e275df71b49168567a78419dd67a4c505fa4a101
[bpt/emacs.git] / src / fns.c
1 /* Random utility Lisp functions.
2
3 Copyright (C) 1985-1987, 1993-1995, 1997-2014 Free Software Foundation,
4 Inc.
5
6 This file is part of GNU Emacs.
7
8 GNU Emacs is free software: you can redistribute it and/or modify
9 it under the terms of the GNU General Public License as published by
10 the Free Software Foundation, either version 3 of the License, or
11 (at your option) any later version.
12
13 GNU Emacs is distributed in the hope that it will be useful,
14 but WITHOUT ANY WARRANTY; without even the implied warranty of
15 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
16 GNU General Public License for more details.
17
18 You should have received a copy of the GNU General Public License
19 along with GNU Emacs. If not, see <http://www.gnu.org/licenses/>. */
20
21 #include <config.h>
22
23 #include <unistd.h>
24 #include <time.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 #if defined (HAVE_X_WINDOWS)
40 #include "xterm.h"
41 #endif
42
43 Lisp_Object Qstring_lessp;
44 static Lisp_Object Qprovide, Qrequire;
45 static Lisp_Object Qyes_or_no_p_history;
46 Lisp_Object Qcursor_in_echo_area;
47 static Lisp_Object Qwidget_type;
48 static Lisp_Object Qcodeset, Qdays, Qmonths, Qpaper;
49
50 static Lisp_Object Qmd5, Qsha1, Qsha224, Qsha256, Qsha384, Qsha512;
51 \f
52 DEFUN ("identity", Fidentity, Sidentity, 1, 1, 0,
53 doc: /* Return the argument unchanged. */)
54 (Lisp_Object arg)
55 {
56 return arg;
57 }
58
59 DEFUN ("random", Frandom, Srandom, 0, 1, 0,
60 doc: /* Return a pseudo-random number.
61 All integers representable in Lisp, i.e. between `most-negative-fixnum'
62 and `most-positive-fixnum', inclusive, are equally likely.
63
64 With positive integer LIMIT, return random number in interval [0,LIMIT).
65 With argument t, set the random number seed from the current time and pid.
66 With a string argument, set the seed based on the string's contents.
67 Other values of LIMIT are ignored.
68
69 See Info node `(elisp)Random Numbers' for more details. */)
70 (Lisp_Object limit)
71 {
72 EMACS_INT val;
73
74 if (EQ (limit, Qt))
75 init_random ();
76 else if (STRINGP (limit))
77 seed_random (SSDATA (limit), SBYTES (limit));
78
79 val = get_random ();
80 if (INTEGERP (limit) && 0 < XINT (limit))
81 while (true)
82 {
83 /* Return the remainder, except reject the rare case where
84 get_random returns a number so close to INTMASK that the
85 remainder isn't random. */
86 EMACS_INT remainder = val % XINT (limit);
87 if (val - remainder <= INTMASK - XINT (limit) + 1)
88 return make_number (remainder);
89 val = get_random ();
90 }
91 return make_number (val);
92 }
93 \f
94 /* Heuristic on how many iterations of a tight loop can be safely done
95 before it's time to do a QUIT. This must be a power of 2. */
96 enum { QUIT_COUNT_HEURISTIC = 1 << 16 };
97
98 /* Random data-structure functions. */
99
100 static void
101 CHECK_LIST_END (Lisp_Object x, Lisp_Object y)
102 {
103 CHECK_TYPE (NILP (x), Qlistp, y);
104 }
105
106 DEFUN ("length", Flength, Slength, 1, 1, 0,
107 doc: /* Return the length of vector, list or string SEQUENCE.
108 A byte-code function object is also allowed.
109 If the string contains multibyte characters, this is not necessarily
110 the number of bytes in the string; it is the number of characters.
111 To get the number of bytes, use `string-bytes'. */)
112 (register Lisp_Object sequence)
113 {
114 register Lisp_Object val;
115
116 if (STRINGP (sequence))
117 XSETFASTINT (val, SCHARS (sequence));
118 else if (VECTORP (sequence))
119 XSETFASTINT (val, ASIZE (sequence));
120 else if (CHAR_TABLE_P (sequence))
121 XSETFASTINT (val, MAX_CHAR);
122 else if (BOOL_VECTOR_P (sequence))
123 XSETFASTINT (val, bool_vector_size (sequence));
124 else if (COMPILEDP (sequence))
125 XSETFASTINT (val, ASIZE (sequence) & PSEUDOVECTOR_SIZE_MASK);
126 else if (CONSP (sequence))
127 {
128 EMACS_INT i = 0;
129
130 do
131 {
132 ++i;
133 if ((i & (QUIT_COUNT_HEURISTIC - 1)) == 0)
134 {
135 if (MOST_POSITIVE_FIXNUM < i)
136 error ("List too long");
137 QUIT;
138 }
139 sequence = XCDR (sequence);
140 }
141 while (CONSP (sequence));
142
143 CHECK_LIST_END (sequence, sequence);
144
145 val = make_number (i);
146 }
147 else if (NILP (sequence))
148 XSETFASTINT (val, 0);
149 else
150 wrong_type_argument (Qsequencep, sequence);
151
152 return val;
153 }
154
155 DEFUN ("safe-length", Fsafe_length, Ssafe_length, 1, 1, 0,
156 doc: /* Return the length of a list, but avoid error or infinite loop.
157 This function never gets an error. If LIST is not really a list,
158 it returns 0. If LIST is circular, it returns a finite value
159 which is at least the number of distinct elements. */)
160 (Lisp_Object list)
161 {
162 Lisp_Object tail, halftail;
163 double hilen = 0;
164 uintmax_t lolen = 1;
165
166 if (! CONSP (list))
167 return make_number (0);
168
169 /* halftail is used to detect circular lists. */
170 for (tail = halftail = list; ; )
171 {
172 tail = XCDR (tail);
173 if (! CONSP (tail))
174 break;
175 if (EQ (tail, halftail))
176 break;
177 lolen++;
178 if ((lolen & 1) == 0)
179 {
180 halftail = XCDR (halftail);
181 if ((lolen & (QUIT_COUNT_HEURISTIC - 1)) == 0)
182 {
183 QUIT;
184 if (lolen == 0)
185 hilen += UINTMAX_MAX + 1.0;
186 }
187 }
188 }
189
190 /* If the length does not fit into a fixnum, return a float.
191 On all known practical machines this returns an upper bound on
192 the true length. */
193 return hilen ? make_float (hilen + lolen) : make_fixnum_or_float (lolen);
194 }
195
196 DEFUN ("string-bytes", Fstring_bytes, Sstring_bytes, 1, 1, 0,
197 doc: /* Return the number of bytes in STRING.
198 If STRING is multibyte, this may be greater than the length of STRING. */)
199 (Lisp_Object string)
200 {
201 CHECK_STRING (string);
202 return make_number (SBYTES (string));
203 }
204
205 DEFUN ("string-equal", Fstring_equal, Sstring_equal, 2, 2, 0,
206 doc: /* Return t if two strings have identical contents.
207 Case is significant, but text properties are ignored.
208 Symbols are also allowed; their print names are used instead. */)
209 (register Lisp_Object s1, Lisp_Object s2)
210 {
211 if (SYMBOLP (s1))
212 s1 = SYMBOL_NAME (s1);
213 if (SYMBOLP (s2))
214 s2 = SYMBOL_NAME (s2);
215 CHECK_STRING (s1);
216 CHECK_STRING (s2);
217
218 if (SCHARS (s1) != SCHARS (s2)
219 || SBYTES (s1) != SBYTES (s2)
220 || memcmp (SDATA (s1), SDATA (s2), SBYTES (s1)))
221 return Qnil;
222 return Qt;
223 }
224
225 DEFUN ("compare-strings", Fcompare_strings, Scompare_strings, 6, 7, 0,
226 doc: /* Compare the contents of two strings, converting to multibyte if needed.
227 The arguments START1, END1, START2, and END2, if non-nil, are
228 positions specifying which parts of STR1 or STR2 to compare. In
229 string STR1, compare the part between START1 (inclusive) and END1
230 \(exclusive). If START1 is nil, it defaults to 0, the beginning of
231 the string; if END1 is nil, it defaults to the length of the string.
232 Likewise, in string STR2, compare the part between START2 and END2.
233 Like in `substring', negative values are counted from the end.
234
235 The strings are compared by the numeric values of their characters.
236 For instance, STR1 is "less than" STR2 if its first differing
237 character has a smaller numeric value. If IGNORE-CASE is non-nil,
238 characters are converted to lower-case before comparing them. Unibyte
239 strings are converted to multibyte for comparison.
240
241 The value is t if the strings (or specified portions) match.
242 If string STR1 is less, the value is a negative number N;
243 - 1 - N is the number of characters that match at the beginning.
244 If string STR1 is greater, the value is a positive number N;
245 N - 1 is the number of characters that match at the beginning. */)
246 (Lisp_Object str1, Lisp_Object start1, Lisp_Object end1, Lisp_Object str2,
247 Lisp_Object start2, Lisp_Object end2, Lisp_Object ignore_case)
248 {
249 ptrdiff_t from1, to1, from2, to2, i1, i1_byte, i2, i2_byte;
250
251 CHECK_STRING (str1);
252 CHECK_STRING (str2);
253
254 validate_subarray (str1, start1, end1, SCHARS (str1), &from1, &to1);
255 validate_subarray (str2, start2, end2, SCHARS (str2), &from2, &to2);
256
257 i1 = from1;
258 i2 = from2;
259
260 i1_byte = string_char_to_byte (str1, i1);
261 i2_byte = string_char_to_byte (str2, i2);
262
263 while (i1 < to1 && i2 < to2)
264 {
265 /* When we find a mismatch, we must compare the
266 characters, not just the bytes. */
267 int c1, c2;
268
269 FETCH_STRING_CHAR_AS_MULTIBYTE_ADVANCE (c1, str1, i1, i1_byte);
270 FETCH_STRING_CHAR_AS_MULTIBYTE_ADVANCE (c2, str2, i2, i2_byte);
271
272 if (c1 == c2)
273 continue;
274
275 if (! NILP (ignore_case))
276 {
277 c1 = XINT (Fupcase (make_number (c1)));
278 c2 = XINT (Fupcase (make_number (c2)));
279 }
280
281 if (c1 == c2)
282 continue;
283
284 /* Note that I1 has already been incremented
285 past the character that we are comparing;
286 hence we don't add or subtract 1 here. */
287 if (c1 < c2)
288 return make_number (- i1 + from1);
289 else
290 return make_number (i1 - from1);
291 }
292
293 if (i1 < to1)
294 return make_number (i1 - from1 + 1);
295 if (i2 < to2)
296 return make_number (- i1 + from1 - 1);
297
298 return Qt;
299 }
300
301 DEFUN ("string-lessp", Fstring_lessp, Sstring_lessp, 2, 2, 0,
302 doc: /* Return t if first arg string is less than second in lexicographic order.
303 Case is significant.
304 Symbols are also allowed; their print names are used instead. */)
305 (register Lisp_Object s1, Lisp_Object s2)
306 {
307 register ptrdiff_t end;
308 register ptrdiff_t i1, i1_byte, i2, i2_byte;
309
310 if (SYMBOLP (s1))
311 s1 = SYMBOL_NAME (s1);
312 if (SYMBOLP (s2))
313 s2 = SYMBOL_NAME (s2);
314 CHECK_STRING (s1);
315 CHECK_STRING (s2);
316
317 i1 = i1_byte = i2 = i2_byte = 0;
318
319 end = SCHARS (s1);
320 if (end > SCHARS (s2))
321 end = SCHARS (s2);
322
323 while (i1 < end)
324 {
325 /* When we find a mismatch, we must compare the
326 characters, not just the bytes. */
327 int c1, c2;
328
329 FETCH_STRING_CHAR_ADVANCE (c1, s1, i1, i1_byte);
330 FETCH_STRING_CHAR_ADVANCE (c2, s2, i2, i2_byte);
331
332 if (c1 != c2)
333 return c1 < c2 ? Qt : Qnil;
334 }
335 return i1 < SCHARS (s2) ? Qt : Qnil;
336 }
337 \f
338 enum concat_target_type
339 {
340 concat_cons,
341 concat_string,
342 concat_vector
343 };
344
345 static Lisp_Object concat (ptrdiff_t nargs, Lisp_Object *args,
346 enum concat_target_type target_type, bool last_special);
347
348 /* ARGSUSED */
349 Lisp_Object
350 concat2 (Lisp_Object s1, Lisp_Object s2)
351 {
352 Lisp_Object args[2];
353 args[0] = s1;
354 args[1] = s2;
355 return concat (2, args, concat_string, 0);
356 }
357
358 /* ARGSUSED */
359 Lisp_Object
360 concat3 (Lisp_Object s1, Lisp_Object s2, Lisp_Object s3)
361 {
362 Lisp_Object args[3];
363 args[0] = s1;
364 args[1] = s2;
365 args[2] = s3;
366 return concat (3, args, concat_string, 0);
367 }
368
369 DEFUN ("append", Fappend, Sappend, 0, MANY, 0,
370 doc: /* Concatenate all the arguments and make the result a list.
