335 dummy = call1 (fn, XCAR (tail));
2336 if (vals)
2337 vals[i] = dummy;
2338 tail = XCDR (tail);
2339 }
2340 }
2341
2342 UNGCPRO;
2343 }
2344
2345 DEFUN ("mapconcat", Fmapconcat, Smapconcat, 3, 3, 0,
2346 doc: /* Apply FUNCTION to each element of SEQUENCE, and concat the results as strings.
2347 In between each pair of results, stick in SEPARATOR. Thus, " " as
2348 SEPARATOR results in spaces between the values returned by FUNCTION.
2349 SEQUENCE may be a list, a vector, a bool-vector, or a string. */)
2350 (Lisp_Object function, Lisp_Object sequence, Lisp_Object separator)
2351 {
2352 Lisp_Object len;
2353 register EMACS_INT leni;
2354 EMACS_INT nargs;
2355 ptrdiff_t i;
2356 register Lisp_Object *args;
2357 struct gcpro gcpro1;
2358 Lisp_Object ret;
2359 USE_SAFE_ALLOCA;
2360
2361 len = Flength (sequence);
2362 if (CHAR_TABLE_P (sequence))
2363 wrong_type_argument (Qlistp, sequence);
2364 leni = XINT (len);
2365 nargs = leni + leni - 1;
2366 if (nargs < 0) return empty_unibyte_string;
2367
2368 SAFE_ALLOCA_LISP (args, nargs);
2369
2370 GCPRO1 (separator);
2371 mapcar1 (leni, args, function, sequence);
2372 UNGCPRO;
2373
2374 for (i = leni - 1; i > 0; i--)
2375 args[i + i] = args[i];
2376
2377 for (i = 1; i < nargs; i += 2)
2378 args[i] = separator;
2379
2380 ret = Fconcat (nargs, args);
2381 SAFE_FREE ();
2382
2383 return ret;
2384 }
2385
2386 DEFUN ("mapcar", Fmapcar, Smapcar, 2, 2, 0,
2387 doc: /* Apply FUNCTION to each element of SEQUENCE, and make a list of the results.
2388 The result is a list just as long as SEQUENCE.
2389 SEQUENCE may be a list, a vector, a bool-vector, or a string. */)
2390 (Lisp_Object function, Lisp_Object sequence)
2391 {
2392 register Lisp_Object len;
2393 register EMACS_INT leni;
2394 register Lisp_Object *args;
2395 Lisp_Object ret;
2396 USE_SAFE_ALLOCA;
2397
2398 len = Flength (sequence);
2399 if (CHAR_TABLE_P (sequence))
2400 wrong_type_argument (Qlistp, sequence);
2401 leni = XFASTINT (len);
2402
2403 SAFE_ALLOCA_LISP (args, leni);
2404
2405 mapcar1 (leni, args, function, sequence);
2406
2407 ret = Flist (leni, args);
2408 SAFE_FREE ();
2409
2410 return ret;
2411 }
2412
2413 DEFUN ("mapc", Fmapc, Smapc, 2, 2, 0,
2414 doc: /* Apply FUNCTION to each element of SEQUENCE for side effects only.
2415 Unlike `mapcar', don't accumulate the results. Return SEQUENCE.
2416 SEQUENCE may be a list, a vector, a bool-vector, or a string. */)
2417 (Lisp_Object function, Lisp_Object sequence)
2418 {
2419 register EMACS_INT leni;
2420
2421 leni = XFASTINT (Flength (sequence));
2422 if (CHAR_TABLE_P (sequence))
2423 wrong_type_argument (Qlistp, sequence);
2424 mapcar1 (leni, 0, function, sequence);
2425
2426 return sequence;
2427 }
2428 \f
2429 /* This is how C code calls `yes-or-no-p' and allows the user
2430 to redefined it.
2431
2432 Anything that calls this function must protect from GC! */
2433
2434 Lisp_Object
2435 do_yes_or_no_p (Lisp_Object prompt)
2436 {
2437 return call1 (intern ("yes-or-no-p"), prompt);
2438 }
2439
2440 /* Anything that calls this function must protect from GC! */
2441
2442 DEFUN ("yes-or-no-p", Fyes_or_no_p, Syes_or_no_p, 1, 1, 0,
2443 doc: /* Ask user a yes-or-no question. Return t if answer is yes.
2444 PROMPT is the string to display to ask the question. It should end in
2445 a space; `yes-or-no-p' adds \"(yes or no) \" to it.
2446
2447 The user must confirm the answer with RET, and can edit it until it
2448 has been confirmed.
2449
2450 Under a windowing system a dialog box will be used if `last-nonmenu-event'
2451 is nil, and `use-dialog-box' is non-nil. */)
2452 (Lisp_Object prompt)
2453 {
2454 register Lisp_Object ans;
2455 Lisp_Object args[2];
2456 struct gcpro gcpro1;
2457
2458 CHECK_STRING (prompt);
2459
2460 #ifdef HAVE_MENUS
2461 if (FRAME_WINDOW_P (SELECTED_FRAME ())
2462 && (NILP (last_nonmenu_event) || CONSP (last_nonmenu_event))
2463 && use_dialog_box
2464 && have_menus_p ())
2465 {
2466 Lisp_Object pane, menu, obj;
2467 redisplay_preserve_echo_area (4);
2468 pane = Fcons (Fcons (build_string ("Yes"), Qt),
2469 Fcons (Fcons (build_string ("No"), Qnil),
2470 Qnil));
2471 GCPRO1 (pane);
2472 menu = Fcons (prompt, pane);
2473 obj = Fx_popup_dialog (Qt, menu, Qnil);
2474 UNGCPRO;
2475 return obj;
2476 }
2477 #endif /* HAVE_MENUS */
2478
2479 args[0] = prompt;
2480 args[1] = build_string ("(yes or no) ");
2481 prompt = Fconcat (2, args);
2482
2483 GCPRO1 (prompt);
2484
2485 while (1)
2486 {
2487 ans = Fdowncase (Fread_from_minibuffer (prompt, Qnil, Qnil, Qnil,
2488 Qyes_or_no_p_history, Qnil,
2489 Qnil));
2490 if (SCHARS (ans) == 3 && !strcmp (SSDATA (ans), "yes"))
2491 {
2492 UNGCPRO;
2493 return Qt;
2494 }
2495 if (SCHARS (ans) == 2 && !strcmp (SSDATA (ans), "no"))
2496 {
2497 UNGCPRO;
2498 return Qnil;
2499 }
2500
2501 Fding (Qnil);
2502 Fdiscard_input ();
2503 message ("Please answer yes or no.");
2504 Fsleep_for (make_number (2), Qnil);
2505 }
2506 }
2507 \f
2508 DEFUN ("load-average", Fload_average, Sload_average, 0, 1, 0,
2509 doc: /* Return list of 1 minute, 5 minute and 15 minute load averages.
2510
2511 Each of the three load averages is multiplied by 100, then converted
2512 to integer.
2513
2514 When USE-FLOATS is non-nil, floats will be used instead of integers.
2515 These floats are not multiplied by 100.
2516
2517 If the 5-minute or 15-minute load averages are not available, return a
2518 shortened list, containing only those averages which are available.
2519
2520 An error is thrown if the load average can't be obtained. In some
2521 cases making it work would require Emacs being installed setuid or
2522 setgid so that it can read kernel information, and that usually isn't
2523 advisable. */)
2524 (Lisp_Object use_floats)
2525 {
2526 double load_ave[3];
2527 int loads = getloadavg (load_ave, 3);
2528 Lisp_Object ret = Qnil;
2529
2530 if (loads < 0)
2531 error ("load-average not implemented for this operating system");
2532
2533 while (loads-- > 0)
2534 {
2535 Lisp_Object load = (NILP (use_floats)
2536 ? make_number (100.0 * load_ave[loads])
2537 : make_float (load_ave[loads]));
2538 ret = Fcons (load, ret);
2539 }
2540
2541 return ret;
2542 }
2543 \f
2544 static Lisp_Object Qsubfeatures;
2545
2546 DEFUN ("featurep", Ffeaturep, Sfeaturep, 1, 2, 0,
2547 doc: /* Return t if FEATURE is present in this Emacs.
2548
2549 Use this to conditionalize execution of lisp code based on the
2550 presence or absence of Emacs or environment extensions.
2551 Use `provide' to declare that a feature is available. This function
2552 looks at the value of the variable `features'. The optional argument
2553 SUBFEATURE can be used to check a specific subfeature of FEATURE. */)
2554 (Lisp_Object feature, Lisp_Object subfeature)
2555 {
2556 register Lisp_Object tem;
2557 CHECK_SYMBOL (feature);
2558 tem = Fmemq (feature, Vfeatures);
2559 if (!NILP (tem) && !NILP (subfeature))
2560 tem = Fmember (subfeature, Fget (feature, Qsubfeatures));
2561 return (NILP (tem)) ? Qnil : Qt;
2562 }
2563
2564 DEFUN ("provide", Fprovide, Sprovide, 1, 2, 0,
2565 doc: /* Announce that FEATURE is a feature of the current Emacs.
2566 The optional argument SUBFEATURES should be a list of symbols listing
2567 particular subfeatures supported in this version of FEATURE. */)
2568 (Lisp_Object feature, Lisp_Object subfeatures)
2569 {
2570 register Lisp_Object tem;
2571 CHECK_SYMBOL (feature);
2572 CHECK_LIST (subfeatures);
2573 if (!NILP (Vautoload_queue))
2574 Vautoload_queue = Fcons (Fcons (make_number (0), Vfeatures),
2575 Vautoload_queue);
2576 tem = Fmemq (feature, Vfeatures);
2577 if (NILP (tem))
2578 Vfeatures = Fcons (feature, Vfeatures);
2579 if (!NILP (subfeatures))
2580 Fput (feature, Qsubfeatures, subfeatures);
2581 LOADHIST_ATTACH (Fcons (Qprovide, feature));
2582
2583 /* Run any load-hooks for this file. */
2584 tem = Fassq (feature, Vafter_load_alist);
2585 if (CONSP (tem))
2586 Fprogn (XCDR (tem));
2587
2588 return feature;
2589 }
2590 \f
2591 /* `require' and its subroutines. */
2592
2593 /* List of features currently being require'd, innermost first. */
2594
2595 static Lisp_Object require_nesting_list;
2596
2597 static Lisp_Object
2598 require_unwind (Lisp_Object old_value)
2599 {
2600 return require_nesting_list = old_value;
2601 }
2602
2603 DEFUN ("require", Frequire, Srequire, 1, 3, 0,
2604 doc: /* If feature FEATURE is not loaded, load it from FILENAME.
2605 If FEATURE is not a member of the list `features', then the feature
2606 is not loaded; so load the file FILENAME.
2607 If FILENAME is omitted, the printname of FEATURE is used as the file name,
2608 and `load' will try to load this name appended with the suffix `.elc' or
2609 `.el', in that order. The name without appended suffix will not be used.
2610 See `get-load-suffixes' for the complete list of suffixes.
2611 If the optional third argument NOERROR is non-nil,
2612 then return nil if the file is not found instead of signaling an error.
2613 Normally the return value is FEATURE.
2614 The normal messages at start and end of loading FILENAME are suppressed. */)
2615 (Lisp_Object feature, Lisp_Object filename, Lisp_Object noerror)
2616 {
2617 register Lisp_Object tem;
2618 struct gcpro gcpro1, gcpro2;
2619 int from_file = load_in_progress;
2620
2621 CHECK_SYMBOL (feature);
2622
2623 /* Record the presence of `require' in this file
2624 even if the feature specified is already loaded.
2625 But not more than once in any file,
2626 and not when we aren't loading or reading from a file. */
2627 if (!from_file)
2628 for (tem = Vcurrent_load_list; CONSP (tem); tem = XCDR (tem))
2629 if (NILP (XCDR (tem)) && STRINGP (XCAR (tem)))
2630 from_file = 1;
2631
2632 if (from_file)
2633 {
2634 tem = Fcons (Qrequire, feature);
2635 if (NILP (Fmember (tem, Vcurrent_load_list)))
2636 LOADHIST_ATTACH (tem);
2637 }
2638 tem = Fmemq (feature, Vfeatures);
2639
2640 if (NILP (tem))
2641 {
2642 ptrdiff_t count = SPECPDL_INDEX ();
2643 int nesting = 0;
2644
2645 /* This is to make sure that loadup.el gives a clear picture
2646 of what files are preloaded and when. */
2647 if (! NILP (Vpurify_flag))
2648 error ("(require %s) while preparing to dump",
2649 SDATA (SYMBOL_NAME (feature)));
2650
2651 /* A certain amount of recursive `require' is legitimate,
2652 but if we require the same feature recursively 3 times,
2653 signal an error. */
2654 tem = require_nesting_list;
2655 while (! NILP (tem))
2656 {
2657 if (! NILP (Fequal (feature, XCAR (tem))))
2658 nesting++;
2659 tem = XCDR (tem);
2660 }
2661 if (nesting > 3)
2662 error ("Recursive `require' for feature `%s'",
2663 SDATA (SYMBOL_NAME (feature)));
2664
2665 /* Update the list for any nested `require's that occur. */
2666 record_unwind_protect (require_unwind, require_nesting_list);
2667 require_nesting_list = Fcons (feature, require_nesting_list);
2668
2669 /* Value saved here is to be restored into Vautoload_queue */
2670 record_unwind_protect (un_autoload, Vautoload_queue);
2671 Vautoload_queue = Qt;
2672
2673 /* Load the file. */
2674 GCPRO2 (feature, filename);
2675 tem = Fload (NILP (filename) ? Fsymbol_name (feature) : filename,
2676 noerror, Qt, Qnil, (NILP (filename) ? Qt : Qnil));
2677 UNGCPRO;
2678
2679 /* If load failed entirely, return nil. */
2680 if (NILP (tem))
2681 return unbind_to (count, Qnil);
2682
2683 tem = Fmemq (feature, Vfeatures);
2684 if (NILP (tem))
2685 error ("Required feature `%s' was not provided",
2686 SDATA (SYMBOL_NAME (feature)));
2687
2688 /* Once loading finishes, don't undo it. */
2689 Vautoload_queue = Qt;
2690 feature = unbind_to (count, feature);
2691 }
2692
2693 return feature;
2694 }
2695 \f
2696 /* Primitives for work of the "widget" library.