371 The result is a list whose elements are the elements of all the arguments.
372 Each argument may be a list, vector or string.
373 The last argument is not copied, just used as the tail of the new list.
374 usage: (append &rest SEQUENCES) */)
375 (ptrdiff_t nargs, Lisp_Object *args)
376 {
377 return concat (nargs, args, concat_cons, 1);
378 }
379
380 DEFUN ("concat", Fconcat, Sconcat, 0, MANY, 0,
381 doc: /* Concatenate all the arguments and make the result a string.
382 The result is a string whose elements are the elements of all the arguments.
383 Each argument may be a string or a list or vector of characters (integers).
384 usage: (concat &rest SEQUENCES) */)
385 (ptrdiff_t nargs, Lisp_Object *args)
386 {
387 return concat (nargs, args, concat_string, 0);
388 }
389
390 DEFUN ("vconcat", Fvconcat, Svconcat, 0, MANY, 0,
391 doc: /* Concatenate all the arguments and make the result a vector.
392 The result is a vector whose elements are the elements of all the arguments.
393 Each argument may be a list, vector or string.
394 usage: (vconcat &rest SEQUENCES) */)
395 (ptrdiff_t nargs, Lisp_Object *args)
396 {
397 return concat (nargs, args, concat_vector, 0);
398 }
399
400
401 DEFUN ("copy-sequence", Fcopy_sequence, Scopy_sequence, 1, 1, 0,
402 doc: /* Return a copy of a list, vector, string or char-table.
403 The elements of a list or vector are not copied; they are shared
404 with the original. */)
405 (Lisp_Object arg)
406 {
407 if (NILP (arg)) return arg;
408
409 if (CHAR_TABLE_P (arg))
410 {
411 return copy_char_table (arg);
412 }
413
414 if (BOOL_VECTOR_P (arg))
415 {
416 EMACS_INT nbits = bool_vector_size (arg);
417 ptrdiff_t nbytes = bool_vector_bytes (nbits);
418 Lisp_Object val = make_uninit_bool_vector (nbits);
419 memcpy (bool_vector_data (val), bool_vector_data (arg), nbytes);
420 return val;
421 }
422
423 if (CONSP (arg))
424 return concat (1, &arg, concat_cons, 0);
425 else if (STRINGP (arg))
426 return concat (1, &arg, concat_string, 0);
427 else if (VECTORP (arg))
428 return concat (1, &arg, concat_vector, 0);
429 else
430 wrong_type_argument (Qsequencep, arg);
431 }
432
433 /* This structure holds information of an argument of `concat' that is
434 a string and has text properties to be copied. */
435 struct textprop_rec
436 {
437 ptrdiff_t argnum; /* refer to ARGS (arguments of `concat') */
438 ptrdiff_t from; /* refer to ARGS[argnum] (argument string) */
439 ptrdiff_t to; /* refer to VAL (the target string) */
440 };
441
442 static Lisp_Object
443 concat (ptrdiff_t nargs, Lisp_Object *args,
444 enum concat_target_type target_type, bool last_special)
445 {
446 Lisp_Object val;
447 Lisp_Object tail;
448 Lisp_Object this;
449 ptrdiff_t toindex;
450 ptrdiff_t toindex_byte = 0;
451 EMACS_INT result_len;
452 EMACS_INT result_len_byte;
453 ptrdiff_t argnum;
454 Lisp_Object last_tail;
455 Lisp_Object prev;
456 bool some_multibyte;
457 /* When we make a multibyte string, we can't copy text properties
458 while concatenating each string because the length of resulting
459 string can't be decided until we finish the whole concatenation.
460 So, we record strings that have text properties to be copied
461 here, and copy the text properties after the concatenation. */
462 struct textprop_rec *textprops = NULL;
463 /* Number of elements in textprops. */
464 ptrdiff_t num_textprops = 0;
465 USE_SAFE_ALLOCA;
466
467 tail = Qnil;
468
469 /* In append, the last arg isn't treated like the others */
470 if (last_special && nargs > 0)
471 {
472 nargs--;
473 last_tail = args[nargs];
474 }
475 else
476 last_tail = Qnil;
477
478 /* Check each argument. */
479 for (argnum = 0; argnum < nargs; argnum++)
480 {
481 this = args[argnum];
482 if (!(CONSP (this) || NILP (this) || VECTORP (this) || STRINGP (this)
483 || COMPILEDP (this) || BOOL_VECTOR_P (this)))
484 wrong_type_argument (Qsequencep, this);
485 }
486
487 /* Compute total length in chars of arguments in RESULT_LEN.
488 If desired output is a string, also compute length in bytes
489 in RESULT_LEN_BYTE, and determine in SOME_MULTIBYTE
490 whether the result should be a multibyte string. */
491 result_len_byte = 0;
492 result_len = 0;
493 some_multibyte = 0;
494 for (argnum = 0; argnum < nargs; argnum++)
495 {
496 EMACS_INT len;
497 this = args[argnum];
498 len = XFASTINT (Flength (this));
499 if (target_type == concat_string)
500 {
501 /* We must count the number of bytes needed in the string
502 as well as the number of characters. */
503 ptrdiff_t i;
504 Lisp_Object ch;
505 int c;
506 ptrdiff_t this_len_byte;
507
508 if (VECTORP (this) || COMPILEDP (this))
509 for (i = 0; i < len; i++)
510 {
511 ch = AREF (this, i);
512 CHECK_CHARACTER (ch);
513 c = XFASTINT (ch);
514 this_len_byte = CHAR_BYTES (c);
515 if (STRING_BYTES_BOUND - result_len_byte < this_len_byte)
516 string_overflow ();
517 result_len_byte += this_len_byte;
518 if (! ASCII_CHAR_P (c) && ! CHAR_BYTE8_P (c))
519 some_multibyte = 1;
520 }
521 else if (BOOL_VECTOR_P (this) && bool_vector_size (this) > 0)
522 wrong_type_argument (Qintegerp, Faref (this, make_number (0)));
523 else if (CONSP (this))
524 for (; CONSP (this); this = XCDR (this))
525 {
526 ch = XCAR (this);
527 CHECK_CHARACTER (ch);
528 c = XFASTINT (ch);
529 this_len_byte = CHAR_BYTES (c);
530 if (STRING_BYTES_BOUND - result_len_byte < this_len_byte)
531 string_overflow ();
532 result_len_byte += this_len_byte;
533 if (! ASCII_CHAR_P (c) && ! CHAR_BYTE8_P (c))
534 some_multibyte = 1;
535 }
536 else if (STRINGP (this))
537 {
538 if (STRING_MULTIBYTE (this))
539 {
540 some_multibyte = 1;
541 this_len_byte = SBYTES (this);
542 }
543 else
544 this_len_byte = count_size_as_multibyte (SDATA (this),
545 SCHARS (this));
546 if (STRING_BYTES_BOUND - result_len_byte < this_len_byte)
547 string_overflow ();
548 result_len_byte += this_len_byte;
549 }
550 }
551
552 result_len += len;
553 if (MOST_POSITIVE_FIXNUM < result_len)
554 memory_full (SIZE_MAX);
555 }
556
557 if (! some_multibyte)
558 result_len_byte = result_len;
559
560 /* Create the output object. */
561 if (target_type == concat_cons)
562 val = Fmake_list (make_number (result_len), Qnil);
563 else if (target_type == concat_vector)
564 val = Fmake_vector (make_number (result_len), Qnil);
565 else if (some_multibyte)
566 val = make_uninit_multibyte_string (result_len, result_len_byte);
567 else
568 val = make_uninit_string (result_len);
569
570 /* In `append', if all but last arg are nil, return last arg. */
571 if (target_type == concat_cons && EQ (val, Qnil))
572 return last_tail;
573
574 /* Copy the contents of the args into the result. */
575 if (CONSP (val))
576 tail = val, toindex = -1; /* -1 in toindex is flag we are making a list */
577 else
578 toindex = 0, toindex_byte = 0;
579
580 prev = Qnil;
581 if (STRINGP (val))
582 SAFE_NALLOCA (textprops, 1, nargs);
583
584 for (argnum = 0; argnum < nargs; argnum++)
585 {
586 Lisp_Object thislen;
587 ptrdiff_t thisleni = 0;
588 register ptrdiff_t thisindex = 0;
589 register ptrdiff_t thisindex_byte = 0;
590
591 this = args[argnum];
592 if (!CONSP (this))
593 thislen = Flength (this), thisleni = XINT (thislen);
594
595 /* Between strings of the same kind, copy fast. */
596 if (STRINGP (this) && STRINGP (val)
597 && STRING_MULTIBYTE (this) == some_multibyte)
598 {
599 ptrdiff_t thislen_byte = SBYTES (this);
600
601 memcpy (SDATA (val) + toindex_byte, SDATA (this), SBYTES (this));
602 if (string_intervals (this))
603 {
604 textprops[num_textprops].argnum = argnum;
605 textprops[num_textprops].from = 0;
606 textprops[num_textprops++].to = toindex;
607 }
608 toindex_byte += thislen_byte;
609 toindex += thisleni;
610 }
611 /* Copy a single-byte string to a multibyte string. */
612 else if (STRINGP (this) && STRINGP (val))
613 {
614 if (string_intervals (this))
615 {
616 textprops[num_textprops].argnum = argnum;
617 textprops[num_textprops].from = 0;
618 textprops[num_textprops++].to = toindex;
619 }
620 toindex_byte += copy_text (SDATA (this),
621 SDATA (val) + toindex_byte,
622 SCHARS (this), 0, 1);
623 toindex += thisleni;
624 }
625 else
626 /* Copy element by element. */
627 while (1)
628 {
629 register Lisp_Object elt;
630
631 /* Fetch next element of `this' arg into `elt', or break if
632 `this' is exhausted. */
633 if (NILP (this)) break;
634 if (CONSP (this))
635 elt = XCAR (this), this = XCDR (this);
636 else if (thisindex >= thisleni)
637 break;
638 else if (STRINGP (this))
639 {
640 int c;
641 if (STRING_MULTIBYTE (this))
642 FETCH_STRING_CHAR_ADVANCE_NO_CHECK (c, this,
643 thisindex,
644 thisindex_byte);
645 else
646 {
647 c = SREF (this, thisindex); thisindex++;
648 if (some_multibyte && !ASCII_CHAR_P (c))
649 c = BYTE8_TO_CHAR (c);
650 }
651 XSETFASTINT (elt, c);
652 }
653 else if (BOOL_VECTOR_P (this))
654 {
655 elt = bool_vector_ref (this, thisindex);
656 thisindex++;
657 }
658 else
659 {
660 elt = AREF (this, thisindex);
661 thisindex++;
662 }
663
664 /* Store this element into the result. */
665 if (toindex < 0)
666 {
667 XSETCAR (tail, elt);
668 prev = tail;
669 tail = XCDR (tail);
670 }
671 else if (VECTORP (val))
672 {
673 ASET (val, toindex, elt);
674 toindex++;
675 }
676 else
677 {
678 int c;
679 CHECK_CHARACTER (elt);
680 c = XFASTINT (elt);
681 if (some_multibyte)
682 toindex_byte += CHAR_STRING (c, SDATA (val) + toindex_byte);
683 else
684 SSET (val, toindex_byte++, c);
685 toindex++;
686 }
687 }
688 }
689 if (!NILP (prev))
690 XSETCDR (prev, last_tail);
691
692 if (num_textprops > 0)
693 {
694 Lisp_Object props;
695 ptrdiff_t last_to_end = -1;
696
697 for (argnum = 0; argnum < num_textprops; argnum++)
698 {
699 this = args[textprops[argnum].argnum];
700 props = text_property_list (this,
701 make_number (0),
702 make_number (SCHARS (this)),
703 Qnil);
704 /* If successive arguments have properties, be sure that the
705 value of `composition' property be the copy. */
706 if (last_to_end == textprops[argnum].to)
707 make_composition_value_copy (props);
708 add_text_properties_from_list (val, props,
709 make_number (textprops[argnum].to));
710 last_to_end = textprops[argnum].to + SCHARS (this);
711 }
712 }
713
714 SAFE_FREE ();
715 return val;
716 }
717 \f
718 static Lisp_Object string_char_byte_cache_string;
719 static ptrdiff_t string_char_byte_cache_charpos;
720 static ptrdiff_t string_char_byte_cache_bytepos;
721
722 void
723 clear_string_char_byte_cache (void)
724 {
725 string_char_byte_cache_string = Qnil;
726 }
727
728 /* Return the byte index corresponding to CHAR_INDEX in STRING. */
729
730 ptrdiff_t
731 string_char_to_byte (Lisp_Object string, ptrdiff_t char_index)
732 {
733 ptrdiff_t i_byte;
734 ptrdiff_t best_below, best_below_byte;
735 ptrdiff_t best_above, best_above_byte;
736
737 best_below = best_below_byte = 0;
738 best_above = SCHARS (string);
739 best_above_byte = SBYTES (string);
740 if (best_above == best_above_byte)
741 return char_index;
742
743 if (EQ (string, string_char_byte_cache_string))
744 {
745 if (string_char_byte_cache_charpos < char_index)
746 {
747 best_below = string_char_byte_cache_charpos;
748 best_below_byte = string_char_byte_cache_bytepos;
749 }
750 else
751 {
752 best_above = string_char_byte_cache_charpos;
753 best_above_byte = string_char_byte_cache_bytepos;
754 }
755 }
756
757 if (char_index - best_below < best_above - char_index)
758 {
759 unsigned char *p = SDATA (string) + best_below_byte;
760
761 while (best_below < char_index)
762 {
763 p += BYTES_BY_CHAR_HEAD (*p);