2697 In an ideal world, this section would not have been necessary.
2698 However, lisp function calls being as slow as they are, it turns
2699 out that some functions in the widget library (wid-edit.el) are the
2700 bottleneck of Widget operation. Here is their translation to C,
2701 for the sole reason of efficiency. */
2702
2703 DEFUN ("plist-member", Fplist_member, Splist_member, 2, 2, 0,
2704 doc: /* Return non-nil if PLIST has the property PROP.
2705 PLIST is a property list, which is a list of the form
2706 \(PROP1 VALUE1 PROP2 VALUE2 ...\). PROP is a symbol.
2707 Unlike `plist-get', this allows you to distinguish between a missing
2708 property and a property with the value nil.
2709 The value is actually the tail of PLIST whose car is PROP. */)
2710 (Lisp_Object plist, Lisp_Object prop)
2711 {
2712 while (CONSP (plist) && !EQ (XCAR (plist), prop))
2713 {
2714 QUIT;
2715 plist = XCDR (plist);
2716 plist = CDR (plist);
2717 }
2718 return plist;
2719 }
2720
2721 DEFUN ("widget-put", Fwidget_put, Swidget_put, 3, 3, 0,
2722 doc: /* In WIDGET, set PROPERTY to VALUE.
2723 The value can later be retrieved with `widget-get'. */)
2724 (Lisp_Object widget, Lisp_Object property, Lisp_Object value)
2725 {
2726 CHECK_CONS (widget);
2727 XSETCDR (widget, Fplist_put (XCDR (widget), property, value));
2728 return value;
2729 }
2730
2731 DEFUN ("widget-get", Fwidget_get, Swidget_get, 2, 2, 0,
2732 doc: /* In WIDGET, get the value of PROPERTY.
2733 The value could either be specified when the widget was created, or
2734 later with `widget-put'. */)
2735 (Lisp_Object widget, Lisp_Object property)
2736 {
2737 Lisp_Object tmp;
2738
2739 while (1)
2740 {
2741 if (NILP (widget))
2742 return Qnil;
2743 CHECK_CONS (widget);
2744 tmp = Fplist_member (XCDR (widget), property);
2745 if (CONSP (tmp))
2746 {
2747 tmp = XCDR (tmp);
2748 return CAR (tmp);
2749 }
2750 tmp = XCAR (widget);
2751 if (NILP (tmp))
2752 return Qnil;
2753 widget = Fget (tmp, Qwidget_type);
2754 }
2755 }
2756
2757 DEFUN ("widget-apply", Fwidget_apply, Swidget_apply, 2, MANY, 0,
2758 doc: /* Apply the value of WIDGET's PROPERTY to the widget itself.
2759 ARGS are passed as extra arguments to the function.
2760 usage: (widget-apply WIDGET PROPERTY &rest ARGS) */)
2761 (ptrdiff_t nargs, Lisp_Object *args)
2762 {
2763 /* This function can GC. */
2764 Lisp_Object newargs[3];
2765 struct gcpro gcpro1, gcpro2;
2766 Lisp_Object result;
2767
2768 newargs[0] = Fwidget_get (args[0], args[1]);
2769 newargs[1] = args[0];
2770 newargs[2] = Flist (nargs - 2, args + 2);
2771 GCPRO2 (newargs[0], newargs[2]);
2772 result = Fapply (3, newargs);
2773 UNGCPRO;
2774 return result;
2775 }
2776
2777 #ifdef HAVE_LANGINFO_CODESET
2778 #include <langinfo.h>
2779 #endif
2780
2781 DEFUN ("locale-info", Flocale_info, Slocale_info, 1, 1, 0,
2782 doc: /* Access locale data ITEM for the current C locale, if available.
2783 ITEM should be one of the following:
2784
2785 `codeset', returning the character set as a string (locale item CODESET);
2786
2787 `days', returning a 7-element vector of day names (locale items DAY_n);
2788
2789 `months', returning a 12-element vector of month names (locale items MON_n);
2790
2791 `paper', returning a list (WIDTH HEIGHT) for the default paper size,
2792 both measured in millimeters (locale items PAPER_WIDTH, PAPER_HEIGHT).
2793
2794 If the system can't provide such information through a call to
2795 `nl_langinfo', or if ITEM isn't from the list above, return nil.
2796
2797 See also Info node `(libc)Locales'.
2798
2799 The data read from the system are decoded using `locale-coding-system'. */)
2800 (Lisp_Object item)
2801 {
2802 char *str = NULL;
2803 #ifdef HAVE_LANGINFO_CODESET
2804 Lisp_Object val;
2805 if (EQ (item, Qcodeset))
2806 {
2807 str = nl_langinfo (CODESET);
2808 return build_string (str);
2809 }
2810 #ifdef DAY_1
2811 else if (EQ (item, Qdays)) /* e.g. for calendar-day-name-array */
2812 {
2813 Lisp_Object v = Fmake_vector (make_number (7), Qnil);
2814 const int days[7] = {DAY_1, DAY_2, DAY_3, DAY_4, DAY_5, DAY_6, DAY_7};
2815 int i;
2816 struct gcpro gcpro1;
2817 GCPRO1 (v);
2818 synchronize_system_time_locale ();
2819 for (i = 0; i < 7; i++)
2820 {
2821 str = nl_langinfo (days[i]);
2822 val = make_unibyte_string (str, strlen (str));
2823 /* Fixme: Is this coding system necessarily right, even if
2824 it is consistent with CODESET? If not, what to do? */
2825 Faset (v, make_number (i),
2826 code_convert_string_norecord (val, Vlocale_coding_system,
2827 0));
2828 }
2829 UNGCPRO;
2830 return v;
2831 }
2832 #endif /* DAY_1 */
2833 #ifdef MON_1
2834 else if (EQ (item, Qmonths)) /* e.g. for calendar-month-name-array */
2835 {
2836 Lisp_Object v = Fmake_vector (make_number (12), Qnil);
2837 const int months[12] = {MON_1, MON_2, MON_3, MON_4, MON_5, MON_6, MON_7,
2838 MON_8, MON_9, MON_10, MON_11, MON_12};
2839 int i;
2840 struct gcpro gcpro1;
2841 GCPRO1 (v);
2842 synchronize_system_time_locale ();
2843 for (i = 0; i < 12; i++)
2844 {
2845 str = nl_langinfo (months[i]);
2846 val = make_unibyte_string (str, strlen (str));
2847 Faset (v, make_number (i),
2848 code_convert_string_norecord (val, Vlocale_coding_system, 0));
2849 }
2850 UNGCPRO;
2851 return v;
2852 }
2853 #endif /* MON_1 */
2854 /* LC_PAPER stuff isn't defined as accessible in glibc as of 2.3.1,
2855 but is in the locale files. This could be used by ps-print. */
2856 #ifdef PAPER_WIDTH
2857 else if (EQ (item, Qpaper))
2858 {
2859 return list2 (make_number (nl_langinfo (PAPER_WIDTH)),
2860 make_number (nl_langinfo (PAPER_HEIGHT)));
2861 }
2862 #endif /* PAPER_WIDTH */
2863 #endif /* HAVE_LANGINFO_CODESET*/
2864 return Qnil;
2865 }
2866 \f
2867 /* base64 encode/decode functions (RFC 2045).
2868 Based on code from GNU recode. */
2869
2870 #define MIME_LINE_LENGTH 76
2871
2872 #define IS_ASCII(Character) \
2873 ((Character) < 128)
2874 #define IS_BASE64(Character) \
2875 (IS_ASCII (Character) && base64_char_to_value[Character] >= 0)
2876 #define IS_BASE64_IGNORABLE(Character) \
2877 ((Character) == ' ' || (Character) == '\t' || (Character) == '\n' \
2878 || (Character) == '\f' || (Character) == '\r')
2879
2880 /* Used by base64_decode_1 to retrieve a non-base64-ignorable
2881 character or return retval if there are no characters left to
2882 process. */
2883 #define READ_QUADRUPLET_BYTE(retval) \
2884 do \
2885 { \
2886 if (i == length) \
2887 { \
2888 if (nchars_return) \
2889 *nchars_return = nchars; \
2890 return (retval); \
2891 } \
2892 c = from[i++]; \
2893 } \
2894 while (IS_BASE64_IGNORABLE (c))
2895
2896 /* Table of characters coding the 64 values. */
2897 static const char base64_value_to_char[64] =
2898 {
2899 'A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'I', 'J', /* 0- 9 */
2900 'K', 'L', 'M', 'N', 'O', 'P', 'Q', 'R', 'S', 'T', /* 10-19 */
2901 'U', 'V', 'W', 'X', 'Y', 'Z', 'a', 'b', 'c', 'd', /* 20-29 */
2902 'e', 'f', 'g', 'h', 'i', 'j', 'k', 'l', 'm', 'n', /* 30-39 */
2903 'o', 'p', 'q', 'r', 's', 't', 'u', 'v', 'w', 'x', /* 40-49 */
2904 'y', 'z', '0', '1', '2', '3', '4', '5', '6', '7', /* 50-59 */
2905 '8', '9', '+', '/' /* 60-63 */
2906 };
2907
2908 /* Table of base64 values for first 128 characters. */
2909 static const short base64_char_to_value[128] =
2910 {
2911 -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, /* 0- 9 */
2912 -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, /* 10- 19 */
2913 -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, /* 20- 29 */
2914 -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, /* 30- 39 */
2915 -1, -1, -1, 62, -1, -1, -1, 63, 52, 53, /* 40- 49 */
2916 54, 55, 56, 57, 58, 59, 60, 61, -1, -1, /* 50- 59 */
2917 -1, -1, -1, -1, -1, 0, 1, 2, 3, 4, /* 60- 69 */
2918 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, /* 70- 79 */
2919 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, /* 80- 89 */
2920 25, -1, -1, -1, -1, -1, -1, 26, 27, 28, /* 90- 99 */
2921 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, /* 100-109 */
2922 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, /* 110-119 */
2923 49, 50, 51, -1, -1, -1, -1, -1 /* 120-127 */
2924 };
2925
2926 /* The following diagram shows the logical steps by which three octets
2927 get transformed into four base64 characters.
2928
2929 .--------. .--------. .--------.
2930 |aaaaaabb| |bbbbcccc| |ccdddddd|
2931 `--------' `--------' `--------'
2932 6 2 4 4 2 6
2933 .--------+--------+--------+--------.
2934 |00aaaaaa|00bbbbbb|00cccccc|00dddddd|
2935 `--------+--------+--------+--------'
2936
2937 .--------+--------+--------+--------.
2938 |AAAAAAAA|BBBBBBBB|CCCCCCCC|DDDDDDDD|
2939 `--------+--------+--------+--------'
2940
2941 The octets are divided into 6 bit chunks, which are then encoded into
2942 base64 characters. */
2943
2944
2945 static ptrdiff_t base64_encode_1 (const char *, char *, ptrdiff_t, int, int);
2946 static ptrdiff_t base64_decode_1 (const char *, char *, ptrdiff_t, int,
2947 ptrdiff_t *);
2948
2949 DEFUN ("base64-encode-region", Fbase64_encode_region, Sbase64_encode_region,
2950 2, 3, "r",
2951 doc: /* Base64-encode the region between BEG and END.
2952 Return the length of the encoded text.
2953 Optional third argument NO-LINE-BREAK means do not break long lines
2954 into shorter lines. */)
2955 (Lisp_Object beg, Lisp_Object end, Lisp_Object no_line_break)
2956 {
2957 char *encoded;
2958 ptrdiff_t allength, length;
2959 ptrdiff_t ibeg, iend, encoded_length;
2960 ptrdiff_t old_pos = PT;
2961 USE_SAFE_ALLOCA;
2962
2963 validate_region (&beg, &end);
2964
2965 ibeg = CHAR_TO_BYTE (XFASTINT (beg));
2966 iend = CHAR_TO_BYTE (XFASTINT (end));
2967 move_gap_both (XFASTINT (beg), ibeg);
2968
2969 /* We need to allocate enough room for encoding the text.