764 best_below++;
765 }
766 i_byte = p - SDATA (string);
767 }
768 else
769 {
770 unsigned char *p = SDATA (string) + best_above_byte;
771
772 while (best_above > char_index)
773 {
774 p--;
775 while (!CHAR_HEAD_P (*p)) p--;
776 best_above--;
777 }
778 i_byte = p - SDATA (string);
779 }
780
781 string_char_byte_cache_bytepos = i_byte;
782 string_char_byte_cache_charpos = char_index;
783 string_char_byte_cache_string = string;
784
785 return i_byte;
786 }
787 \f
788 /* Return the character index corresponding to BYTE_INDEX in STRING. */
789
790 ptrdiff_t
791 string_byte_to_char (Lisp_Object string, ptrdiff_t byte_index)
792 {
793 ptrdiff_t i, i_byte;
794 ptrdiff_t best_below, best_below_byte;
795 ptrdiff_t best_above, best_above_byte;
796
797 best_below = best_below_byte = 0;
798 best_above = SCHARS (string);
799 best_above_byte = SBYTES (string);
800 if (best_above == best_above_byte)
801 return byte_index;
802
803 if (EQ (string, string_char_byte_cache_string))
804 {
805 if (string_char_byte_cache_bytepos < byte_index)
806 {
807 best_below = string_char_byte_cache_charpos;
808 best_below_byte = string_char_byte_cache_bytepos;
809 }
810 else
811 {
812 best_above = string_char_byte_cache_charpos;
813 best_above_byte = string_char_byte_cache_bytepos;
814 }
815 }
816
817 if (byte_index - best_below_byte < best_above_byte - byte_index)
818 {
819 unsigned char *p = SDATA (string) + best_below_byte;
820 unsigned char *pend = SDATA (string) + byte_index;
821
822 while (p < pend)
823 {
824 p += BYTES_BY_CHAR_HEAD (*p);
825 best_below++;
826 }
827 i = best_below;
828 i_byte = p - SDATA (string);
829 }
830 else
831 {
832 unsigned char *p = SDATA (string) + best_above_byte;
833 unsigned char *pbeg = SDATA (string) + byte_index;
834
835 while (p > pbeg)
836 {
837 p--;
838 while (!CHAR_HEAD_P (*p)) p--;
839 best_above--;
840 }
841 i = best_above;
842 i_byte = p - SDATA (string);
843 }
844
845 string_char_byte_cache_bytepos = i_byte;
846 string_char_byte_cache_charpos = i;
847 string_char_byte_cache_string = string;
848
849 return i;
850 }
851 \f
852 /* Convert STRING to a multibyte string. */
853
854 static Lisp_Object
855 string_make_multibyte (Lisp_Object string)
856 {
857 unsigned char *buf;
858 ptrdiff_t nbytes;
859 Lisp_Object ret;
860 USE_SAFE_ALLOCA;
861
862 if (STRING_MULTIBYTE (string))
863 return string;
864
865 nbytes = count_size_as_multibyte (SDATA (string),
866 SCHARS (string));
867 /* If all the chars are ASCII, they won't need any more bytes
868 once converted. In that case, we can return STRING itself. */
869 if (nbytes == SBYTES (string))
870 return string;
871
872 buf = SAFE_ALLOCA (nbytes);
873 copy_text (SDATA (string), buf, SBYTES (string),
874 0, 1);
875
876 ret = make_multibyte_string ((char *) buf, SCHARS (string), nbytes);
877 SAFE_FREE ();
878
879 return ret;
880 }
881
882
883 /* Convert STRING (if unibyte) to a multibyte string without changing
884 the number of characters. Characters 0200 trough 0237 are
885 converted to eight-bit characters. */
886
887 Lisp_Object
888 string_to_multibyte (Lisp_Object string)
889 {
890 unsigned char *buf;
891 ptrdiff_t nbytes;
892 Lisp_Object ret;
893 USE_SAFE_ALLOCA;
894
895 if (STRING_MULTIBYTE (string))
896 return string;
897
898 nbytes = count_size_as_multibyte (SDATA (string), SBYTES (string));
899 /* If all the chars are ASCII, they won't need any more bytes once
900 converted. */
901 if (nbytes == SBYTES (string))
902 return make_multibyte_string (SSDATA (string), nbytes, nbytes);
903
904 buf = SAFE_ALLOCA (nbytes);
905 memcpy (buf, SDATA (string), SBYTES (string));
906 str_to_multibyte (buf, nbytes, SBYTES (string));
907
908 ret = make_multibyte_string ((char *) buf, SCHARS (string), nbytes);
909 SAFE_FREE ();
910
911 return ret;
912 }
913
914
915 /* Convert STRING to a single-byte string. */
916
917 Lisp_Object
918 string_make_unibyte (Lisp_Object string)
919 {
920 ptrdiff_t nchars;
921 unsigned char *buf;
922 Lisp_Object ret;
923 USE_SAFE_ALLOCA;
924
925 if (! STRING_MULTIBYTE (string))
926 return string;
927
928 nchars = SCHARS (string);
929
930 buf = SAFE_ALLOCA (nchars);
931 copy_text (SDATA (string), buf, SBYTES (string),
932 1, 0);
933
934 ret = make_unibyte_string ((char *) buf, nchars);
935 SAFE_FREE ();
936
937 return ret;
938 }
939
940 DEFUN ("string-make-multibyte", Fstring_make_multibyte, Sstring_make_multibyte,
941 1, 1, 0,
942 doc: /* Return the multibyte equivalent of STRING.
943 If STRING is unibyte and contains non-ASCII characters, the function
944 `unibyte-char-to-multibyte' is used to convert each unibyte character
945 to a multibyte character. In this case, the returned string is a
946 newly created string with no text properties. If STRING is multibyte
947 or entirely ASCII, it is returned unchanged. In particular, when
948 STRING is unibyte and entirely ASCII, the returned string is unibyte.
949 \(When the characters are all ASCII, Emacs primitives will treat the
950 string the same way whether it is unibyte or multibyte.) */)
951 (Lisp_Object string)
952 {
953 CHECK_STRING (string);
954
955 return string_make_multibyte (string);
956 }
957
958 DEFUN ("string-make-unibyte", Fstring_make_unibyte, Sstring_make_unibyte,
959 1, 1, 0,
960 doc: /* Return the unibyte equivalent of STRING.
961 Multibyte character codes are converted to unibyte according to
962 `nonascii-translation-table' or, if that is nil, `nonascii-insert-offset'.
963 If the lookup in the translation table fails, this function takes just
964 the low 8 bits of each character. */)
965 (Lisp_Object string)
966 {
967 CHECK_STRING (string);
968
969 return string_make_unibyte (string);
970 }
971
972 DEFUN ("string-as-unibyte", Fstring_as_unibyte, Sstring_as_unibyte,
973 1, 1, 0,
974 doc: /* Return a unibyte string with the same individual bytes as STRING.
975 If STRING is unibyte, the result is STRING itself.
976 Otherwise it is a newly created string, with no text properties.
977 If STRING is multibyte and contains a character of charset
978 `eight-bit', it is converted to the corresponding single byte. */)
979 (Lisp_Object string)
980 {
981 CHECK_STRING (string);
982
983 if (STRING_MULTIBYTE (string))
984 {
985 unsigned char *str = (unsigned char *) xlispstrdup (string);
986 ptrdiff_t bytes = str_as_unibyte (str, SBYTES (string));
987
988 string = make_unibyte_string ((char *) str, bytes);
989 xfree (str);
990 }
991 return string;
992 }
993
994 DEFUN ("string-as-multibyte", Fstring_as_multibyte, Sstring_as_multibyte,
995 1, 1, 0,
996 doc: /* Return a multibyte string with the same individual bytes as STRING.
997 If STRING is multibyte, the result is STRING itself.
998 Otherwise it is a newly created string, with no text properties.
999
1000 If STRING is unibyte and contains an individual 8-bit byte (i.e. not
1001 part of a correct utf-8 sequence), it is converted to the corresponding
1002 multibyte character of charset `eight-bit'.
1003 See also `string-to-multibyte'.
1004
1005 Beware, this often doesn't really do what you think it does.
1006 It is similar to (decode-coding-string STRING 'utf-8-emacs).
1007 If you're not sure, whether to use `string-as-multibyte' or
1008 `string-to-multibyte', use `string-to-multibyte'. */)
1009 (Lisp_Object string)
1010 {
1011 CHECK_STRING (string);
1012
1013 if (! STRING_MULTIBYTE (string))
1014 {
1015 Lisp_Object new_string;
1016 ptrdiff_t nchars, nbytes;
1017
1018 parse_str_as_multibyte (SDATA (string),
1019 SBYTES (string),
1020 &nchars, &nbytes);
1021 new_string = make_uninit_multibyte_string (nchars, nbytes);
1022 memcpy (SDATA (new_string), SDATA (string), SBYTES (string));
1023 if (nbytes != SBYTES (string))
1024 str_as_multibyte (SDATA (new_string), nbytes,
1025 SBYTES (string), NULL);
1026 string = new_string;
1027 set_string_intervals (string, NULL);
1028 }
1029 return string;
1030 }
1031
1032 DEFUN ("string-to-multibyte", Fstring_to_multibyte, Sstring_to_multibyte,
1033 1, 1, 0,
1034 doc: /* Return a multibyte string with the same individual chars as STRING.
1035 If STRING is multibyte, the result is STRING itself.
1036 Otherwise it is a newly created string, with no text properties.
1037
1038 If STRING is unibyte and contains an 8-bit byte, it is converted to
1039 the corresponding multibyte character of charset `eight-bit'.
1040
1041 This differs from `string-as-multibyte' by converting each byte of a correct
1042 utf-8 sequence to an eight-bit character, not just bytes that don't form a
1043 correct sequence. */)
1044 (Lisp_Object string)
1045 {
1046 CHECK_STRING (string);
1047
1048 return string_to_multibyte (string);
1049 }
1050
1051 DEFUN ("string-to-unibyte", Fstring_to_unibyte, Sstring_to_unibyte,
1052 1, 1, 0,
1053 doc: /* Return a unibyte string with the same individual chars as STRING.
1054 If STRING is unibyte, the result is STRING itself.
1055 Otherwise it is a newly created string, with no text properties,
1056 where each `eight-bit' character is converted to the corresponding byte.
1057 If STRING contains a non-ASCII, non-`eight-bit' character,
1058 an error is signaled. */)
1059 (Lisp_Object string)
1060 {
1061 CHECK_STRING (string);
1062
1063 if (STRING_MULTIBYTE (string))
1064 {
1065 ptrdiff_t chars = SCHARS (string);
1066 unsigned char *str = xmalloc_atomic (chars);
1067 ptrdiff_t converted = str_to_unibyte (SDATA (string), str, chars);
1068
1069 if (converted < chars)
1070 error ("Can't convert the %"pD"dth character to unibyte", converted);
1071 string = make_unibyte_string ((char *) str, chars);
1072 xfree (str);
1073 }
1074 return string;
1075 }
1076
1077 \f
1078 DEFUN ("copy-alist", Fcopy_alist, Scopy_alist, 1, 1, 0,
1079 doc: /* Return a copy of ALIST.
1080 This is an alist which represents the same mapping from objects to objects,
1081 but does not share the alist structure with ALIST.
1082 The objects mapped (cars and cdrs of elements of the alist)
1083 are shared, however.
1084 Elements of ALIST that are not conses are also shared. */)
1085 (Lisp_Object alist)
1086 {
1087 register Lisp_Object tem;
1088
1089 CHECK_LIST (alist);
1090 if (NILP (alist))
1091 return alist;
1092 alist = concat (1, &alist, concat_cons, 0);
1093 for (tem = alist; CONSP (tem); tem = XCDR (tem))
1094 {
1095 register Lisp_Object car;
1096 car = XCAR (tem);
1097
1098 if (CONSP (car))
1099 XSETCAR (tem, Fcons (XCAR (car), XCDR (car)));
1100 }
1101 return alist;
1102 }
1103
1104 /* Check that ARRAY can have a valid subarray [FROM..TO),
1105 given that its size is SIZE.
1106 If FROM is nil, use 0; if TO is nil, use SIZE.
1107 Count negative values backwards from the end.
1108 Set *IFROM and *ITO to the two indexes used. */
1109
1110 void
1111 validate_subarray (Lisp_Object array, Lisp_Object from, Lisp_Object to,
1112 ptrdiff_t size, ptrdiff_t *ifrom, ptrdiff_t *ito)
1113 {
1114 EMACS_INT f, t;
1115
1116 if (INTEGERP (from))
1117 {
1118 f = XINT (from);
1119 if (f < 0)
1120 f += size;
1121 }
1122 else if (NILP (from))
1123 f = 0;
1124 else
1125 wrong_type_argument (Qintegerp, from);
1126
1127 if (INTEGERP (to))
1128 {
1129 t = XINT (to);
1130 if (t < 0)
1131 t += size;
1132 }
1133 else if (NILP (to))
1134 t = size;
1135 else
1136 wrong_type_argument (Qintegerp, to);
1137
1138 if (! (0 <= f && f <= t && t <= size))
1139 args_out_of_range_3 (array, from, to);
1140
1141 *ifrom = f;
1142 *ito = t;
1143 }
1144
1145 DEFUN ("substring", Fsubstring, Ssubstring, 1, 3, 0,
1146 doc: /* Return a new string whose contents are a substring of STRING.
1147 The returned string consists of the characters between index FROM
1148 \(inclusive) and index TO (exclusive) of STRING. FROM and TO are
1149 zero-indexed: 0 means the first character of STRING. Negative values
1150 are counted from the end of STRING. If TO is nil, the substring runs
1151 to the end of STRING.
1152
1153 The STRING argument may also be a vector. In that case, the return
1154 value is a new vector that contains the elements between index FROM
1155 \(inclusive) and index TO (exclusive) of that vector argument.
1156
1157 With one argument, just copy STRING (with properties, if any). */)
1158 (Lisp_Object string, Lisp_Object from, Lisp_Object to)
1159 {
1160 Lisp_Object res;
1161 ptrdiff_t size, ifrom, ito;
1162
1163 if (STRINGP (string))
1164 size = SCHARS (string);
1165 else if (VECTORP (string))
1166 size = ASIZE (string);
1167 else
1168 wrong_type_argument (Qarrayp, string);
1169
1170 validate_subarray (string, from, to, size, &ifrom, &ito);
1171
1172 if (STRINGP (string))
1173 {
1174 ptrdiff_t from_byte
1175 = !ifrom ? 0 : string_char_to_byte (string, ifrom);
1176 ptrdiff_t to_byte
1177 = ito == size ? SBYTES (string) : string_char_to_byte (string, ito);
1178 res = make_specified_string (SSDATA (string) + from_byte,
1179 ito - ifrom, to_byte - from_byte,
1180 STRING_MULTIBYTE (string));
1181 copy_text_properties (make_number (ifrom), make_number (ito),
1182 string, make_number (0), res, Qnil);
1183 }
1184 else
1185 res = Fvector (ito - ifrom, aref_addr (string, ifrom));
1186
1187 return res;
1188 }
1189
1190
1191 DEFUN ("substring-no-properties", Fsubstring_no_properties, Ssubstring_no_properties, 1, 3, 0,
1192 doc: /* Return a substring of STRING, without text properties.
1193 It starts at index FROM and ends before TO.
1194 TO may be nil or omitted; then the substring runs to the end of STRING.
1195 If FROM is nil or omitted, the substring starts at the beginning of STRING.
1196 If FROM or TO is negative, it counts from the end.
1197
1198 With one argument, just copy STRING without its properties. */)
1199 (Lisp_Object string, register Lisp_Object from, Lisp_Object to)
1200 {
1201 ptrdiff_t from_char, to_char, from_byte, to_byte, size;
1202
1203 CHECK_STRING (string);
1204
1205 size = SCHARS (string);
1206 validate_subarray (string, from, to, size, &from_char, &to_char);
1207
1208 from_byte = !from_char ? 0 : string_char_to_byte (string, from_char);
1209 to_byte =
1210 to_char == size ? SBYTES (string) : string_char_to_byte (string, to_char);
1211 return make_specified_string (SSDATA (string) + from_byte,
1212 to_char - from_char, to_byte - from_byte,
1213 STRING_MULTIBYTE (string));
1214 }
1215
1216 /* Extract a substring of STRING, giving start and end positions
1217 both in characters and in bytes. */
1218
1219 Lisp_Object
1220 substring_both (Lisp_Object string, ptrdiff_t from, ptrdiff_t from_byte,
1221 ptrdiff_t to, ptrdiff_t to_byte)
1222 {
1223 Lisp_Object res;
1224 ptrdiff_t size;
1225
1226 CHECK_VECTOR_OR_STRING (string);
1227
1228 size = STRINGP (string) ? SCHARS (string) : ASIZE (string);
1229
1230 if (!(0 <= from && from <= to && to <= size))
1231 args_out_of_range_3 (string, make_number (from), make_number (to));
1232
1233 if (STRINGP (string))
1234 {
1235 res = make_specified_string (SSDATA (string) + from_byte,
1236 to - from, to_byte - from_byte,
1237 STRING_MULTIBYTE (string));
1238 copy_text_properties (make_number (from), make_number (to),
1239 string, make_number (0), res, Qnil);
1240 }
1241 else
1242 res = Fvector (to - from, aref_addr (string, from));
1243
1244 return res;
1245 }
1246 \f
1247 DEFUN ("nthcdr", Fnthcdr, Snthcdr, 2, 2, 0,
1248 doc: /* Take cdr N times on LIST, return the result. */)
1249 (Lisp_Object n, Lisp_Object list)
1250 {
1251 EMACS_INT i, num;
1252 CHECK_NUMBER (n);
1253 num = XINT (n);
1254 for (i = 0; i < num && !NILP (list); i++)
1255 {
1256 QUIT;
1257 CHECK_LIST_CONS (list, list);
1258 list = XCDR (list);
1259 }
1260 return list;
1261 }
1262
1263 DEFUN ("nth", Fnth, Snth, 2, 2, 0,
1264 doc: /* Return the Nth element of LIST.
1265 N counts from zero. If LIST is not that long, nil is returned. */)
1266 (Lisp_Object n, Lisp_Object list)
1267 {
1268 return Fcar (Fnthcdr (n, list));
1269 }
1270
1271 DEFUN ("elt", Felt, Selt, 2, 2, 0,
1272 doc: /* Return element of SEQUENCE at index N. */)
1273 (register Lisp_Object sequence, Lisp_Object n)
1274 {
1275 CHECK_NUMBER (n);
1276 if (CONSP (sequence) || NILP (sequence))
1277 return Fcar (Fnthcdr (n, sequence));
1278
1279 /* Faref signals a "not array" error, so check here. */
1280 CHECK_ARRAY (sequence, Qsequencep);
1281 return Faref (sequence, n);
1282 }
1283
1284 DEFUN ("member", Fmember, Smember, 2, 2, 0,
1285 doc: /* Return non-nil if ELT is an element of LIST. Comparison done with `equal'.
1286 The value is actually the tail of LIST whose car is ELT. */)
1287 (register Lisp_Object elt, Lisp_Object list)
1288 {
1289 register Lisp_Object tail;
1290 for (tail = list; CONSP (tail); tail = XCDR (tail))
1291 {
1292 register Lisp_Object tem;
1293 CHECK_LIST_CONS (tail, list);
1294 tem = XCAR (tail);
1295 if (! NILP (Fequal (elt, tem)))
1296 return tail;
1297 QUIT;
1298 }
1299 return Qnil;
1300 }
1301
1302 DEFUN ("memq", Fmemq, Smemq, 2, 2, 0,
1303 doc: /* Return non-nil if ELT is an element of LIST. Comparison done with `eq'.
1304 The value is actually the tail of LIST whose car is ELT. */)
1305 (register Lisp_Object elt, Lisp_Object list)
1306 {
1307 while (1)
1308 {
1309 if (!CONSP (list) || EQ (XCAR (list), elt))
1310 break;
1311
1312 list = XCDR (list);
1313 if (!CONSP (list) || EQ (XCAR (list), elt))
1314 break;
1315
1316 list = XCDR (list);
1317 if (!CONSP (list) || EQ (XCAR (list), elt))
1318 break;
1319
1320 list = XCDR (list);
1321 QUIT;
1322 }
1323
1324 CHECK_LIST (list);
1325 return list;
1326 }
1327
1328 DEFUN ("memql", Fmemql, Smemql, 2, 2, 0,
1329 doc: /* Return non-nil if ELT is an element of LIST. Comparison done with `eql'.
1330 The value is actually the tail of LIST whose car is ELT. */)
1331 (register Lisp_Object elt, Lisp_Object list)
1332 {
1333 register Lisp_Object tail;
1334
1335 for (tail = list; CONSP (tail); tail = XCDR (tail))
1336 {
1337 register Lisp_Object tem;
1338 CHECK_LIST_CONS (tail, list);
1339 tem = XCAR (tail);
1340 if (!NILP (Feql (elt, tem)))
1341 return tail;
1342 QUIT;
1343 }
1344 return Qnil;
1345 }
1346
1347 DEFUN ("assq", Fassq, Sassq, 2, 2, 0,
1348 doc: /* Return non-nil if KEY is `eq' to the car of an element of LIST.
1349 The value is actually the first element of LIST whose car is KEY.
1350 Elements of LIST that are not conses are ignored. */)
1351 (Lisp_Object key, Lisp_Object list)
1352 {
1353 while (1)
1354 {
1355 if (!CONSP (list)
1356 || (CONSP (XCAR (list))
1357 && EQ (XCAR (XCAR (list)), key)))
1358 break;
1359
1360 list = XCDR (list);
1361 if (!CONSP (list)
1362 || (CONSP (XCAR (list))
1363 && EQ (XCAR (XCAR (list)), key)))
1364 break;
1365
1366 list = XCDR (list);
1367 if (!CONSP (list)
1368 || (CONSP (XCAR (list))
1369 && EQ (XCAR (XCAR (list)), key)))
1370 break;
1371
1372 list = XCDR (list);
1373 QUIT;
1374 }
1375
1376 return CAR (list);
1377 }
1378
1379 /* Like Fassq but never report an error and do not allow quits.
1380 Use only on lists known never to be circular. */
1381
1382 Lisp_Object
1383 assq_no_quit (Lisp_Object key, Lisp_Object list)
1384 {
1385 while (CONSP (list)
1386 && (!CONSP (XCAR (list))
1387 || !EQ (XCAR (XCAR (list)), key)))
1388 list = XCDR (list);
1389
1390 return CAR_SAFE (list);
1391 }
1392
1393 DEFUN ("assoc", Fassoc, Sassoc, 2, 2, 0,
1394 doc: /* Return non-nil if KEY is `equal' to the car of an element of LIST.
1395 The value is actually the first element of LIST whose car equals KEY. */)
1396 (Lisp_Object key, Lisp_Object list)
1397 {
1398 Lisp_Object car;
1399
1400 while (1)
1401 {
1402 if (!CONSP (list)
1403 || (CONSP (XCAR (list))
1404 && (car = XCAR (XCAR (list)),
1405 EQ (car, key) || !NILP (Fequal (car, key)))))
1406 break;
1407
1408 list = XCDR (list);
1409 if (!CONSP (list)
1410 || (CONSP (XCAR (list))
1411 && (car = XCAR (XCAR (list)),
1412 EQ (car, key) || !NILP (Fequal (car, key)))))
1413 break;
1414
1415 list = XCDR (list);
1416 if (!CONSP (list)
1417 || (CONSP (XCAR (list))
1418 && (car = XCAR (XCAR (list)),
1419 EQ (car, key) || !NILP (Fequal (car, key)))))
1420 break;
1421
1422 list = XCDR (list);
1423 QUIT;
1424 }
1425
1426 return CAR (list);
1427 }
1428
1429 /* Like Fassoc but never report an error and do not allow quits.
1430 Use only on lists known never to be circular. */
1431
1432 Lisp_Object
1433 assoc_no_quit (Lisp_Object key, Lisp_Object list)
1434 {
1435 while (CONSP (list)
1436 && (!CONSP (XCAR (list))
1437 || (!EQ (XCAR (XCAR (list)), key)
1438 && NILP (Fequal (XCAR (XCAR (list)), key)))))
1439 list = XCDR (list);
1440
1441 return CONSP (list) ? XCAR (list) : Qnil;
1442 }
1443
1444 DEFUN ("rassq", Frassq, Srassq, 2, 2, 0,
1445 doc: /* Return non-nil if KEY is `eq' to the cdr of an element of LIST.
1446 The value is actually the first element of LIST whose cdr is KEY. */)
1447 (register Lisp_Object key, Lisp_Object list)
1448 {
1449 while (1)
1450 {
1451 if (!CONSP (list)
1452 || (CONSP (XCAR (list))
1453 && EQ (XCDR (XCAR (list)), key)))
1454 break;
1455
1456 list = XCDR (list);
1457 if (!CONSP (list)
1458 || (CONSP (XCAR (list))
1459 && EQ (XCDR (XCAR (list)), key)))
1460 break;
1461
1462 list = XCDR (list);
1463 if (!CONSP (list)
1464 || (CONSP (XCAR (list))
1465 && EQ (XCDR (XCAR (list)), key)))
1466 break;
1467
1468 list = XCDR (list);
1469 QUIT;
1470 }
1471
1472 return CAR (list);
1473 }
1474
1475 DEFUN ("rassoc", Frassoc, Srassoc, 2, 2, 0,
1476 doc: /* Return non-nil if KEY is `equal' to the cdr of an element of LIST.
1477 The value is actually the first element of LIST whose cdr equals KEY. */)
1478 (Lisp_Object key, Lisp_Object list)
1479 {
1480 Lisp_Object cdr;
1481
1482 while (1)
1483 {
1484 if (!CONSP (list)
1485 || (CONSP (XCAR (list))
1486 && (cdr = XCDR (XCAR (list)),
1487 EQ (cdr, key) || !NILP (Fequal (cdr, key)))))
1488 break;
1489
1490 list = XCDR (list);
1491 if (!CONSP (list)
1492 || (CONSP (XCAR (list))
1493 && (cdr = XCDR (XCAR (list)),
1494 EQ (cdr, key) || !NILP (Fequal (cdr, key)))))
1495 break;
1496
1497 list = XCDR (list);
1498 if (!CONSP (list)
1499 || (CONSP (XCAR (list))
1500 && (cdr = XCDR (XCAR (list)),
1501 EQ (cdr, key) || !NILP (Fequal (cdr, key)))))
1502 break;
1503
1504 list = XCDR (list);
1505 QUIT;
1506 }
1507
1508 return CAR (list);
1509 }
1510 \f
1511 DEFUN ("delq", Fdelq, Sdelq, 2, 2, 0,
1512 doc: /* Delete members of LIST which are `eq' to ELT, and return the result.