2970 We need 33 1/3% more space, plus a newline every 76
2971 characters, and then we round up. */
2972 length = iend - ibeg;
2973 allength = length + length/3 + 1;
2974 allength += allength / MIME_LINE_LENGTH + 1 + 6;
2975
2976 SAFE_ALLOCA (encoded, char *, allength);
2977 encoded_length = base64_encode_1 ((char *) BYTE_POS_ADDR (ibeg),
2978 encoded, length, NILP (no_line_break),
2979 !NILP (BVAR (current_buffer, enable_multibyte_characters)));
2980 if (encoded_length > allength)
2981 abort ();
2982
2983 if (encoded_length < 0)
2984 {
2985 /* The encoding wasn't possible. */
2986 SAFE_FREE ();
2987 error ("Multibyte character in data for base64 encoding");
2988 }
2989
2990 /* Now we have encoded the region, so we insert the new contents
2991 and delete the old. (Insert first in order to preserve markers.) */
2992 SET_PT_BOTH (XFASTINT (beg), ibeg);
2993 insert (encoded, encoded_length);
2994 SAFE_FREE ();
2995 del_range_byte (ibeg + encoded_length, iend + encoded_length, 1);
2996
2997 /* If point was outside of the region, restore it exactly; else just
2998 move to the beginning of the region. */
2999 if (old_pos >= XFASTINT (end))
3000 old_pos += encoded_length - (XFASTINT (end) - XFASTINT (beg));
3001 else if (old_pos > XFASTINT (beg))
3002 old_pos = XFASTINT (beg);
3003 SET_PT (old_pos);
3004
3005 /* We return the length of the encoded text. */
3006 return make_number (encoded_length);
3007 }
3008
3009 DEFUN ("base64-encode-string", Fbase64_encode_string, Sbase64_encode_string,
3010 1, 2, 0,
3011 doc: /* Base64-encode STRING and return the result.
3012 Optional second argument NO-LINE-BREAK means do not break long lines
3013 into shorter lines. */)
3014 (Lisp_Object string, Lisp_Object no_line_break)
3015 {
3016 ptrdiff_t allength, length, encoded_length;
3017 char *encoded;
3018 Lisp_Object encoded_string;
3019 USE_SAFE_ALLOCA;
3020
3021 CHECK_STRING (string);
3022
3023 /* We need to allocate enough room for encoding the text.
3024 We need 33 1/3% more space, plus a newline every 76
3025 characters, and then we round up. */
3026 length = SBYTES (string);
3027 allength = length + length/3 + 1;
3028 allength += allength / MIME_LINE_LENGTH + 1 + 6;
3029
3030 /* We need to allocate enough room for decoding the text. */
3031 SAFE_ALLOCA (encoded, char *, allength);
3032
3033 encoded_length = base64_encode_1 (SSDATA (string),
3034 encoded, length, NILP (no_line_break),
3035 STRING_MULTIBYTE (string));
3036 if (encoded_length > allength)
3037 abort ();
3038
3039 if (encoded_length < 0)
3040 {
3041 /* The encoding wasn't possible. */
3042 SAFE_FREE ();
3043 error ("Multibyte character in data for base64 encoding");
3044 }
3045
3046 encoded_string = make_unibyte_string (encoded, encoded_length);
3047 SAFE_FREE ();
3048
3049 return encoded_string;
3050 }
3051
3052 static ptrdiff_t
3053 base64_encode_1 (const char *from, char *to, ptrdiff_t length,
3054 int line_break, int multibyte)
3055 {
3056 int counter = 0;
3057 ptrdiff_t i = 0;
3058 char *e = to;
3059 int c;
3060 unsigned int value;
3061 int bytes;
3062
3063 while (i < length)
3064 {
3065 if (multibyte)
3066 {
3067 c = STRING_CHAR_AND_LENGTH ((unsigned char *) from + i, bytes);
3068 if (CHAR_BYTE8_P (c))
3069 c = CHAR_TO_BYTE8 (c);
3070 else if (c >= 256)
3071 return -1;
3072 i += bytes;
3073 }
3074 else
3075 c = from[i++];
3076
3077 /* Wrap line every 76 characters. */
3078
3079 if (line_break)
3080 {
3081 if (counter < MIME_LINE_LENGTH / 4)
3082 counter++;
3083 else
3084 {
3085 *e++ = '\n';
3086 counter = 1;
3087 }
3088 }
3089
3090 /* Process first byte of a triplet. */
3091
3092 *e++ = base64_value_to_char[0x3f & c >> 2];
3093 value = (0x03 & c) << 4;
3094
3095 /* Process second byte of a triplet. */
3096
3097 if (i == length)
3098 {
3099 *e++ = base64_value_to_char[value];
3100 *e++ = '=';
3101 *e++ = '=';
3102 break;
3103 }
3104
3105 if (multibyte)
3106 {
3107 c = STRING_CHAR_AND_LENGTH ((unsigned char *) from + i, bytes);
3108 if (CHAR_BYTE8_P (c))
3109 c = CHAR_TO_BYTE8 (c);
3110 else if (c >= 256)
3111 return -1;
3112 i += bytes;
3113 }
3114 else
3115 c = from[i++];
3116
3117 *e++ = base64_value_to_char[value | (0x0f & c >> 4)];
3118 value = (0x0f & c) << 2;
3119
3120 /* Process third byte of a triplet. */
3121
3122 if (i == length)
3123 {
3124 *e++ = base64_value_to_char[value];
3125 *e++ = '=';
3126 break;
3127 }
3128
3129 if (multibyte)
3130 {
3131 c = STRING_CHAR_AND_LENGTH ((unsigned char *) from + i, bytes);
3132 if (CHAR_BYTE8_P (c))
3133 c = CHAR_TO_BYTE8 (c);
3134 else if (c >= 256)
3135 return -1;
3136 i += bytes;
3137 }
3138 else
3139 c = from[i++];
3140
3141 *e++ = base64_value_to_char[value | (0x03 & c >> 6)];
3142 *e++ = base64_value_to_char[0x3f & c];
3143 }
3144
3145 return e - to;
3146 }
3147
3148
3149 DEFUN ("base64-decode-region", Fbase64_decode_region, Sbase64_decode_region,
3150 2, 2, "r",
3151 doc: /* Base64-decode the region between BEG and END.
3152 Return the length of the decoded text.
3153 If the region can't be decoded, signal an error and don't modify the buffer. */)
3154 (Lisp_Object beg, Lisp_Object end)
3155 {
3156 ptrdiff_t ibeg, iend, length, allength;
3157 char *decoded;
3158 ptrdiff_t old_pos = PT;
3159 ptrdiff_t decoded_length;
3160 ptrdiff_t inserted_chars;
3161 int multibyte = !NILP (BVAR (current_buffer, enable_multibyte_characters));
3162 USE_SAFE_ALLOCA;
3163
3164 validate_region (&beg, &end);
3165
3166 ibeg = CHAR_TO_BYTE (XFASTINT (beg));
3167 iend = CHAR_TO_BYTE (XFASTINT (end));
3168
3169 length = iend - ibeg;
3170
3171 /* We need to allocate enough room for decoding the text. If we are
3172 working on a multibyte buffer, each decoded code may occupy at
3173 most two bytes. */
3174 allength = multibyte ? length * 2 : length;
3175 SAFE_ALLOCA (decoded, char *, allength);
3176
3177 move_gap_both (XFASTINT (beg), ibeg);
3178 decoded_length = base64_decode_1 ((char *) BYTE_POS_ADDR (ibeg),
3179 decoded, length,
3180 multibyte, &inserted_chars);
3181 if (decoded_length > allength)
3182 abort ();
3183
3184 if (decoded_length < 0)
3185 {
3186 /* The decoding wasn't possible. */
3187 SAFE_FREE ();
3188 error ("Invalid base64 data");
3189 }
3190
3191 /* Now we have decoded the region, so we insert the new contents
3192 and delete the old. (Insert first in order to preserve markers.) */
3193 TEMP_SET_PT_BOTH (XFASTINT (beg), ibeg);
3194 insert_1_both (decoded, inserted_chars, decoded_length, 0, 1, 0);
3195 SAFE_FREE ();
3196
3197 /* Delete the original text. */
3198 del_range_both (PT, PT_BYTE, XFASTINT (end) + inserted_chars,
3199 iend + decoded_length, 1);
3200
3201 /* If point was outside of the region, restore it exactly; else just
3202 move to the beginning of the region. */
3203 if (old_pos >= XFASTINT (end))
3204 old_pos += inserted_chars - (XFASTINT (end) - XFASTINT (beg));
3205 else if (old_pos > XFASTINT (beg))
3206 old_pos = XFASTINT (beg);
3207 SET_PT (old_pos > ZV ? ZV : old_pos);
3208
3209 return make_number (inserted_chars);
3210 }
3211
3212 DEFUN ("base64-decode-string", Fbase64_decode_string, Sbase64_decode_string,
3213 1, 1, 0,
3214 doc: /* Base64-decode STRING and return the result. */)
3215 (Lisp_Object string)
3216 {
3217 char *decoded;
3218 ptrdiff_t length, decoded_length;
3219 Lisp_Object decoded_string;
3220 USE_SAFE_ALLOCA;
3221
3222 CHECK_STRING (string);
3223
3224 length = SBYTES (string);
3225 /* We need to allocate enough room for decoding the text. */
3226 SAFE_ALLOCA (decoded, char *, length);
3227
3228 /* The decoded result should be unibyte. */
3229 decoded_length = base64_decode_1 (SSDATA (string), decoded, length,
3230 0, NULL);
3231 if (decoded_length > length)
3232 abort ();
3233 else if (decoded_length >= 0)
3234 decoded_string = make_unibyte_string (decoded, decoded_length);
3235 else
3236 decoded_string = Qnil;
3237
3238 SAFE_FREE ();
3239 if (!STRINGP (decoded_string))
3240 error ("Invalid base64 data");
3241
3242 return decoded_string;
3243 }
3244
3245 /* Base64-decode the data at FROM of LENGTH bytes into TO. If
3246 MULTIBYTE is nonzero, the decoded result should be in multibyte
3247 form. If NCHARS_RETURN is not NULL, store the number of produced
3248 characters in *NCHARS_RETURN. */
3249
3250 static ptrdiff_t
3251 base64_decode_1 (const char *from, char *to, ptrdiff_t length,
3252 int multibyte, ptrdiff_t *nchars_return)
3253 {
3254 ptrdiff_t i = 0; /* Used inside READ_QUADRUPLET_BYTE */
3255 char *e = to;
3256 unsigned char c;
3257 unsigned long value;
3258 ptrdiff_t nchars = 0;
3259
3260 while (1)
3261 {
3262 /* Process first byte of a quadruplet. */
3263
3264 READ_QUADRUPLET_BYTE (e-to);
3265
3266 if (!IS_BASE64 (c))
3267 return -1;
3268 value = base64_char_to_value[c] << 18;
3269
3270 /* Process second byte of a quadruplet. */
3271
3272 READ_QUADRUPLET_BYTE (-1);
3273
3274 if (!IS_BASE64 (c))
3275 return -1;
3276 value |= base64_char_to_value[c] << 12;
3277
3278 c = (unsigned char) (value >> 16);
3279 if (multibyte && c >= 128)
3280 e += BYTE8_STRING (c, e);
3281 else
3282 *e++ = c;
3283 nchars++;
3284
3285 /* Process third byte of a quadruplet. */
3286
3287 READ_QUADRUPLET_BYTE (-1);
3288
3289 if (c == '=')
3290 {
3291 READ_QUADRUPLET_BYTE (-1);
3292
3293 if (c != '=')
3294 return -1;
3295 continue;
3296 }
3297
3298 if (!IS_BASE64 (c))
3299 return -1;
3300 value |= base64_char_to_value[c] << 6;
3301
3302 c = (unsigned char) (0xff & value >> 8);
3303 if (multibyte && c >= 128)
3304 e += BYTE8_STRING (c, e);
3305 else
3306 *e++ = c;
3307 nchars++;
3308
3309 /* Process fourth byte of a quadruplet. */
3310
3311 READ_QUADRUPLET_BYTE (-1);
3312
3313 if (c == '=')
3314 continue;
3315
3316 if (!IS_BASE64 (c))
3317 return -1;
3318 value |= base64_char_to_value[c];
3319
3320 c = (unsigned char) (0xff & value);
3321 if (multibyte && c >= 128)
3322 e += BYTE8_STRING (c, e);
3323 else
3324 *e++ = c;
3325 nchars++;
3326 }
3327 }
3328
3329
3330 \f
3331 /***********************************************************************
3332 ***** *****
3333 ***** Hash Tables *****
3334 ***** *****
3335 ***********************************************************************/
3336
3337 /* Implemented by gerd@gnu.org. This hash table implementation was
3338 inspired by CMUCL hash tables. */
3339
3340 /* Ideas:
3341
3342 1. For small tables, association lists are probably faster than
3343 hash tables because they have lower overhead.