1513 More precisely, this function skips any members `eq' to ELT at the
1514 front of LIST, then removes members `eq' to ELT from the remaining
1515 sublist by modifying its list structure, then returns the resulting
1516 list.
1517
1518 Write `(setq foo (delq element foo))' to be sure of correctly changing
1519 the value of a list `foo'. */)
1520 (register Lisp_Object elt, Lisp_Object list)
1521 {
1522 Lisp_Object tail, tortoise, prev = Qnil;
1523 bool skip;
1524
1525 FOR_EACH_TAIL (tail, list, tortoise, skip)
1526 {
1527 Lisp_Object tem = XCAR (tail);
1528 if (EQ (elt, tem))
1529 {
1530 if (NILP (prev))
1531 list = XCDR (tail);
1532 else
1533 Fsetcdr (prev, XCDR (tail));
1534 }
1535 else
1536 prev = tail;
1537 }
1538 return list;
1539 }
1540
1541 DEFUN ("delete", Fdelete, Sdelete, 2, 2, 0,
1542 doc: /* Delete members of SEQ which are `equal' to ELT, and return the result.
1543 SEQ must be a sequence (i.e. a list, a vector, or a string).
1544 The return value is a sequence of the same type.
1545
1546 If SEQ is a list, this behaves like `delq', except that it compares
1547 with `equal' instead of `eq'. In particular, it may remove elements
1548 by altering the list structure.
1549
1550 If SEQ is not a list, deletion is never performed destructively;
1551 instead this function creates and returns a new vector or string.
1552
1553 Write `(setq foo (delete element foo))' to be sure of correctly
1554 changing the value of a sequence `foo'. */)
1555 (Lisp_Object elt, Lisp_Object seq)
1556 {
1557 if (VECTORP (seq))
1558 {
1559 ptrdiff_t i, n;
1560
1561 for (i = n = 0; i < ASIZE (seq); ++i)
1562 if (NILP (Fequal (AREF (seq, i), elt)))
1563 ++n;
1564
1565 if (n != ASIZE (seq))
1566 {
1567 struct Lisp_Vector *p = allocate_vector (n);
1568
1569 for (i = n = 0; i < ASIZE (seq); ++i)
1570 if (NILP (Fequal (AREF (seq, i), elt)))
1571 p->contents[n++] = AREF (seq, i);
1572
1573 XSETVECTOR (seq, p);
1574 }
1575 }
1576 else if (STRINGP (seq))
1577 {
1578 ptrdiff_t i, ibyte, nchars, nbytes, cbytes;
1579 int c;
1580
1581 for (i = nchars = nbytes = ibyte = 0;
1582 i < SCHARS (seq);
1583 ++i, ibyte += cbytes)
1584 {
1585 if (STRING_MULTIBYTE (seq))
1586 {
1587 c = STRING_CHAR (SDATA (seq) + ibyte);
1588 cbytes = CHAR_BYTES (c);
1589 }
1590 else
1591 {
1592 c = SREF (seq, i);
1593 cbytes = 1;
1594 }
1595
1596 if (!INTEGERP (elt) || c != XINT (elt))
1597 {
1598 ++nchars;
1599 nbytes += cbytes;
1600 }
1601 }
1602
1603 if (nchars != SCHARS (seq))
1604 {
1605 Lisp_Object tem;
1606
1607 tem = make_uninit_multibyte_string (nchars, nbytes);
1608 if (!STRING_MULTIBYTE (seq))
1609 STRING_SET_UNIBYTE (tem);
1610
1611 for (i = nchars = nbytes = ibyte = 0;
1612 i < SCHARS (seq);
1613 ++i, ibyte += cbytes)
1614 {
1615 if (STRING_MULTIBYTE (seq))
1616 {
1617 c = STRING_CHAR (SDATA (seq) + ibyte);
1618 cbytes = CHAR_BYTES (c);
1619 }
1620 else
1621 {
1622 c = SREF (seq, i);
1623 cbytes = 1;
1624 }
1625
1626 if (!INTEGERP (elt) || c != XINT (elt))
1627 {
1628 unsigned char *from = SDATA (seq) + ibyte;
1629 unsigned char *to = SDATA (tem) + nbytes;
1630 ptrdiff_t n;
1631
1632 ++nchars;
1633 nbytes += cbytes;
1634
1635 for (n = cbytes; n--; )
1636 *to++ = *from++;
1637 }
1638 }
1639
1640 seq = tem;
1641 }
1642 }
1643 else
1644 {
1645 Lisp_Object tail, prev;
1646
1647 for (tail = seq, prev = Qnil; CONSP (tail); tail = XCDR (tail))
1648 {
1649 CHECK_LIST_CONS (tail, seq);
1650
1651 if (!NILP (Fequal (elt, XCAR (tail))))
1652 {
1653 if (NILP (prev))
1654 seq = XCDR (tail);
1655 else
1656 Fsetcdr (prev, XCDR (tail));
1657 }
1658 else
1659 prev = tail;
1660 QUIT;
1661 }
1662 }
1663
1664 return seq;
1665 }
1666
1667 DEFUN ("nreverse", Fnreverse, Snreverse, 1, 1, 0,
1668 doc: /* Reverse order of items in a list, vector or string SEQ.
1669 If SEQ is a list, it should be nil-terminated.
1670 This function may destructively modify SEQ to produce the value. */)
1671 (Lisp_Object seq)
1672 {
1673 if (NILP (seq))
1674 return seq;
1675 else if (STRINGP (seq))
1676 return Freverse (seq);
1677 else if (CONSP (seq))
1678 {
1679 Lisp_Object prev, tail, next;
1680
1681 for (prev = Qnil, tail = seq; !NILP (tail); tail = next)
1682 {
1683 QUIT;
1684 CHECK_LIST_CONS (tail, tail);
1685 next = XCDR (tail);
1686 Fsetcdr (tail, prev);
1687 prev = tail;
1688 }
1689 seq = prev;
1690 }
1691 else if (VECTORP (seq))
1692 {
1693 ptrdiff_t i, size = ASIZE (seq);
1694
1695 for (i = 0; i < size / 2; i++)
1696 {
1697 Lisp_Object tem = AREF (seq, i);
1698 ASET (seq, i, AREF (seq, size - i - 1));
1699 ASET (seq, size - i - 1, tem);
1700 }
1701 }
1702 else if (BOOL_VECTOR_P (seq))
1703 {
1704 ptrdiff_t i, size = bool_vector_size (seq);
1705
1706 for (i = 0; i < size / 2; i++)
1707 {
1708 bool tem = bool_vector_bitref (seq, i);
1709 bool_vector_set (seq, i, bool_vector_bitref (seq, size - i - 1));
1710 bool_vector_set (seq, size - i - 1, tem);
1711 }
1712 }
1713 else
1714 wrong_type_argument (Qarrayp, seq);
1715 return seq;
1716 }
1717
1718 DEFUN ("reverse", Freverse, Sreverse, 1, 1, 0,
1719 doc: /* Return the reversed copy of list, vector, or string SEQ.
1720 See also the function `nreverse', which is used more often. */)
1721 (Lisp_Object seq)
1722 {
1723 Lisp_Object new;
1724
1725 if (NILP (seq))
1726 return Qnil;
1727 else if (CONSP (seq))
1728 {
1729 for (new = Qnil; CONSP (seq); seq = XCDR (seq))
1730 {
1731 QUIT;
1732 new = Fcons (XCAR (seq), new);
1733 }
1734 CHECK_LIST_END (seq, seq);
1735 }
1736 else if (VECTORP (seq))
1737 {
1738 ptrdiff_t i, size = ASIZE (seq);
1739
1740 new = make_uninit_vector (size);
1741 for (i = 0; i < size; i++)
1742 ASET (new, i, AREF (seq, size - i - 1));
1743 }
1744 else if (BOOL_VECTOR_P (seq))
1745 {
1746 ptrdiff_t i;
1747 EMACS_INT nbits = bool_vector_size (seq);
1748
1749 new = make_uninit_bool_vector (nbits);
1750 for (i = 0; i < nbits; i++)
1751 bool_vector_set (new, i, bool_vector_bitref (seq, nbits - i - 1));
1752 }
1753 else if (STRINGP (seq))
1754 {
1755 ptrdiff_t size = SCHARS (seq), bytes = SBYTES (seq);
1756
1757 if (size == bytes)
1758 {
1759 ptrdiff_t i;
1760
1761 new = make_uninit_string (size);
1762 for (i = 0; i < size; i++)
1763 SSET (new, i, SREF (seq, size - i - 1));
1764 }
1765 else
1766 {
1767 unsigned char *p, *q;
1768
1769 new = make_uninit_multibyte_string (size, bytes);
1770 p = SDATA (seq), q = SDATA (new) + bytes;
1771 while (q > SDATA (new))
1772 {
1773 int ch, len;
1774
1775 ch = STRING_CHAR_AND_LENGTH (p, len);
1776 p += len, q -= len;
1777 CHAR_STRING (ch, q);
1778 }
1779 }
1780 }
1781 else
1782 wrong_type_argument (Qsequencep, seq);
1783 return new;
1784 }
1785 \f
1786 DEFUN ("sort", Fsort, Ssort, 2, 2, 0,
1787 doc: /* Sort LIST, stably, comparing elements using PREDICATE.
1788 Returns the sorted list. LIST is modified by side effects.
1789 PREDICATE is called with two elements of LIST, and should return non-nil
1790 if the first element should sort before the second. */)
1791 (Lisp_Object list, Lisp_Object predicate)
1792 {
1793 Lisp_Object front, back;
1794 register Lisp_Object len, tem;
1795 struct gcpro gcpro1, gcpro2;
1796 EMACS_INT length;
1797
1798 front = list;
1799 len = Flength (list);
1800 length = XINT (len);
1801 if (length < 2)
1802 return list;
1803
1804 XSETINT (len, (length / 2) - 1);
1805 tem = Fnthcdr (len, list);
1806 back = Fcdr (tem);
1807 Fsetcdr (tem, Qnil);
1808
1809 GCPRO2 (front, back);
1810 front = Fsort (front, predicate);
1811 back = Fsort (back, predicate);
1812 UNGCPRO;
1813 return merge (front, back, predicate);
1814 }
1815
1816 Lisp_Object
1817 merge (Lisp_Object org_l1, Lisp_Object org_l2, Lisp_Object pred)
1818 {
1819 Lisp_Object value;
1820 register Lisp_Object tail;
1821 Lisp_Object tem;
1822 register Lisp_Object l1, l2;
1823 struct gcpro gcpro1, gcpro2, gcpro3, gcpro4;
1824
1825 l1 = org_l1;
1826 l2 = org_l2;
1827 tail = Qnil;
1828 value = Qnil;
1829
1830 /* It is sufficient to protect org_l1 and org_l2.
1831 When l1 and l2 are updated, we copy the new values
1832 back into the org_ vars. */
1833 GCPRO4 (org_l1, org_l2, pred, value);
1834
1835 while (1)
1836 {
1837 if (NILP (l1))
1838 {
1839 UNGCPRO;
1840 if (NILP (tail))
1841 return l2;
1842 Fsetcdr (tail, l2);
1843 return value;
1844 }
1845 if (NILP (l2))
1846 {
1847 UNGCPRO;
1848 if (NILP (tail))
1849 return l1;
1850 Fsetcdr (tail, l1);
1851 return value;
1852 }
1853 tem = call2 (pred, Fcar (l2), Fcar (l1));
1854 if (NILP (tem))
1855 {
1856 tem = l1;
1857 l1 = Fcdr (l1);
1858 org_l1 = l1;
1859 }
1860 else
1861 {
1862 tem = l2;
1863 l2 = Fcdr (l2);
1864 org_l2 = l2;
1865 }
1866 if (NILP (tail))
1867 value = tem;
1868 else
1869 Fsetcdr (tail, tem);
1870 tail = tem;
1871 }
1872 }
1873
1874 \f
1875 /* This does not check for quits. That is safe since it must terminate. */
1876
1877 DEFUN ("plist-get", Fplist_get, Splist_get, 2, 2, 0,
1878 doc: /* Extract a value from a property list.
1879 PLIST is a property list, which is a list of the form
1880 \(PROP1 VALUE1 PROP2 VALUE2...). This function returns the value
1881 corresponding to the given PROP, or nil if PROP is not one of the
1882 properties on the list. This function never signals an error. */)
1883 (Lisp_Object plist, Lisp_Object prop)
1884 {
1885 Lisp_Object tail, halftail;
1886
1887 /* halftail is used to detect circular lists. */
1888 tail = halftail = plist;
1889 while (CONSP (tail) && CONSP (XCDR (tail)))
1890 {
1891 if (EQ (prop, XCAR (tail)))
1892 return XCAR (XCDR (tail));
1893
1894 tail = XCDR (XCDR (tail));
1895 halftail = XCDR (halftail);
1896 if (EQ (tail, halftail))
1897 break;
1898 }
1899
1900 return Qnil;
1901 }
1902
1903 DEFUN ("get", Fget, Sget, 2, 2, 0,
1904 doc: /* Return the value of SYMBOL's PROPNAME property.
1905 This is the last value stored with `(put SYMBOL PROPNAME VALUE)'. */)
1906 (Lisp_Object symbol, Lisp_Object propname)
1907 {
1908 CHECK_SYMBOL (symbol);
1909 return Fplist_get (XSYMBOL (symbol)->plist, propname);
1910 }
1911
1912 DEFUN ("plist-put", Fplist_put, Splist_put, 3, 3, 0,
1913 doc: /* Change value in PLIST of PROP to VAL.
1914 PLIST is a property list, which is a list of the form
1915 \(PROP1 VALUE1 PROP2 VALUE2 ...). PROP is a symbol and VAL is any object.
1916 If PROP is already a property on the list, its value is set to VAL,
1917 otherwise the new PROP VAL pair is added. The new plist is returned;
1918 use `(setq x (plist-put x prop val))' to be sure to use the new value.
1919 The PLIST is modified by side effects. */)
1920 (Lisp_Object plist, register Lisp_Object prop, Lisp_Object val)
1921 {
1922 register Lisp_Object tail, prev;
1923 Lisp_Object newcell;
1924 prev = Qnil;
1925 for (tail = plist; CONSP (tail) && CONSP (XCDR (tail));
1926 tail = XCDR (XCDR (tail)))
1927 {
1928 if (EQ (prop, XCAR (tail)))
1929 {
1930 Fsetcar (XCDR (tail), val);
1931 return plist;
1932 }
1933
1934 prev = tail;
1935 QUIT;
1936 }
1937 newcell = Fcons (prop, Fcons (val, NILP (prev) ? plist : XCDR (XCDR (prev))));
1938 if (NILP (prev))
1939 return newcell;
1940 else
1941 Fsetcdr (XCDR (prev), newcell);
1942 return plist;
1943 }
1944
1945 DEFUN ("put", Fput, Sput, 3, 3, 0,
1946 doc: /* Store SYMBOL's PROPNAME property with value VALUE.
1947 It can be retrieved with `(get SYMBOL PROPNAME)'. */)
1948 (Lisp_Object symbol, Lisp_Object propname, Lisp_Object value)
1949 {
1950 CHECK_SYMBOL (symbol);
1951 set_symbol_plist
1952 (symbol, Fplist_put (XSYMBOL (symbol)->plist, propname, value));
1953 return value;
1954 }
1955 \f
1956 DEFUN ("lax-plist-get", Flax_plist_get, Slax_plist_get, 2, 2, 0,
1957 doc: /* Extract a value from a property list, comparing with `equal'.
1958 PLIST is a property list, which is a list of the form
1959 \(PROP1 VALUE1 PROP2 VALUE2...). This function returns the value
1960 corresponding to the given PROP, or nil if PROP is not
1961 one of the properties on the list. */)
1962 (Lisp_Object plist, Lisp_Object prop)
1963 {
1964 Lisp_Object tail;
1965
1966 for (tail = plist;
1967 CONSP (tail) && CONSP (XCDR (tail));
1968 tail = XCDR (XCDR (tail)))
1969 {
1970 if (! NILP (Fequal (prop, XCAR (tail))))
1971 return XCAR (XCDR (tail));
1972
1973 QUIT;
1974 }
1975
1976 CHECK_LIST_END (tail, prop);
1977
1978 return Qnil;
1979 }
1980
1981 DEFUN ("lax-plist-put", Flax_plist_put, Slax_plist_put, 3, 3, 0,
1982 doc: /* Change value in PLIST of PROP to VAL, comparing with `equal'.
1983 PLIST is a property list, which is a list of the form
1984 \(PROP1 VALUE1 PROP2 VALUE2 ...). PROP and VAL are any objects.
1985 If PROP is already a property on the list, its value is set to VAL,
1986 otherwise the new PROP VAL pair is added. The new plist is returned;
1987 use `(setq x (lax-plist-put x prop val))' to be sure to use the new value.
1988 The PLIST is modified by side effects. */)
1989 (Lisp_Object plist, register Lisp_Object prop, Lisp_Object val)
1990 {
1991 register Lisp_Object tail, prev;
1992 Lisp_Object newcell;
1993 prev = Qnil;
1994 for (tail = plist; CONSP (tail) && CONSP (XCDR (tail));
1995 tail = XCDR (XCDR (tail)))
1996 {
1997 if (! NILP (Fequal (prop, XCAR (tail))))
1998 {
1999 Fsetcar (XCDR (tail), val);
2000 return plist;
2001 }
2002
2003 prev = tail;
2004 QUIT;
2005 }
2006 newcell = list2 (prop, val);
2007 if (NILP (prev))
2008 return newcell;
2009 else
2010 Fsetcdr (XCDR (prev), newcell);
2011 return plist;
2012 }
2013 \f
2014 DEFUN ("eql", Feql, Seql, 2, 2, 0,
2015 doc: /* Return t if the two args are the same Lisp object.
2016 Floating-point numbers of equal value are `eql', but they may not be `eq'. */)
2017 (Lisp_Object obj1, Lisp_Object obj2)
2018 {
2019 return scm_is_true (scm_eqv_p (obj1, obj2)) ? Qt : Qnil;
2020 }
2021
2022 DEFUN ("equal", Fequal, Sequal, 2, 2, 0,
2023 doc: /* Return t if two Lisp objects have similar structure and contents.
2024 They must have the same data type.
2025 Conses are compared by comparing the cars and the cdrs.
2026 Vectors and strings are compared element by element.
2027 Numbers are compared by value, but integers cannot equal floats.
2028 (Use `=' if you want integers and floats to be able to be equal.)
2029 Symbols must match exactly. */)
2030 (register Lisp_Object o1, Lisp_Object o2)
2031 {
2032 return scm_is_true (scm_equal_p (o1, o2)) ? Qt : Qnil;
2033 }
2034
2035 SCM compare_text_properties = SCM_BOOL_F;
2036
2037 DEFUN ("equal-including-properties", Fequal_including_properties, Sequal_including_properties, 2, 2, 0,
2038 doc: /* Return t if two Lisp objects have similar structure and contents.
2039 This is like `equal' except that it compares the text properties
2040 of strings. (`equal' ignores text properties.) */)
2041 (register Lisp_Object o1, Lisp_Object o2)
2042 {
2043 Lisp_Object tem;
2044
2045 scm_dynwind_begin (0);
2046 scm_dynwind_fluid (compare_text_properties, SCM_BOOL_T);
2047 tem = Fequal (o1, o2);
2048 scm_dynwind_end ();
2049 return tem;
2050 }
2051
2052 static SCM
2053 misc_equal_p (SCM o1, SCM o2)
2054 {
2055 if (XMISCTYPE (o1) != XMISCTYPE (o2))
2056 return SCM_BOOL_F;
2057 if (OVERLAYP (o1))
2058 {
2059 if (NILP (Fequal (OVERLAY_START (o1), OVERLAY_START (o2)))
2060 || NILP (Fequal (OVERLAY_END (o1), OVERLAY_END (o2))))
2061 return SCM_BOOL_F;
2062 return scm_equal_p (XOVERLAY (o1)->plist, XOVERLAY (o2)->plist);
2063 }
2064 if (MARKERP (o1))
2065 {
2066 struct Lisp_Marker *m1 = XMARKER (o1), *m2 = XMARKER (o2);
2067 return scm_from_bool (m1->buffer == m2->buffer
2068 && (m1->buffer == 0
2069 || m1->bytepos == m2->bytepos));
2070 }
2071 return SCM_BOOL_F;
2072 }
2073
2074 static SCM
2075 vectorlike_equal_p (SCM o1, SCM o2)
2076 {
2077 int i;
2078 ptrdiff_t size = ASIZE (o1);
2079 /* Pseudovectors have the type encoded in the size field, so this
2080 test actually checks that the objects have the same type as well
2081 as the same size. */
2082 if (ASIZE (o2) != size)
2083 return SCM_BOOL_F;
2084 /* Boolvectors are compared much like strings. */
2085 if (BOOL_VECTOR_P (o1))
2086 {
2087 if (XBOOL_VECTOR (o1)->size != XBOOL_VECTOR (o2)->size)
2088 return SCM_BOOL_F;
2089 if (memcmp (XBOOL_VECTOR (o1)->data, XBOOL_VECTOR (o2)->data,
2090 ((XBOOL_VECTOR (o1)->size
2091 + BOOL_VECTOR_BITS_PER_CHAR - 1)
2092 / BOOL_VECTOR_BITS_PER_CHAR)))
2093 return SCM_BOOL_F;
2094 return SCM_BOOL_T;
2095 }
2096 if (WINDOW_CONFIGURATIONP (o1))
2097 return scm_from_bool (compare_window_configurations (o1, o2, 0));
2098
2099 /* Aside from them, only true vectors, char-tables, compiled
2100 functions, and fonts (font-spec, font-entity, font-object) are
2101 sensible to compare, so eliminate the others now. */
2102 if (size & PSEUDOVECTOR_FLAG)
2103 {
2104 if (((size & PVEC_TYPE_MASK) >> PSEUDOVECTOR_AREA_BITS)
2105 < PVEC_COMPILED)
2106 return SCM_BOOL_F;
2107 size &= PSEUDOVECTOR_SIZE_MASK;
2108 }
2109 for (i = 0; i < size; i++)
2110 {
2111 Lisp_Object v1, v2;
2112 v1 = AREF (o1, i);
2113 v2 = AREF (o2, i);
2114 if (scm_is_false (scm_equal_p (v1, v2)))
2115 return SCM_BOOL_F;
2116 }
2117 return SCM_BOOL_T;
2118 }
2119
2120 static SCM
2121 string_equal_p (SCM o1, SCM o2)
2122 {
2123 if (SCHARS (o1) != SCHARS (o2))
2124 return SCM_BOOL_F;
2125 if (SBYTES (o1) != SBYTES (o2))
2126 return SCM_BOOL_F;
2127 if (memcmp (SDATA (o1), SDATA (o2), SBYTES (o1)))
2128 return SCM_BOOL_F;
2129 if (scm_is_true (scm_fluid_ref (compare_text_properties))
2130 && !compare_string_intervals (o1, o2))
2131 return SCM_BOOL_F;
2132 return SCM_BOOL_T;
2133 }
2134 \f
2135
2136 DEFUN ("fillarray", Ffillarray, Sfillarray, 2, 2, 0,
2137 doc: /* Store each element of ARRAY with ITEM.
2138 ARRAY is a vector, string, char-table, or bool-vector. */)
2139 (Lisp_Object array, Lisp_Object item)
2140 {
2141 register ptrdiff_t size, idx;
2142
2143 if (VECTORP (array))
2144 for (idx = 0, size = ASIZE (array); idx < size; idx++)
2145 ASET (array, idx, item);
2146 else if (CHAR_TABLE_P (array))
2147 {
2148 int i;
2149
2150 for (i = 0; i < (1 << CHARTAB_SIZE_BITS_0); i++)
2151 set_char_table_contents (array, i, item);
2152 set_char_table_defalt (array, item);
2153 }
2154 else if (STRINGP (array))
2155 {
2156 register unsigned char *p = SDATA (array);
2157 int charval;
2158 CHECK_CHARACTER (item);
2159 charval = XFASTINT (item);
2160 size = SCHARS (array);
2161 if (STRING_MULTIBYTE (array))
2162 {
2163 unsigned char str[MAX_MULTIBYTE_LENGTH];
2164 int len = CHAR_STRING (charval, str);
2165 ptrdiff_t size_byte = SBYTES (array);
2166
2167 if (INT_MULTIPLY_OVERFLOW (SCHARS (array), len)
2168 || SCHARS (array) * len != size_byte)
2169 error ("Attempt to change byte length of a string");
2170 for (idx = 0; idx < size_byte; idx++)
2171 *p++ = str[idx % len];
2172 }
2173 else
2174 for (idx = 0; idx < size; idx++)
2175 p[idx] = charval;
2176 }
2177 else if (BOOL_VECTOR_P (array))
2178 return bool_vector_fill (array, item);
2179 else
2180 wrong_type_argument (Qarrayp, array);
2181 return array;
2182 }
2183
2184 DEFUN ("clear-string", Fclear_string, Sclear_string,
2185 1, 1, 0,
2186 doc: /* Clear the contents of STRING.
2187 This makes STRING unibyte and may change its length. */)
2188 (Lisp_Object string)
2189 {
2190 ptrdiff_t len;
2191 CHECK_STRING (string);
2192 len = SBYTES (string);
2193 memset (SDATA (string), 0, len);
2194 STRING_SET_CHARS (string, len);
2195 STRING_SET_UNIBYTE (string);
2196 return Qnil;
2197 }
2198 \f
2199 /* ARGSUSED */
2200 Lisp_Object
2201 nconc2 (Lisp_Object s1, Lisp_Object s2)
2202 {
2203 Lisp_Object args[2];
2204 args[0] = s1;
2205 args[1] = s2;
2206 return Fnconc (2, args);
2207 }
2208
2209 DEFUN ("nconc", Fnconc, Snconc, 0, MANY, 0,
2210 doc: /* Concatenate any number of lists by altering them.