3344
3345 For uses of hash tables where the O(1) behavior of table
3346 operations is not a requirement, it might therefore be a good idea
3347 not to hash. Instead, we could just do a linear search in the
3348 key_and_value vector of the hash table. This could be done
3349 if a `:linear-search t' argument is given to make-hash-table. */
3350
3351
3352 /* The list of all weak hash tables. Don't staticpro this one. */
3353
3354 static struct Lisp_Hash_Table *weak_hash_tables;
3355
3356 /* Various symbols. */
3357
3358 static Lisp_Object Qhash_table_p, Qkey, Qvalue;
3359 Lisp_Object Qeq, Qeql, Qequal;
3360 Lisp_Object QCtest, QCsize, QCrehash_size, QCrehash_threshold, QCweakness;
3361 static Lisp_Object Qhash_table_test, Qkey_or_value, Qkey_and_value;
3362
3363 /* Function prototypes. */
3364
3365 static struct Lisp_Hash_Table *check_hash_table (Lisp_Object);
3366 static ptrdiff_t get_key_arg (Lisp_Object, ptrdiff_t, Lisp_Object *, char *);
3367 static void maybe_resize_hash_table (struct Lisp_Hash_Table *);
3368 static int sweep_weak_table (struct Lisp_Hash_Table *, int);
3369
3370
3371 \f
3372 /***********************************************************************
3373 Utilities
3374 ***********************************************************************/
3375
3376 /* If OBJ is a Lisp hash table, return a pointer to its struct
3377 Lisp_Hash_Table. Otherwise, signal an error. */
3378
3379 static struct Lisp_Hash_Table *
3380 check_hash_table (Lisp_Object obj)
3381 {
3382 CHECK_HASH_TABLE (obj);
3383 return XHASH_TABLE (obj);
3384 }
3385
3386
3387 /* Value is the next integer I >= N, N >= 0 which is "almost" a prime
3388 number. A number is "almost" a prime number if it is not divisible
3389 by any integer in the range 2 .. (NEXT_ALMOST_PRIME_LIMIT - 1). */
3390
3391 EMACS_INT
3392 next_almost_prime (EMACS_INT n)
3393 {
3394 verify (NEXT_ALMOST_PRIME_LIMIT == 11);
3395 for (n |= 1; ; n += 2)
3396 if (n % 3 != 0 && n % 5 != 0 && n % 7 != 0)
3397 return n;
3398 }
3399
3400
3401 /* Find KEY in ARGS which has size NARGS. Don't consider indices for
3402 which USED[I] is non-zero. If found at index I in ARGS, set
3403 USED[I] and USED[I + 1] to 1, and return I + 1. Otherwise return
3404 0. This function is used to extract a keyword/argument pair from
3405 a DEFUN parameter list. */
3406
3407 static ptrdiff_t
3408 get_key_arg (Lisp_Object key, ptrdiff_t nargs, Lisp_Object *args, char *used)
3409 {
3410 ptrdiff_t i;
3411
3412 for (i = 1; i < nargs; i++)
3413 if (!used[i - 1] && EQ (args[i - 1], key))
3414 {
3415 used[i - 1] = 1;
3416 used[i] = 1;
3417 return i;
3418 }
3419
3420 return 0;
3421 }
3422
3423
3424 /* Return a Lisp vector which has the same contents as VEC but has
3425 at least INCR_MIN more entries, where INCR_MIN is positive.
3426 If NITEMS_MAX is not -1, do not grow the vector to be any larger
3427 than NITEMS_MAX. Entries in the resulting
3428 vector that are not copied from VEC are set to nil. */
3429
3430 Lisp_Object
3431 larger_vector (Lisp_Object vec, ptrdiff_t incr_min, ptrdiff_t nitems_max)
3432 {
3433 struct Lisp_Vector *v;
3434 ptrdiff_t i, incr, incr_max, old_size, new_size;
3435 ptrdiff_t C_language_max = min (PTRDIFF_MAX, SIZE_MAX) / sizeof *v->contents;
3436 ptrdiff_t n_max = (0 <= nitems_max && nitems_max < C_language_max
3437 ? nitems_max : C_language_max);
3438 eassert (VECTORP (vec));
3439 eassert (0 < incr_min && -1 <= nitems_max);
3440 old_size = ASIZE (vec);
3441 incr_max = n_max - old_size;
3442 incr = max (incr_min, min (old_size >> 1, incr_max));
3443 if (incr_max < incr)
3444 memory_full (SIZE_MAX);
3445 new_size = old_size + incr;
3446 v = allocate_vector (new_size);
3447 memcpy (v->contents, XVECTOR (vec)->contents, old_size * sizeof *v->contents);
3448 for (i = old_size; i < new_size; ++i)
3449 v->contents[i] = Qnil;
3450 XSETVECTOR (vec, v);
3451 return vec;
3452 }
3453
3454
3455 /***********************************************************************
3456 Low-level Functions
3457 ***********************************************************************/
3458
3459 /* Compare KEY1 which has hash code HASH1 and KEY2 with hash code
3460 HASH2 in hash table H using `eql'. Value is non-zero if KEY1 and
3461 KEY2 are the same. */
3462
3463 static int
3464 cmpfn_eql (struct Lisp_Hash_Table *h,
3465 Lisp_Object key1, EMACS_UINT hash1,
3466 Lisp_Object key2, EMACS_UINT hash2)
3467 {
3468 return (FLOATP (key1)
3469 && FLOATP (key2)
3470 && XFLOAT_DATA (key1) == XFLOAT_DATA (key2));
3471 }
3472
3473
3474 /* Compare KEY1 which has hash code HASH1 and KEY2 with hash code
3475 HASH2 in hash table H using `equal'. Value is non-zero if KEY1 and
3476 KEY2 are the same. */
3477
3478 static int
3479 cmpfn_equal (struct Lisp_Hash_Table *h,
3480 Lisp_Object key1, EMACS_UINT hash1,
3481 Lisp_Object key2, EMACS_UINT hash2)
3482 {
3483 return hash1 == hash2 && !NILP (Fequal (key1, key2));
3484 }
3485
3486
3487 /* Compare KEY1 which has hash code HASH1, and KEY2 with hash code
3488 HASH2 in hash table H using H->user_cmp_function. Value is non-zero
3489 if KEY1 and KEY2 are the same. */
3490
3491 static int
3492 cmpfn_user_defined (struct Lisp_Hash_Table *h,
3493 Lisp_Object key1, EMACS_UINT hash1,
3494 Lisp_Object key2, EMACS_UINT hash2)
3495 {
3496 if (hash1 == hash2)
3497 {
3498 Lisp_Object args[3];
3499
3500 args[0] = h->user_cmp_function;
3501 args[1] = key1;
3502 args[2] = key2;
3503 return !NILP (Ffuncall (3, args));
3504 }
3505 else
3506 return 0;
3507 }
3508
3509
3510 /* Value is a hash code for KEY for use in hash table H which uses
3511 `eq' to compare keys. The hash code returned is guaranteed to fit
3512 in a Lisp integer. */
3513
3514 static EMACS_UINT
3515 hashfn_eq (struct Lisp_Hash_Table *h, Lisp_Object key)
3516 {
3517 EMACS_UINT hash = XUINT (key) ^ XTYPE (key);
3518 eassert ((hash & ~INTMASK) == 0);
3519 return hash;
3520 }
3521
3522
3523 /* Value is a hash code for KEY for use in hash table H which uses
3524 `eql' to compare keys. The hash code returned is guaranteed to fit
3525 in a Lisp integer. */
3526
3527 static EMACS_UINT
3528 hashfn_eql (struct Lisp_Hash_Table *h, Lisp_Object key)
3529 {
3530 EMACS_UINT hash;
3531 if (FLOATP (key))
3532 hash = sxhash (key, 0);
3533 else
3534 hash = XUINT (key) ^ XTYPE (key);
3535 eassert ((hash & ~INTMASK) == 0);
3536 return hash;
3537 }
3538
3539
3540 /* Value is a hash code for KEY for use in hash table H which uses
3541 `equal' to compare keys. The hash code returned is guaranteed to fit
3542 in a Lisp integer. */
3543
3544 static EMACS_UINT
3545 hashfn_equal (struct Lisp_Hash_Table *h, Lisp_Object key)
3546 {
3547 EMACS_UINT hash = sxhash (key, 0);
3548 eassert ((hash & ~INTMASK) == 0);
3549 return hash;
3550 }
3551
3552
3553 /* Value is a hash code for KEY for use in hash table H which uses as
3554 user-defined function to compare keys. The hash code returned is
3555 guaranteed to fit in a Lisp integer. */
3556
3557 static EMACS_UINT
3558 hashfn_user_defined (struct Lisp_Hash_Table *h, Lisp_Object key)
3559 {
3560 Lisp_Object args[2], hash;
3561
3562 args[0] = h->user_hash_function;
3563 args[1] = key;
3564 hash = Ffuncall (2, args);
3565 if (!INTEGERP (hash))
3566 signal_error ("Invalid hash code returned from user-supplied hash function", hash);
3567 return XUINT (hash);
3568 }
3569
3570 /* An upper bound on the size of a hash table index. It must fit in
3571 ptrdiff_t and be a valid Emacs fixnum. */
3572 #define INDEX_SIZE_BOUND \
3573 ((ptrdiff_t) min (MOST_POSITIVE_FIXNUM, PTRDIFF_MAX / sizeof (Lisp_Object)))
3574
3575 /* Create and initialize a new hash table.
3576
3577 TEST specifies the test the hash table will use to compare keys.
3578 It must be either one of the predefined tests `eq', `eql' or
3579 `equal' or a symbol denoting a user-defined test named TEST with
3580 test and hash functions USER_TEST and USER_HASH.
3581
3582 Give the table initial capacity SIZE, SIZE >= 0, an integer.
3583
3584 If REHASH_SIZE is an integer, it must be > 0, and this hash table's
3585 new size when it becomes full is computed by adding REHASH_SIZE to
3586 its old size. If REHASH_SIZE is a float, it must be > 1.0, and the
3587 table's new size is computed by multiplying its old size with
3588 REHASH_SIZE.
3589
3590 REHASH_THRESHOLD must be a float <= 1.0, and > 0. The table will
3591 be resized when the ratio of (number of entries in the table) /
3592 (table size) is >= REHASH_THRESHOLD.