2211 Only the last argument is not altered, and need not be a list.
2212 usage: (nconc &rest LISTS) */)
2213 (ptrdiff_t nargs, Lisp_Object *args)
2214 {
2215 ptrdiff_t argnum;
2216 register Lisp_Object tail, tem, val;
2217
2218 val = tail = Qnil;
2219
2220 for (argnum = 0; argnum < nargs; argnum++)
2221 {
2222 tem = args[argnum];
2223 if (NILP (tem)) continue;
2224
2225 if (NILP (val))
2226 val = tem;
2227
2228 if (argnum + 1 == nargs) break;
2229
2230 CHECK_LIST_CONS (tem, tem);
2231
2232 while (CONSP (tem))
2233 {
2234 tail = tem;
2235 tem = XCDR (tail);
2236 QUIT;
2237 }
2238
2239 tem = args[argnum + 1];
2240 Fsetcdr (tail, tem);
2241 if (NILP (tem))
2242 args[argnum + 1] = tail;
2243 }
2244
2245 return val;
2246 }
2247 \f
2248 /* This is the guts of all mapping functions.
2249 Apply FN to each element of SEQ, one by one,
2250 storing the results into elements of VALS, a C vector of Lisp_Objects.
2251 LENI is the length of VALS, which should also be the length of SEQ. */
2252
2253 static void
2254 mapcar1 (EMACS_INT leni, Lisp_Object *vals, Lisp_Object fn, Lisp_Object seq)
2255 {
2256 register Lisp_Object tail;
2257 Lisp_Object dummy;
2258 register EMACS_INT i;
2259 struct gcpro gcpro1, gcpro2, gcpro3;
2260
2261 if (vals)
2262 {
2263 /* Don't let vals contain any garbage when GC happens. */
2264 for (i = 0; i < leni; i++)
2265 vals[i] = Qnil;
2266
2267 GCPRO3 (dummy, fn, seq);
2268 gcpro1.var = vals;
2269 gcpro1.nvars = leni;
2270 }
2271 else
2272 GCPRO2 (fn, seq);
2273 /* We need not explicitly protect `tail' because it is used only on lists, and
2274 1) lists are not relocated and 2) the list is marked via `seq' so will not
2275 be freed */
2276
2277 if (VECTORP (seq) || COMPILEDP (seq))
2278 {
2279 for (i = 0; i < leni; i++)
2280 {
2281 dummy = call1 (fn, AREF (seq, i));
2282 if (vals)
2283 vals[i] = dummy;
2284 }
2285 }
2286 else if (BOOL_VECTOR_P (seq))
2287 {
2288 for (i = 0; i < leni; i++)
2289 {
2290 dummy = call1 (fn, bool_vector_ref (seq, i));
2291 if (vals)
2292 vals[i] = dummy;
2293 }
2294 }
2295 else if (STRINGP (seq))
2296 {
2297 ptrdiff_t i_byte;
2298
2299 for (i = 0, i_byte = 0; i < leni;)
2300 {
2301 int c;
2302 ptrdiff_t i_before = i;
2303
2304 FETCH_STRING_CHAR_ADVANCE (c, seq, i, i_byte);
2305 XSETFASTINT (dummy, c);
2306 dummy = call1 (fn, dummy);
2307 if (vals)
2308 vals[i_before] = dummy;
2309 }
2310 }
2311 else /* Must be a list, since Flength did not get an error */
2312 {
2313 tail = seq;
2314 for (i = 0; i < leni && CONSP (tail); i++)
2315 {
2316 dummy = call1 (fn, XCAR (tail));
2317 if (vals)
2318 vals[i] = dummy;
2319 tail = XCDR (tail);
2320 }
2321 }
2322
2323 UNGCPRO;
2324 }
2325
2326 DEFUN ("mapconcat", Fmapconcat, Smapconcat, 3, 3, 0,
2327 doc: /* Apply FUNCTION to each element of SEQUENCE, and concat the results as strings.
2328 In between each pair of results, stick in SEPARATOR. Thus, " " as
2329 SEPARATOR results in spaces between the values returned by FUNCTION.
2330 SEQUENCE may be a list, a vector, a bool-vector, or a string. */)
2331 (Lisp_Object function, Lisp_Object sequence, Lisp_Object separator)
2332 {
2333 Lisp_Object len;
2334 register EMACS_INT leni;
2335 EMACS_INT nargs;
2336 ptrdiff_t i;
2337 register Lisp_Object *args;
2338 struct gcpro gcpro1;
2339 Lisp_Object ret;
2340 USE_SAFE_ALLOCA;
2341
2342 len = Flength (sequence);
2343 if (CHAR_TABLE_P (sequence))
2344 wrong_type_argument (Qlistp, sequence);
2345 leni = XINT (len);
2346 nargs = leni + leni - 1;
2347 if (nargs < 0) return empty_unibyte_string;
2348
2349 SAFE_ALLOCA_LISP (args, nargs);
2350
2351 GCPRO1 (separator);
2352 mapcar1 (leni, args, function, sequence);
2353 UNGCPRO;
2354
2355 for (i = leni - 1; i > 0; i--)
2356 args[i + i] = args[i];
2357
2358 for (i = 1; i < nargs; i += 2)
2359 args[i] = separator;
2360
2361 ret = Fconcat (nargs, args);
2362 SAFE_FREE ();
2363
2364 return ret;
2365 }
2366
2367 DEFUN ("mapcar", Fmapcar, Smapcar, 2, 2, 0,
2368 doc: /* Apply FUNCTION to each element of SEQUENCE, and make a list of the results.
2369 The result is a list just as long as SEQUENCE.
2370 SEQUENCE may be a list, a vector, a bool-vector, or a string. */)
2371 (Lisp_Object function, Lisp_Object sequence)
2372 {
2373 register Lisp_Object len;
2374 register EMACS_INT leni;
2375 register Lisp_Object *args;
2376 Lisp_Object ret;
2377 USE_SAFE_ALLOCA;
2378
2379 len = Flength (sequence);
2380 if (CHAR_TABLE_P (sequence))
2381 wrong_type_argument (Qlistp, sequence);
2382 leni = XFASTINT (len);
2383
2384 SAFE_ALLOCA_LISP (args, leni);
2385
2386 mapcar1 (leni, args, function, sequence);
2387
2388 ret = Flist (leni, args);
2389 SAFE_FREE ();
2390
2391 return ret;
2392 }
2393
2394 DEFUN ("mapc", Fmapc, Smapc, 2, 2, 0,
2395 doc: /* Apply FUNCTION to each element of SEQUENCE for side effects only.
2396 Unlike `mapcar', don't accumulate the results. Return SEQUENCE.
2397 SEQUENCE may be a list, a vector, a bool-vector, or a string. */)
2398 (Lisp_Object function, Lisp_Object sequence)
2399 {
2400 register EMACS_INT leni;
2401
2402 leni = XFASTINT (Flength (sequence));
2403 if (CHAR_TABLE_P (sequence))
2404 wrong_type_argument (Qlistp, sequence);
2405 mapcar1 (leni, 0, function, sequence);
2406
2407 return sequence;
2408 }
2409 \f
2410 /* This is how C code calls `yes-or-no-p' and allows the user
2411 to redefined it.
2412
2413 Anything that calls this function must protect from GC! */
2414
2415 Lisp_Object
2416 do_yes_or_no_p (Lisp_Object prompt)
2417 {
2418 return call1 (intern ("yes-or-no-p"), prompt);
2419 }
2420
2421 /* Anything that calls this function must protect from GC! */
2422
2423 DEFUN ("yes-or-no-p", Fyes_or_no_p, Syes_or_no_p, 1, 1, 0,
2424 doc: /* Ask user a yes-or-no question.
2425 Return t if answer is yes, and nil if the answer is no.
2426 PROMPT is the string to display to ask the question. It should end in
2427 a space; `yes-or-no-p' adds \"(yes or no) \" to it.
2428
2429 The user must confirm the answer with RET, and can edit it until it
2430 has been confirmed.
2431
2432 If dialog boxes are supported, a dialog box will be used
2433 if `last-nonmenu-event' is nil, and `use-dialog-box' is non-nil. */)
2434 (Lisp_Object prompt)
2435 {
2436 register Lisp_Object ans;
2437 Lisp_Object args[2];
2438 struct gcpro gcpro1;
2439
2440 CHECK_STRING (prompt);
2441
2442 if ((NILP (last_nonmenu_event) || CONSP (last_nonmenu_event))
2443 && use_dialog_box)
2444 {
2445 Lisp_Object pane, menu, obj;
2446 redisplay_preserve_echo_area (4);
2447 pane = list2 (Fcons (build_string ("Yes"), Qt),
2448 Fcons (build_string ("No"), Qnil));
2449 GCPRO1 (pane);
2450 menu = Fcons (prompt, pane);
2451 obj = Fx_popup_dialog (Qt, menu, Qnil);
2452 UNGCPRO;
2453 return obj;
2454 }
2455
2456 args[0] = prompt;
2457 args[1] = build_string ("(yes or no) ");
2458 prompt = Fconcat (2, args);
2459
2460 GCPRO1 (prompt);
2461
2462 while (1)
2463 {
2464 ans = Fdowncase (Fread_from_minibuffer (prompt, Qnil, Qnil, Qnil,
2465 Qyes_or_no_p_history, Qnil,
2466 Qnil));
2467 if (SCHARS (ans) == 3 && !strcmp (SSDATA (ans), "yes"))
2468 {
2469 UNGCPRO;
2470 return Qt;
2471 }
2472 if (SCHARS (ans) == 2 && !strcmp (SSDATA (ans), "no"))
2473 {
2474 UNGCPRO;
2475 return Qnil;
2476 }
2477
2478 Fding (Qnil);
2479 Fdiscard_input ();
2480 message1 ("Please answer yes or no.");
2481 Fsleep_for (make_number (2), Qnil);
2482 }
2483 }
2484 \f
2485 DEFUN ("load-average", Fload_average, Sload_average, 0, 1, 0,
2486 doc: /* Return list of 1 minute, 5 minute and 15 minute load averages.
2487
2488 Each of the three load averages is multiplied by 100, then converted
2489 to integer.
2490
2491 When USE-FLOATS is non-nil, floats will be used instead of integers.
2492 These floats are not multiplied by 100.
2493
2494 If the 5-minute or 15-minute load averages are not available, return a
2495 shortened list, containing only those averages which are available.
2496
2497 An error is thrown if the load average can't be obtained. In some
2498 cases making it work would require Emacs being installed setuid or
2499 setgid so that it can read kernel information, and that usually isn't
2500 advisable. */)
2501 (Lisp_Object use_floats)
2502 {
2503 double load_ave[3];
2504 int loads = getloadavg (load_ave, 3);
2505 Lisp_Object ret = Qnil;
2506
2507 if (loads < 0)
2508 error ("load-average not implemented for this operating system");
2509
2510 while (loads-- > 0)
2511 {
2512 Lisp_Object load = (NILP (use_floats)
2513 ? make_number (100.0 * load_ave[loads])
2514 : make_float (load_ave[loads]));
2515 ret = Fcons (load, ret);
2516 }
2517
2518 return ret;
2519 }
2520 \f
2521 static Lisp_Object Qsubfeatures;
2522
2523 DEFUN ("featurep", Ffeaturep, Sfeaturep, 1, 2, 0,
2524 doc: /* Return t if FEATURE is present in this Emacs.
2525
2526 Use this to conditionalize execution of lisp code based on the
2527 presence or absence of Emacs or environment extensions.
2528 Use `provide' to declare that a feature is available. This function
2529 looks at the value of the variable `features'. The optional argument
2530 SUBFEATURE can be used to check a specific subfeature of FEATURE. */)
2531 (Lisp_Object feature, Lisp_Object subfeature)
2532 {
2533 register Lisp_Object tem;
2534 CHECK_SYMBOL (feature);
2535 tem = Fmemq (feature, Vfeatures);
2536 if (!NILP (tem) && !NILP (subfeature))
2537 tem = Fmember (subfeature, Fget (feature, Qsubfeatures));
2538 return (NILP (tem)) ? Qnil : Qt;
2539 }
2540
2541 static Lisp_Object Qfuncall;
2542
2543 DEFUN ("provide", Fprovide, Sprovide, 1, 2, 0,
2544 doc: /* Announce that FEATURE is a feature of the current Emacs.
2545 The optional argument SUBFEATURES should be a list of symbols listing
2546 particular subfeatures supported in this version of FEATURE. */)
2547 (Lisp_Object feature, Lisp_Object subfeatures)
2548 {
2549 register Lisp_Object tem;
2550 CHECK_SYMBOL (feature);
2551 CHECK_LIST (subfeatures);
2552 if (!NILP (Vautoload_queue))
2553 Vautoload_queue = Fcons (Fcons (make_number (0), Vfeatures),
2554 Vautoload_queue);
2555 tem = Fmemq (feature, Vfeatures);
2556 if (NILP (tem))
2557 Vfeatures = Fcons (feature, Vfeatures);
2558 if (!NILP (subfeatures))
2559 Fput (feature, Qsubfeatures, subfeatures);
2560 LOADHIST_ATTACH (Fcons (Qprovide, feature));
2561
2562 /* Run any load-hooks for this file. */
2563 tem = Fassq (feature, Vafter_load_alist);
2564 if (CONSP (tem))
2565 Fmapc (Qfuncall, XCDR (tem));
2566
2567 return feature;
2568 }
2569 \f
2570 /* `require' and its subroutines. */
2571
2572 /* List of features currently being require'd, innermost first. */
2573
2574 static Lisp_Object require_nesting_list;
2575
2576 static void
2577 require_unwind (Lisp_Object old_value)
2578 {
2579 require_nesting_list = old_value;
2580 }
2581
2582 DEFUN ("require", Frequire, Srequire, 1, 3, 0,
2583 doc: /* If feature FEATURE is not loaded, load it from FILENAME.