3593
3594 WEAK specifies the weakness of the table. If non-nil, it must be
3595 one of the symbols `key', `value', `key-or-value', or `key-and-value'. */
3596
3597 Lisp_Object
3598 make_hash_table (Lisp_Object test, Lisp_Object size, Lisp_Object rehash_size,
3599 Lisp_Object rehash_threshold, Lisp_Object weak,
3600 Lisp_Object user_test, Lisp_Object user_hash)
3601 {
3602 struct Lisp_Hash_Table *h;
3603 Lisp_Object table;
3604 EMACS_INT index_size, sz;
3605 ptrdiff_t i;
3606 double index_float;
3607
3608 /* Preconditions. */
3609 eassert (SYMBOLP (test));
3610 eassert (INTEGERP (size) && XINT (size) >= 0);
3611 eassert ((INTEGERP (rehash_size) && XINT (rehash_size) > 0)
3612 || (FLOATP (rehash_size) && 1 < XFLOAT_DATA (rehash_size)));
3613 eassert (FLOATP (rehash_threshold)
3614 && 0 < XFLOAT_DATA (rehash_threshold)
3615 && XFLOAT_DATA (rehash_threshold) <= 1.0);
3616
3617 if (XFASTINT (size) == 0)
3618 size = make_number (1);
3619
3620 sz = XFASTINT (size);
3621 index_float = sz / XFLOAT_DATA (rehash_threshold);
3622 index_size = (index_float < INDEX_SIZE_BOUND + 1
3623 ? next_almost_prime (index_float)
3624 : INDEX_SIZE_BOUND + 1);
3625 if (INDEX_SIZE_BOUND < max (index_size, 2 * sz))
3626 error ("Hash table too large");
3627
3628 /* Allocate a table and initialize it. */
3629 h = allocate_hash_table ();
3630
3631 /* Initialize hash table slots. */
3632 h->test = test;
3633 if (EQ (test, Qeql))
3634 {
3635 h->cmpfn = cmpfn_eql;
3636 h->hashfn = hashfn_eql;
3637 }
3638 else if (EQ (test, Qeq))
3639 {
3640 h->cmpfn = NULL;
3641 h->hashfn = hashfn_eq;
3642 }
3643 else if (EQ (test, Qequal))
3644 {
3645 h->cmpfn = cmpfn_equal;
3646 h->hashfn = hashfn_equal;
3647 }
3648 else
3649 {
3650 h->user_cmp_function = user_test;
3651 h->user_hash_function = user_hash;
3652 h->cmpfn = cmpfn_user_defined;
3653 h->hashfn = hashfn_user_defined;
3654 }
3655
3656 h->weak = weak;
3657 h->rehash_threshold = rehash_threshold;
3658 h->rehash_size = rehash_size;
3659 h->count = 0;
3660 h->key_and_value = Fmake_vector (make_number (2 * sz), Qnil);
3661 h->hash = Fmake_vector (size, Qnil);
3662 h->next = Fmake_vector (size, Qnil);
3663 h->index = Fmake_vector (make_number (index_size), Qnil);
3664
3665 /* Set up the free list. */
3666 for (i = 0; i < sz - 1; ++i)
3667 HASH_NEXT (h, i) = make_number (i + 1);
3668 h->next_free = make_number (0);
3669
3670 XSET_HASH_TABLE (table, h);
3671 eassert (HASH_TABLE_P (table));
3672 eassert (XHASH_TABLE (table) == h);
3673
3674 /* Maybe add this hash table to the list of all weak hash tables. */
3675 if (NILP (h->weak))
3676 h->next_weak = NULL;
3677 else
3678 {
3679 h->next_weak = weak_hash_tables;
3680 weak_hash_tables = h;
3681 }
3682
3683 return table;
3684 }
3685
3686
3687 /* Return a copy of hash table H1. Keys and values are not copied,
3688 only the table itself is. */
3689
3690 static Lisp_Object
3691 copy_hash_table (struct Lisp_Hash_Table *h1)
3692 {
3693 Lisp_Object table;
3694 struct Lisp_Hash_Table *h2;
3695 struct Lisp_Vector *next;
3696
3697 h2 = allocate_hash_table ();
3698 next = h2->header.next.vector;
3699 memcpy (h2, h1, sizeof *h2);
3700 h2->header.next.vector = next;
3701 h2->key_and_value = Fcopy_sequence (h1->key_and_value);
3702 h2->hash = Fcopy_sequence (h1->hash);
3703 h2->next = Fcopy_sequence (h1->next);
3704 h2->index = Fcopy_sequence (h1->index);
3705 XSET_HASH_TABLE (table, h2);
3706
3707 /* Maybe add this hash table to the list of all weak hash tables. */
3708 if (!NILP (h2->weak))
3709 {
3710 h2->next_weak = weak_hash_tables;
3711 weak_hash_tables = h2;
3712 }
3713
3714 return table;
3715 }
3716
3717
3718 /* Resize hash table H if it's too full. If H cannot be resized
3719 because it's already too large, throw an error. */
3720
3721 static inline void
3722 maybe_resize_hash_table (struct Lisp_Hash_Table *h)
3723 {
3724 if (NILP (h->next_free))
3725 {
3726 ptrdiff_t old_size = HASH_TABLE_SIZE (h);
3727 EMACS_INT new_size, index_size, nsize;
3728 ptrdiff_t i;
3729 double index_float;
3730
3731 if (INTEGERP (h->rehash_size))
3732 new_size = old_size + XFASTINT (h->rehash_size);
3733 else
3734 {
3735 double float_new_size = old_size * XFLOAT_DATA (h->rehash_size);
3736 if (float_new_size < INDEX_SIZE_BOUND + 1)
3737 {
3738 new_size = float_new_size;
3739 if (new_size <= old_size)
3740 new_size = old_size + 1;
3741 }
3742 else
3743 new_size = INDEX_SIZE_BOUND + 1;
3744 }
3745 index_float = new_size / XFLOAT_DATA (h->rehash_threshold);
3746 index_size = (index_float < INDEX_SIZE_BOUND + 1
3747 ? next_almost_prime (index_float)
3748 : INDEX_SIZE_BOUND + 1);
3749 nsize = max (index_size, 2 * new_size);
3750 if (INDEX_SIZE_BOUND < nsize)
3751 error ("Hash table too large to resize");
3752
3753 #ifdef ENABLE_CHECKING
3754 if (HASH_TABLE_P (Vpurify_flag)
3755 && XHASH_TABLE (Vpurify_flag) == h)
3756 {
3757 Lisp_Object args[2];
3758 args[0] = build_string ("Growing hash table to: %d");
3759 args[1] = make_number (new_size);
3760 Fmessage (2, args);
3761 }
3762 #endif
3763
3764 h->key_and_value = larger_vector (h->key_and_value,
3765 2 * (new_size - old_size), -1);
3766 h->next = larger_vector (h->next, new_size - old_size, -1);
3767 h->hash = larger_vector (h->hash, new_size - old_size, -1);
3768 h->index = Fmake_vector (make_number (index_size), Qnil);
3769
3770 /* Update the free list. Do it so that new entries are added at
3771 the end of the free list. This makes some operations like
3772 maphash faster. */
3773 for (i = old_size; i < new_size - 1; ++i)
3774 HASH_NEXT (h, i) = make_number (i + 1);
3775
3776 if (!NILP (h->next_free))
3777 {
3778 Lisp_Object last, next;
3779
3780 last = h->next_free;
3781 while (next = HASH_NEXT (h, XFASTINT (last)),
3782 !NILP (next))
3783 last = next;
3784
3785 HASH_NEXT (h, XFASTINT (last)) = make_number (old_size);
3786 }
3787 else
3788 XSETFASTINT (h->next_free, old_size);
3789
3790 /* Rehash. */
3791 for (i = 0; i < old_size; ++i)
3792 if (!NILP (HASH_HASH (h, i)))
3793 {
3794 EMACS_UINT hash_code = XUINT (HASH_HASH (h, i));
3795 ptrdiff_t start_of_bucket = hash_code % ASIZE (h->index);
3796 HASH_NEXT (h, i) = HASH_INDEX (h, start_of_bucket);
3797 HASH_INDEX (h, start_of_bucket) = make_number (i);
3798 }
3799 }
3800 }
3801
3802
3803 /* Lookup KEY in hash table H. If HASH is non-null, return in *HASH
3804 the hash code of KEY. Value is the index of the entry in H
3805 matching KEY, or -1 if not found. */
3806
3807 ptrdiff_t
3808 hash_lookup (struct Lisp_Hash_Table *h, Lisp_Object key, EMACS_UINT *hash)
3809 {
3810 EMACS_UINT hash_code;
3811 ptrdiff_t start_of_bucket;
3812 Lisp_Object idx;
3813
3814 hash_code = h->hashfn (h, key);
3815 if (hash)
3816 *hash = hash_code;
3817
3818 start_of_bucket = hash_code % ASIZE (h->index);
3819 idx = HASH_INDEX (h, start_of_bucket);
3820
3821 /* We need not gcpro idx since it's either an integer or nil. */
3822 while (!NILP (idx))
3823 {
3824 ptrdiff_t i = XFASTINT (idx);
3825 if (EQ (key, HASH_KEY (h, i))
3826 || (h->cmpfn
3827 && h->cmpfn (h, key, hash_code,
3828 HASH_KEY (h, i), XUINT (HASH_HASH (h, i)))))
3829 break;
3830 idx = HASH_NEXT (h, i);
3831 }
3832
3833 return NILP (idx) ? -1 : XFASTINT (idx);
3834 }
3835
3836
3837 /* Put an entry into hash table H that associates KEY with VALUE.
3838 HASH is a previously computed hash code of KEY.
3839 Value is the index of the entry in H matching KEY. */
3840
3841 ptrdiff_t
3842 hash_put (struct Lisp_Hash_Table *h, Lisp_Object key, Lisp_Object value,
3843 EMACS_UINT hash)
3844 {
3845 ptrdiff_t start_of_bucket, i;
3846
3847 eassert ((hash & ~INTMASK) == 0);
3848
3849 /* Increment count after resizing because resizing may fail. */
3850 maybe_resize_hash_table (h);
3851 h->count++;
3852
3853 /* Store key/value in the key_and_value vector. */
3854 i = XFASTINT (h->next_free);
3855 h->next_free = HASH_NEXT (h, i);
3856 HASH_KEY (h, i) = key;
3857 HASH_VALUE (h, i) = value;
3858
3859 /* Remember its hash code. */
3860 HASH_HASH (h, i) = make_number (hash);
3861
3862 /* Add new entry to its collision chain. */
3863 start_of_bucket = hash % ASIZE (h->index);
3864 HASH_NEXT (h, i) = HASH_INDEX (h, start_of_bucket);
3865 HASH_INDEX (h, start_of_bucket) = make_number (i);
3866 return i;
3867 }
3868
3869
3870 /* Remove the entry matching KEY from hash table H, if there is one. */
3871
3872 static void
3873 hash_remove_from_table (struct Lisp_Hash_Table *h, Lisp_Object key)
3874 {
3875 EMACS_UINT hash_code;
3876 ptrdiff_t start_of_bucket;
3877 Lisp_Object idx, prev;
3878
3879 hash_code = h->hashfn (h, key);
3880 start_of_bucket = hash_code % ASIZE (h->index);
3881 idx = HASH_INDEX (h, start_of_bucket);
3882 prev = Qnil;
3883
3884 /* We need not gcpro idx, prev since they're either integers or nil. */
3885 while (!NILP (idx))
3886 {
3887 ptrdiff_t i = XFASTINT (idx);
3888
3889 if (EQ (key, HASH_KEY (h, i))
3890 || (h->cmpfn
3891 && h->cmpfn (h, key, hash_code,
3892 HASH_KEY (h, i), XUINT (HASH_HASH (h, i)))))
3893 {
3894 /* Take entry out of collision chain. */
3895 if (NILP (prev))
3896 HASH_INDEX (h, start_of_bucket) = HASH_NEXT (h, i);
3897 else
3898 HASH_NEXT (h, XFASTINT (prev)) = HASH_NEXT (h, i);
3899
3900 /* Clear slots in key_and_value and add the slots to
3901 the free list. */
3902 HASH_KEY (h, i) = HASH_VALUE (h, i) = HASH_HASH (h, i) = Qnil;
3903 HASH_NEXT (h, i) = h->next_free;
3904 h->next_free = make_number (i);
3905 h->count--;
3906 eassert (h->count >= 0);
3907 break;
3908 }
3909 else
3910 {
3911 prev = idx;
3912 idx = HASH_NEXT (h, i);
3913 }
3914 }
3915 }
3916
3917
3918 /* Clear hash table H. */
3919
3920 static void
3921 hash_clear (struct Lisp_Hash_Table *h)
3922 {
3923 if (h->count > 0)
3924 {
3925 ptrdiff_t i, size = HASH_TABLE_SIZE (h);
3926
3927 for (i = 0; i < size; ++i)
3928 {
3929 HASH_NEXT (h, i) = i < size - 1 ? make_number (i + 1) : Qnil;
3930 HASH_KEY (h, i) = Qnil;
3931 HASH_VALUE (h, i) = Qnil;
3932 HASH_HASH (h, i) = Qnil;
3933 }
3934
3935 for (i = 0; i < ASIZE (h->index); ++i)
3936 ASET (h->index, i, Qnil);
3937
3938 h->next_free = make_number (0);
3939 h->count = 0;
3940 }
3941 }
3942
3943
3944 \f
3945 /************************************************************************
3946 Weak Hash Tables
3947 ************************************************************************/
3948
3949 void
3950 init_weak_hash_tables (void)
3951 {
3952 weak_hash_tables = NULL;
3953 }
3954
3955 /* Sweep weak hash table H. REMOVE_ENTRIES_P non-zero means remove
3956 entries from the table that don't survive the current GC.
3957 REMOVE_ENTRIES_P zero means mark entries that are in use. Value is
3958 non-zero if anything was marked. */
3959
3960 static int
3961 sweep_weak_table (struct Lisp_Hash_Table *h, int remove_entries_p)
3962 {
3963 ptrdiff_t bucket, n;
3964 int marked;
3965
3966 n = ASIZE (h->index) & ~ARRAY_MARK_FLAG;
3967 marked = 0;
3968
3969 for (bucket = 0; bucket < n; ++bucket)
3970 {
3971 Lisp_Object idx, next, prev;
3972
3973 /* Follow collision chain, removing entries that
3974 don't survive this garbage collection. */
3975 prev = Qnil;
3976 for (idx = HASH_INDEX (h, bucket); !NILP (idx); idx = next)
3977 {
3978 ptrdiff_t i = XFASTINT (idx);
3979 int key_known_to_survive_p = survives_gc_p (HASH_KEY (h, i));
3980 int value_known_to_survive_p = survives_gc_p (HASH_VALUE (h, i));
3981 int remove_p;
3982
3983 if (EQ (h->weak, Qkey))
3984 remove_p = !key_known_to_survive_p;
3985 else if (EQ (h->weak, Qvalue))
3986 remove_p = !value_known_to_survive_p;
3987 else if (EQ (h->weak, Qkey_or_value))
3988 remove_p = !(key_known_to_survive_p || value_known_to_survive_p);
3989 else if (EQ (h->weak, Qkey_and_value))
3990 remove_p = !(key_known_to_survive_p && value_known_to_survive_p);
3991 else
3992 abort ();
3993
3994 next = HASH_NEXT (h, i);
3995
3996 if (remove_entries_p)
3997 {
3998 if (remove_p)
3999 {
4000 /* Take out of collision chain. */
4001 if (NILP (prev))
4002 HASH_INDEX (h, bucket) = next;
4003 else
4004 HASH_NEXT (h, XFASTINT (prev)) = next;
4005
4006 /* Add to free list. */
4007 HASH_NEXT (h, i) = h->next_free;
4008 h->next_free = idx;
4009
4010 /* Clear key, value, and hash. */
4011 HASH_KEY (h, i) = HASH_VALUE (h, i) = Qnil;
4012 HASH_HASH (h, i) = Qnil;
4013
4014 h->count--;
4015 }
4016 else
4017 {
4018 prev = idx;
4019 }
4020 }
4021 else
4022 {
4023 if (!remove_p)
4024 {
4025 /* Make sure key and value survive. */
4026 if (!key_known_to_survive_p)
4027 {
4028 mark_object (HASH_KEY (h, i));
4029 marked = 1;
4030 }
4031
4032 if (!value_known_to_survive_p)
4033 {
4034 mark_object (HASH_VALUE (h, i));
4035 marked = 1;
4036 }
4037 }
4038 }
4039 }
4040 }
4041
4042 return marked;
4043 }
4044
4045 /* Remove elements from weak hash tables that don't survive the
4046 current garbage collection. Remove weak tables that don't survive
4047 from Vweak_hash_tables. Called from gc_sweep. */
4048
4049 void
4050 sweep_weak_hash_tables (void)
4051 {
4052 struct Lisp_Hash_Table *h, *used, *next;
4053 int marked;
4054
4055 /* Mark all keys and values that are in use. Keep on marking until
4056 there is no more change. This is necessary for cases like
4057 value-weak table A containing an entry X -> Y, where Y is used in a
4058 key-weak table B, Z -> Y. If B comes after A in the list of weak
4059 tables, X -> Y might be removed from A, although when looking at B
4060 one finds that it shouldn't. */
4061 do
4062 {
4063 marked = 0;
4064 for (h = weak_hash_tables; h; h = h->next_weak)
4065 {
4066 if (h->header.size & ARRAY_MARK_FLAG)
4067 marked |= sweep_weak_table (h, 0);
4068 }
4069 }
4070 while (marked);
4071
4072 /* Remove tables and entries that aren't used. */
4073 for (h = weak_hash_tables, used = NULL; h; h = next)
4074 {
4075 next = h->next_weak;
4076
4077 if (h->header.size & ARRAY_MARK_FLAG)
4078 {
4079 /* TABLE is marked as used. Sweep its contents. */
4080 if (h->count > 0)
4081 sweep_weak_table (h, 1);
4082
4083 /* Add table to the list of used weak hash tables. */
4084 h->next_weak = used;
4085 used = h;
4086 }
4087 }
4088
4089 weak_hash_tables = used;
4090 }
4091
4092
4093 \f
4094 /***********************************************************************
4095 Hash Code Computation
4096 ***********************************************************************/
4097
4098 /* Maximum depth up to which to dive into Lisp structures. */
4099
4100 #define SXHASH_MAX_DEPTH 3
4101
4102 /* Maximum length up to which to take list and vector elements into
4103 account. */
4104
4105 #define SXHASH_MAX_LEN 7
4106
4107 /* Combine two integers X and Y for hashing. The result might not fit
4108 into a Lisp integer. */
4109
4110 #define SXHASH_COMBINE(X, Y) \
4111 ((((EMACS_UINT) (X) << 4) + ((EMACS_UINT) (X) >> (BITS_PER_EMACS_INT - 4))) \
4112 + (EMACS_UINT) (Y))
4113
4114 /* Hash X, returning a value that fits into a Lisp integer. */
4115 #define SXHASH_REDUCE(X) \
4116 ((((X) ^ (X) >> (BITS_PER_EMACS_INT - FIXNUM_BITS))) & INTMASK)
4117
4118 /* Return a hash for string PTR which has length LEN. The hash value
4119 can be any EMACS_UINT value. */
4120
4121 EMACS_UINT
4122 hash_string (char const *ptr, ptrdiff_t len)
4123 {
4124 char const *p = ptr;
4125 char const *end = p + len;
4126 unsigned char c;
4127 EMACS_UINT hash = 0;
4128
4129 while (p != end)
4130 {
4131 c = *p++;
4132 hash = SXHASH_COMBINE (hash, c);
4133 }
4134
4135 return hash;
4136 }
4137
4138 /* Return a hash for string PTR which has length LEN. The hash
4139 code returned is guaranteed to fit in a Lisp integer. */
4140
4141 static EMACS_UINT
4142 sxhash_string (char const *ptr, ptrdiff_t len)
4143 {
4144 EMACS_UINT hash = hash_string (ptr, len);
4145 return SXHASH_REDUCE (hash);
4146 }
4147
4148 /* Return a hash for the floating point value VAL. */
4149
4150 static EMACS_INT
4151 sxhash_float (double val)
4152 {
4153 EMACS_UINT hash = 0;
4154 enum {
4155 WORDS_PER_DOUBLE = (sizeof val / sizeof hash
4156 + (sizeof val % sizeof hash != 0))
4157 };
4158 union {
4159 double val;
4160 EMACS_UINT word[WORDS_PER_DOUBLE];
4161 } u;
4162 int i;
4163 u.val = val;
4164 memset (&u.val + 1, 0, sizeof u - sizeof u.val);
4165 for (i = 0; i < WORDS_PER_DOUBLE; i++)
4166 hash = SXHASH_COMBINE (hash, u.word[i]);
4167 return SXHASH_REDUCE (hash);
4168 }
4169
4170 /* Return a hash for list LIST. DEPTH is the current depth in the
4171 list. We don't recurse deeper than SXHASH_MAX_DEPTH in it. */
4172
4173 static EMACS_UINT
4174 sxhash_list (Lisp_Object list, int depth)
4175 {
4176 EMACS_UINT hash = 0;
4177 int i;
4178
4179 if (depth < SXHASH_MAX_DEPTH)
4180 for (i = 0;
4181 CONSP (list) && i < SXHASH_MAX_LEN;
4182 list = XCDR (list), ++i)
4183 {
4184 EMACS_UINT hash2 = sxhash (XCAR (list), depth + 1);
4185 hash = SXHASH_COMBINE (hash, hash2);
4186 }
4187
4188 if (!NILP (list))
4189 {
4190 EMACS_UINT hash2 = sxhash (list, depth + 1);
4191 hash = SXHASH_COMBINE (hash, hash2);
4192 }
4193
4194 return SXHASH_REDUCE (hash);
4195 }
4196
4197
4198 /* Return a hash for vector VECTOR. DEPTH is the current depth in
4199 the Lisp structure. */
4200
4201 static EMACS_UINT
4202 sxhash_vector (Lisp_Object vec, int depth)
4203 {
4204 EMACS_UINT hash = ASIZE (vec);
4205 int i, n;
4206
4207 n = min (SXHASH_MAX_LEN, ASIZE (vec));
4208 for (i = 0; i < n; ++i)
4209 {
4210 EMACS_UINT hash2 = sxhash (AREF (vec, i), depth + 1);
4211 hash = SXHASH_COMBINE (hash, hash2);
4212 }
4213
4214 return SXHASH_REDUCE (hash);
4215 }
4216
4217 /* Return a hash for bool-vector VECTOR. */
4218
4219 static EMACS_UINT
4220 sxhash_bool_vector (Lisp_Object vec)
4221 {
4222 EMACS_UINT hash = XBOOL_VECTOR (vec)->size;
4223 int i, n;
4224
4225 n = min (SXHASH_MAX_LEN, XBOOL_VECTOR (vec)->header.size);
4226 for (i = 0; i < n; ++i)
4227 hash = SXHASH_COMBINE (hash, XBOOL_VECTOR (vec)->data[i]);
4228
4229 return SXHASH_REDUCE (hash);
4230 }
4231
4232
4233 /* Return a hash code for OBJ. DEPTH is the current depth in the Lisp
4234 structure. Value is an unsigned integer clipped to INTMASK. */
4235
4236 EMACS_UINT
4237 sxhash (Lisp_Object obj, int depth)
4238 {
4239 EMACS_UINT hash;
4240
4241 if (depth > SXHASH_MAX_DEPTH)
4242 return 0;
4243
4244 switch (XTYPE (obj))
4245 {
4246 case_Lisp_Int:
4247 hash = XUINT (obj);
4248 break;
4249
4250 case Lisp_Misc:
4251 hash = XUINT (obj);
4252 break;
4253
4254 case Lisp_Symbol:
4255 obj = SYMBOL_NAME (obj);
4256 /* Fall through. */
4257
4258 case Lisp_String:
4259 hash = sxhash_string (SSDATA (obj), SBYTES (obj));
4260 break;
4261
4262 /* This can be everything from a vector to an overlay. */
4263 case Lisp_Vectorlike:
4264 if (VECTORP (obj))
4265 /* According to the CL HyperSpec, two arrays are equal only if
4266 they are `eq', except for strings and bit-vectors. In
4267 Emacs, this works differently. We have to compare element
4268 by element. */
4269 hash = sxhash_vector (obj, depth);
4270 else if (BOOL_VECTOR_P (obj))
4271 hash = sxhash_bool_vector (obj);
4272 else
4273 /* Others are `equal' if they are `eq', so let's take their
4274 address as hash. */
4275 hash = XUINT (obj);
4276 break;
4277
4278 case Lisp_Cons:
4279 hash = sxhash_list (obj, depth);
4280 break;
4281
4282 case Lisp_Float:
4283 hash = sxhash_float (XFLOAT_DATA (obj));
4284 break;
4285
4286 default:
4287 abort ();
4288 }
4289
4290 return hash;
4291 }
4292
4293
4294 \f
4295 /***********************************************************************
4296 Lisp Interface
4297 ***********************************************************************/
4298
4299
4300 DEFUN ("sxhash", Fsxhash, Ssxhash, 1, 1, 0,
4301 doc: /* Compute a hash code for OBJ and return it as integer. */)
4302 (Lisp_Object obj)
4303 {
4304 EMACS_UINT hash = sxhash (obj, 0);
4305 return make_number (hash);
4306 }
4307
4308
4309 DEFUN ("make-hash-table", Fmake_hash_table, Smake_hash_table, 0, MANY, 0,
4310 doc: /* Create and return a new hash table.
4311
4312 Arguments are specified as keyword/argument pairs. The following
4313 arguments are defined:
4314
4315 :test TEST -- TEST must be a symbol that specifies how to compare
4316 keys. Default is `eql'. Predefined are the tests `eq', `eql', and
4317 `equal'. User-supplied test and hash functions can be specified via
4318 `define-hash-table-test'.
4319
4320 :size SIZE -- A hint as to how many elements will be put in the table.
4321 Default is 65.
4322
4323 :rehash-size REHASH-SIZE - Indicates how to expand the table when it
4324 fills up. If REHASH-SIZE is an integer, increase the size by that
4325 amount. If it is a float, it must be > 1.0, and the new size is the
4326 old size multiplied by that factor. Default is 1.5.
4327
4328 :rehash-threshold THRESHOLD -- THRESHOLD must a float > 0, and <= 1.0.
4329 Resize the hash table when the ratio (number of entries / table size)
4330 is greater than or equal to THRESHOLD. Default is 0.8.
4331
4332 :weakness WEAK -- WEAK must be one of nil, t, `key', `value',
4333 `key-or-value', or `key-and-value'. If WEAK is not nil, the table
4334 returned is a weak table. Key/value pairs are removed from a weak
4335 hash table when there are no non-weak references pointing to their
4336 key, value, one of key or value, or both key and value, depending on
4337 WEAK. WEAK t is equivalent to `key-and-value'. Default value of WEAK
4338 is nil.
4339
4340 usage: (make-hash-table &rest KEYWORD-ARGS) */)
4341 (ptrdiff_t nargs, Lisp_Object *args)
4342 {
4343 Lisp_Object test, size, rehash_size, rehash_threshold, weak;
4344 Lisp_Object user_test, user_hash;
4345 char *used;
4346 ptrdiff_t i;
4347
4348 /* The vector `used' is used to keep track of arguments that
4349 have been consumed. */
4350 used = (char *) alloca (nargs * sizeof *used);
4351 memset (used, 0, nargs * sizeof *used);
4352
4353 /* See if there's a `:test TEST' among the arguments. */
4354 i = get_key_arg (QCtest, nargs, args, used);
4355 test = i ? args[i] : Qeql;
4356 if (!EQ (test, Qeq) && !EQ (test, Qeql) && !EQ (test, Qequal))
4357 {
4358 /* See if it is a user-defined test. */
4359 Lisp_Object prop;
4360
4361 prop = Fget (test, Qhash_table_test);
4362 if (!CONSP (prop) || !CONSP (XCDR (prop)))
4363 signal_error ("Invalid hash table test", test);
4364 user_test = XCAR (prop);
4365 user_hash = XCAR (XCDR (prop));
4366 }
4367 else
4368 user_test = user_hash = Qnil;
4369
4370 /* See if there's a `:size SIZE' argument. */
4371 i = get_key_arg (QCsize, nargs, args, used);
4372 size = i ? args[i] : Qnil;
4373 if (NILP (size))
4374 size = make_number (DEFAULT_HASH_SIZE);
4375 else if (!INTEGERP (size) || XINT (size) < 0)
4376 signal_error ("Invalid hash table size", size);
4377
4378 /* Look for `:rehash-size SIZE'. */
4379 i = get_key_arg (QCrehash_size, nargs, args, used);
4380 rehash_size = i ? args[i] : make_float (DEFAULT_REHASH_SIZE);
4381 if (! ((INTEGERP (rehash_size) && 0 < XINT (rehash_size))
4382 || (FLOATP (rehash_size) && 1 < XFLOAT_DATA (rehash_size))))
4383 signal_error ("Invalid hash table rehash size", rehash_size);
4384
4385 /* Look for `:rehash-threshold THRESHOLD'. */
4386 i = get_key_arg (QCrehash_threshold, nargs, args, used);
4387 rehash_threshold = i ? args[i] : make_float (DEFAULT_REHASH_THRESHOLD);
4388 if (! (FLOATP (rehash_threshold)
4389 && 0 < XFLOAT_DATA (rehash_threshold)
4390 && XFLOAT_DATA (rehash_threshold) <= 1))
4391 signal_error ("Invalid hash table rehash threshold", rehash_threshold);
4392
4393 /* Look for `:weakness WEAK'. */
4394 i = get_key_arg (QCweakness, nargs, args, used);
4395 weak = i ? args[i] : Qnil;
4396 if (EQ (weak, Qt))
4397 weak = Qkey_and_value;
4398 if (!NILP (weak)
4399 && !EQ (weak, Qkey)
4400 && !EQ (weak, Qvalue)
4401 && !EQ (weak, Qkey_or_value)
4402 && !EQ (weak, Qkey_and_value))
4403 signal_error ("Invalid hash table weakness", weak);
4404
4405 /* Now, all args should have been used up, or there's a problem. */
4406 for (i = 0; i < nargs; ++i)
4407 if (!used[i])
4408 signal_error ("Invalid argument list", args[i]);
4409
4410 return make_hash_table (test, size, rehash_size, rehash_threshold, weak,
4411 user_test, user_hash);
4412 }
4413
4414
4415 DEFUN ("copy-hash-table", Fcopy_hash_table, Scopy_hash_table, 1, 1, 0,
4416 doc: /* Return a copy of hash table TABLE. */)
4417 (Lisp_Object table)
4418 {
4419 return copy_hash_table (check_hash_table (table));
4420 }
4421
4422
4423 DEFUN ("hash-table-count", Fhash_table_count, Shash_table_count, 1, 1, 0,
4424 doc: /* Return the number of elements in TABLE. */)
4425 (Lisp_Object table)
4426 {
4427 return make_number (check_hash_table (table)->count);
4428 }
4429
4430
4431 DEFUN ("hash-table-rehash-size", Fhash_table_rehash_size,
4432 Shash_table_rehash_size, 1, 1, 0,
4433 doc: /* Return the current rehash size of TABLE. */)
4434 (Lisp_Object table)
4435 {
4436 return check_hash_table (table)->rehash_size;
4437 }
4438
4439
4440 DEFUN ("hash-table-rehash-threshold", Fhash_table_rehash_threshold,
4441 Shash_table_rehash_threshold, 1, 1, 0,
4442 doc: /* Return the current rehash threshold of TABLE. */)
4443 (Lisp_Object table)
4444 {
4445 return check_hash_table (table)->rehash_threshold;
4446 }
4447
4448
4449 DEFUN ("hash-table-size", Fhash_table_size, Shash_table_size, 1, 1, 0,
4450 doc: /* Return the size of TABLE.
4451 The size can be used as an argument to `make-hash-table' to create
4452 a hash table than can hold as many elements as TABLE holds
4453 without need for resizing. */)
4454 (Lisp_Object table)
4455 {
4456 struct Lisp_Hash_Table *h = check_hash_table (table);
4457 return make_number (HASH_TABLE_SIZE (h));
4458 }
4459
4460
4461 DEFUN ("hash-table-test", Fhash_table_test, Shash_table_test, 1, 1, 0,
4462 doc: /* Return the test TABLE uses. */)
4463 (Lisp_Object table)
4464 {
4465 return check_hash_table (table)->test;
4466 }
4467
4468
4469 DEFUN ("hash-table-weakness", Fhash_table_weakness, Shash_table_weakness,
4470 1, 1, 0,
4471 doc: /* Return the weakness of TABLE. */)
4472 (Lisp_Object table)
4473 {
4474 return check_hash_table (table)->weak;
4475 }
4476
4477
4478 DEFUN ("hash-table-p", Fhash_table_p, Shash_table_p, 1, 1, 0,
4479 doc: /* Return t if OBJ is a Lisp hash table object. */)
4480 (Lisp_Object obj)
4481 {
4482 return HASH_TABLE_P (obj) ? Qt : Qnil;
4483 }
4484
4485
4486 DEFUN ("clrhash", Fclrhash, Sclrhash, 1, 1, 0,
4487 doc: /* Clear hash table TABLE and return it. */)
4488 (Lisp_Object table)
4489 {
4490 hash_clear (check_hash_table (table));
4491 /* Be compatible with XEmacs. */
4492 return table;
4493 }
4494
4495
4496 DEFUN ("gethash", Fgethash, Sgethash, 2, 3, 0,
4497 doc: /* Look up KEY in TABLE and return its associated value.
4498 If KEY is not found, return DFLT which defaults to nil. */)
4499 (Lisp_Object key, Lisp_Object table, Lisp_Object dflt)
4500 {
4501 struct Lisp_Hash_Table *h = check_hash_table (table);
4502 ptrdiff_t i = hash_lookup (h, key, NULL);
4503 return i >= 0 ? HASH_VALUE (h, i) : dflt;
4504 }
4505
4506
4507 DEFUN ("puthash", Fputhash, Sputhash, 3, 3, 0,
4508 doc: /* Associate KEY with VALUE in hash table TABLE.
4509 If KEY is already present in table, replace its current value with
4510 VALUE. In any case, return VALUE. */)
4511 (Lisp_Object key, Lisp_Object value, Lisp_Object table)
4512 {
4513 struct Lisp_Hash_Table *h = check_hash_table (table);
4514 ptrdiff_t i;
4515 EMACS_UINT hash;
4516
4517 i = hash_lookup (h, key, &hash);
4518 if (i >= 0)
4519 HASH_VALUE (h, i) = value;
4520 else
4521 hash_put (h, key, value, hash);
4522
4523 return value;
4524 }
4525
4526
4527 DEFUN ("remhash", Fremhash, Sremhash, 2, 2, 0,
4528 doc: /* Remove KEY from TABLE. */)
4529 (Lisp_Object key, Lisp_Object table)
4530 {
4531 struct Lisp_Hash_Table *h = check_hash_table (table);
4532 hash_remove_from_table (h, key);
4533 return Qnil;
4534 }
4535
4536
4537 DEFUN ("maphash", Fmaphash, Smaphash, 2, 2, 0,
4538 doc: /* Call FUNCTION for all entries in hash table TABLE.
4539 FUNCTION is called with two arguments, KEY and VALUE. */)
4540 (Lisp_Object function, Lisp_Object table)
4541 {
4542 struct Lisp_Hash_Table *h = check_hash_table (table);
4543 Lisp_Object args[3];
4544 ptrdiff_t i;
4545
4546 for (i = 0; i < HASH_TABLE_SIZE (h); ++i)
4547 if (!NILP (HASH_HASH (h, i)))
4548 {
4549 args[0] = function;
4550 args[1] = HASH_KEY (h, i);
4551 args[2] = HASH_VALUE (h, i);
4552 Ffuncall (3, args);
4553 }
4554
4555 return Qnil;
4556 }
4557
4558
4559 DEFUN ("define-hash-table-test", Fdefine_hash_table_test,
4560 Sdefine_hash_table_test, 3, 3, 0,
4561 doc: /* Define a new hash table test with name NAME, a symbol.
4562
4563 In hash tables created with NAME specified as test, use TEST to
4564 compare keys, and HASH for computing hash codes of keys.
4565
4566 TEST must be a function taking two arguments and returning non-nil if
4567 both arguments are the same. HASH must be a function taking one
4568 argument and return an integer that is the hash code of the argument.
4569 Hash code computation should use the whole value range of integers,
4570 including negative integers. */)
4571 (Lisp_Object name, Lisp_Object test, Lisp_Object hash)
4572 {
4573 return Fput (name, Qhash_table_test, list2 (test, hash));
4574 }
4575
4576
4577 \f
4578 /************************************************************************
4579 MD5, SHA-1, and SHA-2
4580 ************************************************************************/
4581
4582 #include "md5.h"
4583 #include "sha1.h"
4584 #include "sha256.h"
4585 #include "sha512.h"
4586
4587 /* ALGORITHM is a symbol: md5, sha1, sha224 and so on. */
4588
4589 static Lisp_Object
4590 secure_hash (Lisp_Object algorithm, Lisp_Object object, Lisp_Object start, Lisp_Object end, Lisp_Object coding_system, Lisp_Object noerror, Lisp_Object binary)
4591 {
4592 int i;
4593 ptrdiff_t size;
4594 EMACS_INT start_char = 0, end_char = 0;
4595 ptrdiff_t start_byte, end_byte;
4596 register EMACS_INT b, e;
4597 register struct buffer *bp;
4598 EMACS_INT temp;
4599 int digest_size;
4600 void *(*hash_func) (const char *, size_t, void *);
4601 Lisp_Object digest;
4602
4603 CHECK_SYMBOL (algorithm);
4604
4605 if (STRINGP (object))
4606 {
4607 if (NILP (coding_system))
4608 {
4609 /* Decide the coding-system to encode the data with. */
4610
4611 if (STRING_MULTIBYTE (object))
4612 /* use default, we can't guess correct value */
4613 coding_system = preferred_coding_system ();
4614 else
4615 coding_system = Qraw_text;
4616 }
4617
4618 if (NILP (Fcoding_system_p (coding_system)))
4619 {
4620 /* Invalid coding system. */
4621
4622 if (!NILP (noerror))
4623 coding_system = Qraw_text;
4624 else
4625 xsignal1 (Qcoding_system_error, coding_system);
4626 }
4627
4628 if (STRING_MULTIBYTE (object))
4629 object = code_convert_string (object, coding_system, Qnil, 1, 0, 1);
4630
4631 size = SCHARS (object);
4632
4633 if (!NILP (start))
4634 {
4635 CHECK_NUMBER (start);
4636
4637 start_char = XINT (start);
4638
4639 if (start_char < 0)
4640 start_char += size;
4641 }
4642
4643 if (NILP (end))
4644 end_char = size;
4645 else
4646 {
4647 CHECK_NUMBER (end);
4648
4649 end_char = XINT (end);
4650
4651 if (end_char < 0)
4652 end_char += size;
4653 }
4654
4655 if (!(0 <= start_char && start_char <= end_char && end_char <= size))
4656 args_out_of_range_3 (object, make_number (start_char),
4657 make_number (end_char));
4658
4659 start_byte = NILP (start) ? 0 : string_char_to_byte (object, start_char);
4660 end_byte =
4661 NILP (end) ? SBYTES (object) : string_char_to_byte (object, end_char);
4662 }
4663 else
4664 {
4665 struct buffer *prev = current_buffer;
4666
4667 record_unwind_protect (Fset_buffer, Fcurrent_buffer ());
4668
4669 CHECK_BUFFER (object);
4670
4671 bp = XBUFFER (object);
4672 if (bp != current_buffer)
4673 set_buffer_internal (bp);
4674
4675 if (NILP (start))
4676 b = BEGV;
4677 else
4678 {
4679 CHECK_NUMBER_COERCE_MARKER (start);
4680 b = XINT (start);
4681 }
4682
4683 if (NILP (end))
4684 e = ZV;
4685 else
4686 {
4687 CHECK_NUMBER_COERCE_MARKER (end);
4688 e = XINT (end);
4689 }
4690
4691 if (b > e)
4692 temp = b, b = e, e = temp;
4693
4694 if (!(BEGV <= b && e <= ZV))
4695 args_out_of_range (start, end);
4696
4697 if (NILP (coding_system))
4698 {
4699 /* Decide the coding-system to encode the data with.
4700 See fileio.c:Fwrite-region */
4701
4702 if (!NILP (Vcoding_system_for_write))
4703 coding_system = Vcoding_system_for_write;
4704 else
4705 {
4706 int force_raw_text = 0;
4707
4708 coding_system = BVAR (XBUFFER (object), buffer_file_coding_system);
4709 if (NILP (coding_system)
4710 || NILP (Flocal_variable_p (Qbuffer_file_coding_system, Qnil)))
4711 {
4712 coding_system = Qnil;
4713 if (NILP (BVAR (current_buffer, enable_multibyte_characters)))
4714 force_raw_text = 1;
4715 }
4716
4717 if (NILP (coding_system) && !NILP (Fbuffer_file_name (object)))
4718 {
4719 /* Check file-coding-system-alist. */
4720 Lisp_Object args[4], val;
4721
4722 args[0] = Qwrite_region; args[1] = start; args[2] = end;
4723 args[3] = Fbuffer_file_name (object);
4724 val = Ffind_operation_coding_system (4, args);
4725 if (CONSP (val) && !NILP (XCDR (val)))
4726 coding_system = XCDR (val);
4727 }
4728
4729 if (NILP (coding_system)
4730 && !NILP (BVAR (XBUFFER (object), buffer_file_coding_system)))
4731 {
4732 /* If we still have not decided a coding system, use the
4733 default value of buffer-file-coding-system. */
4734 coding_system = BVAR (XBUFFER (object), buffer_file_coding_system);
4735 }
4736
4737 if (!force_raw_text
4738 && !NILP (Ffboundp (Vselect_safe_coding_system_function)))
4739 /* Confirm that VAL can surely encode the current region. */
4740 coding_system = call4 (Vselect_safe_coding_system_function,
4741 make_number (b), make_number (e),
4742 coding_system, Qnil);
4743
4744 if (force_raw_text)
4745 coding_system = Qraw_text;
4746 }
4747
4748 if (NILP (Fcoding_system_p (coding_system)))
4749 {
4750 /* Invalid coding system. */
4751
4752 if (!NILP (noerror))
4753 coding_system = Qraw_text;
4754 else
4755 xsignal1 (Qcoding_system_error, coding_system);
4756 }
4757 }
4758
4759 object = make_buffer_string (b, e, 0);
4760 if (prev != current_buffer)
4761 set_buffer_internal (prev);
4762 /* Discard the unwind protect for recovering the current
4763 buffer. */
4764 specpdl_ptr--;
4765
4766 if (STRING_MULTIBYTE (object))
4767 object = code_convert_string (object, coding_system, Qnil, 1, 0, 0);
4768 start_byte = 0;
4769 end_byte = SBYTES (object);
4770 }
4771
4772 if (EQ (algorithm, Qmd5))
4773 {
4774 digest_size = MD5_DIGEST_SIZE;
4775 hash_func = md5_buffer;
4776 }
4777 else if (EQ (algorithm, Qsha1))
4778 {
4779 digest_size = SHA1_DIGEST_SIZE;
4780 hash_func = sha1_buffer;
4781 }
4782 else if (EQ (algorithm, Qsha224))
4783 {
4784 digest_size = SHA224_DIGEST_SIZE;
4785 hash_func = sha224_buffer;
4786 }
4787 else if (EQ (algorithm, Qsha256))
4788 {
4789 digest_size = SHA256_DIGEST_SIZE;
4790 hash_func = sha256_buffer;
4791 }
4792 else if (EQ (algorithm, Qsha384))
4793 {
4794 digest_size = SHA384_DIGEST_SIZE;
4795 hash_func = sha384_buffer;
4796 }
4797 else if (EQ (algorithm, Qsha512))
4798 {
4799 digest_size = SHA512_DIGEST_SIZE;
4800 hash_func = sha512_buffer;
4801 }
4802 else
4803 error ("Invalid algorithm arg: %s", SDATA (Fsymbol_name (algorithm)));
4804
4805 /* allocate 2 x digest_size so that it can be re-used to hold the
4806 hexified value */
4807 digest = make_uninit_string (digest_size * 2);
4808
4809 hash_func (SSDATA (object) + start_byte,
4810 end_byte - start_byte,
4811 SSDATA (digest));
4812
4813 if (NILP (binary))
4814 {
4815 unsigned char *p = SDATA (digest);
4816 for (i = digest_size - 1; i >= 0; i--)
4817 {
4818 static char const hexdigit[16] = "0123456789abcdef";
4819 int p_i = p[i];
4820 p[2 * i] = hexdigit[p_i >> 4];
4821 p[2 * i + 1] = hexdigit[p_i & 0xf];
4822 }
4823 return digest;
4824 }
4825 else
4826 return make_unibyte_string (SSDATA (digest), digest_size);
4827 }
4828
4829 DEFUN ("md5", Fmd5, Smd5, 1, 5, 0,
4830 doc: /* Return MD5 message digest of OBJECT, a buffer or string.
4831
4832 A message digest is a cryptographic checksum of a document, and the
4833 algorithm to calculate it is defined in RFC 1321.
4834
4835 The two optional arguments START and END are character positions
4836 specifying for which part of OBJECT the message digest should be
4837 computed. If nil or omitted, the digest is computed for the whole
4838 OBJECT.
4839
4840 The MD5 message digest is computed from the result of encoding the
4841 text in a coding system, not directly from the internal Emacs form of
4842 the text. The optional fourth argument CODING-SYSTEM specifies which
4843 coding system to encode the text with. It should be the same coding
4844 system that you used or will use when actually writing the text into a
4845 file.
4846
4847 If CODING-SYSTEM is nil or omitted, the default depends on OBJECT. If
4848 OBJECT is a buffer, the default for CODING-SYSTEM is whatever coding
4849 system would be chosen by default for writing this text into a file.
4850
4851 If OBJECT is a string, the most preferred coding system (see the
4852 command `prefer-coding-system') is used.
4853
4854 If NOERROR is non-nil, silently assume the `raw-text' coding if the
4855 guesswork fails. Normally, an error is signaled in such case. */)
4856 (Lisp_Object object, Lisp_Object start, Lisp_Object end, Lisp_Object coding_system, Lisp_Object noerror)
4857 {
4858 return secure_hash (Qmd5, object, start, end, coding_system, noerror, Qnil);
4859 }
4860
4861 DEFUN ("secure-hash", Fsecure_hash, Ssecure_hash, 2, 5, 0,
4862 doc: /* Return the secure hash of OBJECT, a buffer or string.
4863 ALGORITHM is a symbol specifying the hash to use:
4864 md5, sha1, sha224, sha256, sha384 or sha512.
4865
4866 The two optional arguments START and END are positions specifying for
4867 which part of OBJECT to compute the hash. If nil or omitted, uses the
4868 whole OBJECT.
4869
4870 If BINARY is non-nil, returns a string in binary form. */)
4871 (Lisp_Object algorithm, Lisp_Object object, Lisp_Object start, Lisp_Object end, Lisp_Object binary)
4872 {
4873 return secure_hash (algorithm, object, start, end, Qnil, Qnil, binary);
4874 }
4875 \f
4876 void
4877 syms_of_fns (void)
4878 {
4879 DEFSYM (Qmd5, "md5");
4880 DEFSYM (Qsha1, "sha1");
4881 DEFSYM (Qsha224, "sha224");
4882 DEFSYM (Qsha256, "sha256");
4883 DEFSYM (Qsha384, "sha384");
4884 DEFSYM (Qsha512, "sha512");
4885
4886 /* Hash table stuff. */
4887 DEFSYM (Qhash_table_p, "hash-table-p");
4888 DEFSYM (Qeq, "eq");
4889 DEFSYM (Qeql, "eql");
4890 DEFSYM (Qequal, "equal");
4891 DEFSYM (QCtest, ":test");
4892 DEFSYM (QCsize, ":size");
4893 DEFSYM (QCrehash_size, ":rehash-size");
4894 DEFSYM (QCrehash_threshold, ":rehash-threshold");
4895 DEFSYM (QCweakness, ":weakness");
4896 DEFSYM (Qkey, "key");
4897 DEFSYM (Qvalue, "value");
4898 DEFSYM (Qhash_table_test, "hash-table-test");
4899 DEFSYM (Qkey_or_value, "key-or-value");
4900 DEFSYM (Qkey_and_value, "key-and-value");
4901
4902 defsubr (&Ssxhash);
4903 defsubr (&Smake_hash_table);
4904 defsubr (&Scopy_hash_table);
4905 defsubr (&Shash_table_count);
4906 defsubr (&Shash_table_rehash_size);
4907 defsubr (&Shash_table_rehash_threshold);
4908 defsubr (&Shash_table_size);
4909 defsubr (&Shash_table_test);
4910 defsubr (&Shash_table_weakness);
4911 defsubr (&Shash_table_p);
4912 defsubr (&Sclrhash);
4913 defsubr (&Sgethash);
4914 defsubr (&Sputhash);
4915 defsubr (&Sremhash);
4916 defsubr (&Smaphash);
4917 defsubr (&Sdefine_hash_table_test);
4918
4919 DEFSYM (Qstring_lessp, "string-lessp");
4920 DEFSYM (Qprovide, "provide");
4921 DEFSYM (Qrequire, "require");
4922 DEFSYM (Qyes_or_no_p_history, "yes-or-no-p-history");
4923 DEFSYM (Qcursor_in_echo_area, "cursor-in-echo-area");
4924 DEFSYM (Qwidget_type, "widget-type");
4925
4926 staticpro (&string_char_byte_cache_string);
4927 string_char_byte_cache_string = Qnil;
4928
4929 require_nesting_list = Qnil;
4930 staticpro (&require_nesting_list);
4931
4932 Fset (Qyes_or_no_p_history, Qnil);
4933
4934 DEFVAR_LISP ("features", Vfeatures,
4935 doc: /* A list of symbols which are the features of the executing Emacs.
4936 Used by `featurep' and `require', and altered by `provide'. */);
4937 Vfeatures = Fcons (intern_c_string ("emacs"), Qnil);
4938 DEFSYM (Qsubfeatures, "subfeatures");
4939
4940 #ifdef HAVE_LANGINFO_CODESET
4941 DEFSYM (Qcodeset, "codeset");
4942 DEFSYM (Qdays, "days");
4943 DEFSYM (Qmonths, "months");
4944 DEFSYM (Qpaper, "paper");
4945 #endif /* HAVE_LANGINFO_CODESET */
4946
4947 DEFVAR_BOOL ("use-dialog-box", use_dialog_box,
4948 doc: /* Non-nil means mouse commands use dialog boxes to ask questions.
4949 This applies to `y-or-n-p' and `yes-or-no-p' questions asked by commands
4950 invoked by mouse clicks and mouse menu items.
4951
4952 On some platforms, file selection dialogs are also enabled if this is
4953 non-nil. */);
4954 use_dialog_box = 1;
4955
4956 DEFVAR_BOOL ("use-file-dialog", use_file_dialog,
4957 doc: /* Non-nil means mouse commands use a file dialog to ask for files.
4958 This applies to commands from menus and tool bar buttons even when
4959 they are initiated from the keyboard. If `use-dialog-box' is nil,
4960 that disables the use of a file dialog, regardless of the value of
4961 this variable. */);
4962 use_file_dialog = 1;
4963
4964 defsubr (&Sidentity);
4965 defsubr (&Srandom);
4966 defsubr (&Slength);
4967 defsubr (&Ssafe_length);
4968 defsubr (&Sstring_bytes);
4969 defsubr (&Sstring_equal);
4970 defsubr (&Scompare_strings);
4971 defsubr (&Sstring_lessp);
4972 defsubr (&Sappend);
4973 defsubr (&Sconcat);
4974 defsubr (&Svconcat);
4975 defsubr (&Scopy_sequence);
4976 defsubr (&Sstring_make_multibyte);
4977 defsubr (&Sstring_make_unibyte);
4978 defsubr (&Sstring_as_multibyte);
4979 defsubr (&Sstring_as_unibyte);
4980 defsubr (&Sstring_to_multibyte);
4981 defsubr (&Sstring_to_unibyte);
4982 defsubr (&Scopy_alist);
4983 defsubr (&Ssubstring);
4984 defsubr (&Ssubstring_no_properties);
4985 defsubr (&Snthcdr);
4986 defsubr (&Snth);
4987 defsubr (&Selt);
4988 defsubr (&Smember);
4989 defsubr (&Smemq);
4990 defsubr (&Smemql);
4991 defsubr (&Sassq);
4992 defsubr (&Sassoc);
4993 defsubr (&Srassq);
4994 defsubr (&Srassoc);
4995 defsubr (&Sdelq);
4996 defsubr (&Sdelete);
4997 defsubr (&Snreverse);
4998 defsubr (&Sreverse);
4999 defsubr (&Ssort);
5000 defsubr (&Splist_get);
5001 defsubr (&Sget);
5002 defsubr (&Splist_put);
5003 defsubr (&Sput);
5004 defsubr (&Slax_plist_get);
5005 defsubr (&Slax_plist_put);
5006 defsubr (&Seql);
5007 defsubr (&Sequal);
5008 defsubr (&Sequal_including_properties);
5009 defsubr (&Sfillarray);
5010 defsubr (&Sclear_string);
5011 defsubr (&Snconc);
5012 defsubr (&Smapcar);
5013 defsubr (&Smapc);
5014 defsubr (&Smapconcat);
5015 defsubr (&Syes_or_no_p);
5016 defsubr (&Sload_average);
5017 defsubr (&Sfeaturep);
5018 defsubr (&Srequire);
5019 defsubr (&Sprovide);
5020 defsubr (&Splist_member);
5021 defsubr (&Swidget_put);
5022 defsubr (&Swidget_get);
5023 defsubr (&Swidget_apply);
5024 defsubr (&Sbase64_encode_region);
5025 defsubr (&Sbase64_decode_region);
5026 defsubr (&Sbase64_encode_string);
5027 defsubr (&Sbase64_decode_string);
5028 defsubr (&Smd5);
5029 defsubr (&Ssecure_hash);
5030 defsubr (&Slocale_info);
5031 }
5032
5033
5034 void
5035 init_fns (void)
5036 {
5037 }