2584 If FEATURE is not a member of the list `features', then the feature
2585 is not loaded; so load the file FILENAME.
2586 If FILENAME is omitted, the printname of FEATURE is used as the file name,
2587 and `load' will try to load this name appended with the suffix `.elc' or
2588 `.el', in that order. The name without appended suffix will not be used.
2589 See `get-load-suffixes' for the complete list of suffixes.
2590 If the optional third argument NOERROR is non-nil,
2591 then return nil if the file is not found instead of signaling an error.
2592 Normally the return value is FEATURE.
2593 The normal messages at start and end of loading FILENAME are suppressed. */)
2594 (Lisp_Object feature, Lisp_Object filename, Lisp_Object noerror)
2595 {
2596 Lisp_Object tem;
2597 struct gcpro gcpro1, gcpro2;
2598 bool from_file = load_in_progress;
2599
2600 CHECK_SYMBOL (feature);
2601
2602 /* Record the presence of `require' in this file
2603 even if the feature specified is already loaded.
2604 But not more than once in any file,
2605 and not when we aren't loading or reading from a file. */
2606 if (!from_file)
2607 for (tem = Vcurrent_load_list; CONSP (tem); tem = XCDR (tem))
2608 if (NILP (XCDR (tem)) && STRINGP (XCAR (tem)))
2609 from_file = 1;
2610
2611 if (from_file)
2612 {
2613 tem = Fcons (Qrequire, feature);
2614 if (NILP (Fmember (tem, Vcurrent_load_list)))
2615 LOADHIST_ATTACH (tem);
2616 }
2617 tem = Fmemq (feature, Vfeatures);
2618
2619 if (NILP (tem))
2620 {
2621 ptrdiff_t count = SPECPDL_INDEX ();
2622 int nesting = 0;
2623
2624 /* This is to make sure that loadup.el gives a clear picture
2625 of what files are preloaded and when. */
2626 if (! NILP (Vpurify_flag))
2627 error ("(require %s) while preparing to dump",
2628 SDATA (SYMBOL_NAME (feature)));
2629
2630 /* A certain amount of recursive `require' is legitimate,
2631 but if we require the same feature recursively 3 times,
2632 signal an error. */
2633 tem = require_nesting_list;
2634 while (! NILP (tem))
2635 {
2636 if (! NILP (Fequal (feature, XCAR (tem))))
2637 nesting++;
2638 tem = XCDR (tem);
2639 }
2640 if (nesting > 3)
2641 error ("Recursive `require' for feature `%s'",
2642 SDATA (SYMBOL_NAME (feature)));
2643
2644 /* Update the list for any nested `require's that occur. */
2645 record_unwind_protect (require_unwind, require_nesting_list);
2646 require_nesting_list = Fcons (feature, require_nesting_list);
2647
2648 /* Value saved here is to be restored into Vautoload_queue */
2649 record_unwind_protect (un_autoload, Vautoload_queue);
2650 Vautoload_queue = Qt;
2651
2652 /* Load the file. */
2653 GCPRO2 (feature, filename);
2654 tem = Fload (NILP (filename) ? Fsymbol_name (feature) : filename,
2655 noerror, Qt, Qnil, (NILP (filename) ? Qt : Qnil));
2656 UNGCPRO;
2657
2658 /* If load failed entirely, return nil. */
2659 if (NILP (tem))
2660 return unbind_to (count, Qnil);
2661
2662 tem = Fmemq (feature, Vfeatures);
2663 if (NILP (tem))
2664 error ("Required feature `%s' was not provided",
2665 SDATA (SYMBOL_NAME (feature)));
2666
2667 /* Once loading finishes, don't undo it. */
2668 Vautoload_queue = Qt;
2669 unbind_to (count, feature);
2670 }
2671
2672 return feature;
2673 }
2674 \f
2675 /* Primitives for work of the "widget" library.
2676 In an ideal world, this section would not have been necessary.
2677 However, lisp function calls being as slow as they are, it turns
2678 out that some functions in the widget library (wid-edit.el) are the
2679 bottleneck of Widget operation. Here is their translation to C,
2680 for the sole reason of efficiency. */
2681
2682 DEFUN ("plist-member", Fplist_member, Splist_member, 2, 2, 0,
2683 doc: /* Return non-nil if PLIST has the property PROP.
2684 PLIST is a property list, which is a list of the form
2685 \(PROP1 VALUE1 PROP2 VALUE2 ...\). PROP is a symbol.
2686 Unlike `plist-get', this allows you to distinguish between a missing
2687 property and a property with the value nil.
2688 The value is actually the tail of PLIST whose car is PROP. */)
2689 (Lisp_Object plist, Lisp_Object prop)
2690 {
2691 while (CONSP (plist) && !EQ (XCAR (plist), prop))
2692 {
2693 QUIT;
2694 plist = XCDR (plist);
2695 plist = CDR (plist);
2696 }
2697 return plist;
2698 }
2699
2700 DEFUN ("widget-put", Fwidget_put, Swidget_put, 3, 3, 0,
2701 doc: /* In WIDGET, set PROPERTY to VALUE.
2702 The value can later be retrieved with `widget-get'. */)
2703 (Lisp_Object widget, Lisp_Object property, Lisp_Object value)
2704 {
2705 CHECK_CONS (widget);
2706 XSETCDR (widget, Fplist_put (XCDR (widget), property, value));
2707 return value;
2708 }
2709
2710 DEFUN ("widget-get", Fwidget_get, Swidget_get, 2, 2, 0,
2711 doc: /* In WIDGET, get the value of PROPERTY.
2712 The value could either be specified when the widget was created, or
2713 later with `widget-put'. */)
2714 (Lisp_Object widget, Lisp_Object property)
2715 {
2716 Lisp_Object tmp;
2717
2718 while (1)
2719 {
2720 if (NILP (widget))
2721 return Qnil;
2722 CHECK_CONS (widget);
2723 tmp = Fplist_member (XCDR (widget), property);
2724 if (CONSP (tmp))
2725 {
2726 tmp = XCDR (tmp);
2727 return CAR (tmp);
2728 }
2729 tmp = XCAR (widget);
2730 if (NILP (tmp))
2731 return Qnil;
2732 widget = Fget (tmp, Qwidget_type);
2733 }
2734 }
2735
2736 DEFUN ("widget-apply", Fwidget_apply, Swidget_apply, 2, MANY, 0,
2737 doc: /* Apply the value of WIDGET's PROPERTY to the widget itself.
2738 ARGS are passed as extra arguments to the function.
2739 usage: (widget-apply WIDGET PROPERTY &rest ARGS) */)
2740 (ptrdiff_t nargs, Lisp_Object *args)
2741 {
2742 /* This function can GC. */
2743 Lisp_Object newargs[3];
2744 struct gcpro gcpro1, gcpro2;
2745 Lisp_Object result;
2746
2747 newargs[0] = Fwidget_get (args[0], args[1]);
2748 newargs[1] = args[0];
2749 newargs[2] = Flist (nargs - 2, args + 2);
2750 GCPRO2 (newargs[0], newargs[2]);
2751 result = Fapply (3, newargs);
2752 UNGCPRO;
2753 return result;
2754 }
2755
2756 #ifdef HAVE_LANGINFO_CODESET
2757 #include <langinfo.h>
2758 #endif
2759
2760 DEFUN ("locale-info", Flocale_info, Slocale_info, 1, 1, 0,
2761 doc: /* Access locale data ITEM for the current C locale, if available.
2762 ITEM should be one of the following:
2763
2764 `codeset', returning the character set as a string (locale item CODESET);
2765
2766 `days', returning a 7-element vector of day names (locale items DAY_n);
2767
2768 `months', returning a 12-element vector of month names (locale items MON_n);
2769
2770 `paper', returning a list (WIDTH HEIGHT) for the default paper size,
2771 both measured in millimeters (locale items PAPER_WIDTH, PAPER_HEIGHT).
2772
2773 If the system can't provide such information through a call to
2774 `nl_langinfo', or if ITEM isn't from the list above, return nil.
2775
2776 See also Info node `(libc)Locales'.
2777
2778 The data read from the system are decoded using `locale-coding-system'. */)
2779 (Lisp_Object item)
2780 {
2781 char *str = NULL;
2782 #ifdef HAVE_LANGINFO_CODESET
2783 Lisp_Object val;
2784 if (EQ (item, Qcodeset))
2785 {
2786 str = nl_langinfo (CODESET);
2787 return build_string (str);
2788 }
2789 #ifdef DAY_1
2790 else if (EQ (item, Qdays)) /* e.g. for calendar-day-name-array */
2791 {
2792 Lisp_Object v = Fmake_vector (make_number (7), Qnil);
2793 const int days[7] = {DAY_1, DAY_2, DAY_3, DAY_4, DAY_5, DAY_6, DAY_7};
2794 int i;
2795 struct gcpro gcpro1;
2796 GCPRO1 (v);
2797 synchronize_system_time_locale ();
2798 for (i = 0; i < 7; i++)
2799 {
2800 str = nl_langinfo (days[i]);
2801 val = build_unibyte_string (str);
2802 /* Fixme: Is this coding system necessarily right, even if
2803 it is consistent with CODESET? If not, what to do? */
2804 ASET (v, i, code_convert_string_norecord (val, Vlocale_coding_system,
2805 0));
2806 }
2807 UNGCPRO;
2808 return v;
2809 }
2810 #endif /* DAY_1 */
2811 #ifdef MON_1
2812 else if (EQ (item, Qmonths)) /* e.g. for calendar-month-name-array */
2813 {
2814 Lisp_Object v = Fmake_vector (make_number (12), Qnil);
2815 const int months[12] = {MON_1, MON_2, MON_3, MON_4, MON_5, MON_6, MON_7,
2816 MON_8, MON_9, MON_10, MON_11, MON_12};
2817 int i;
2818 struct gcpro gcpro1;
2819 GCPRO1 (v);
2820 synchronize_system_time_locale ();
2821 for (i = 0; i < 12; i++)
2822 {
2823 str = nl_langinfo (months[i]);
2824 val = build_unibyte_string (str);
2825 ASET (v, i, code_convert_string_norecord (val, Vlocale_coding_system,
2826 0));
2827 }
2828 UNGCPRO;
2829 return v;
2830 }
2831 #endif /* MON_1 */
2832 /* LC_PAPER stuff isn't defined as accessible in glibc as of 2.3.1,
2833 but is in the locale files. This could be used by ps-print. */
2834 #ifdef PAPER_WIDTH
2835 else if (EQ (item, Qpaper))
2836 return list2i (nl_langinfo (PAPER_WIDTH), nl_langinfo (PAPER_HEIGHT));
2837 #endif /* PAPER_WIDTH */
2838 #endif /* HAVE_LANGINFO_CODESET*/
2839 return Qnil;
2840 }
2841 \f
2842 /* base64 encode/decode functions (RFC 2045).
2843 Based on code from GNU recode. */
2844
2845 #define MIME_LINE_LENGTH 76
2846
2847 #define IS_ASCII(Character) \
2848 ((Character) < 128)
2849 #define IS_BASE64(Character) \
2850 (IS_ASCII (Character) && base64_char_to_value[Character] >= 0)
2851 #define IS_BASE64_IGNORABLE(Character) \
2852 ((Character) == ' ' || (Character) == '\t' || (Character) == '\n' \
2853 || (Character) == '\f' || (Character) == '\r')
2854
2855 /* Used by base64_decode_1 to retrieve a non-base64-ignorable
2856 character or return retval if there are no characters left to
2857 process. */
2858 #define READ_QUADRUPLET_BYTE(retval) \
2859 do \
2860 { \
2861 if (i == length) \
2862 { \
2863 if (nchars_return) \
2864 *nchars_return = nchars; \
2865 return (retval); \
2866 } \
2867 c = from[i++]; \
2868 } \
2869 while (IS_BASE64_IGNORABLE (c))
2870
2871 /* Table of characters coding the 64 values. */
2872 static const char base64_value_to_char[64] =
2873 {
2874 'A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'I', 'J', /* 0- 9 */
2875 'K', 'L', 'M', 'N', 'O', 'P', 'Q', 'R', 'S', 'T', /* 10-19 */
2876 'U', 'V', 'W', 'X', 'Y', 'Z', 'a', 'b', 'c', 'd', /* 20-29 */
2877 'e', 'f', 'g', 'h', 'i', 'j', 'k', 'l', 'm', 'n', /* 30-39 */
2878 'o', 'p', 'q', 'r', 's', 't', 'u', 'v', 'w', 'x', /* 40-49 */
2879 'y', 'z', '0', '1', '2', '3', '4', '5', '6', '7', /* 50-59 */
2880 '8', '9', '+', '/' /* 60-63 */
2881 };
2882
2883 /* Table of base64 values for first 128 characters. */
2884 static const short base64_char_to_value[128] =
2885 {
2886 -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, /* 0- 9 */
2887 -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, /* 10- 19 */
2888 -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, /* 20- 29 */
2889 -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, /* 30- 39 */
2890 -1, -1, -1, 62, -1, -1, -1, 63, 52, 53, /* 40- 49 */
2891 54, 55, 56, 57, 58, 59, 60, 61, -1, -1, /* 50- 59 */
2892 -1, -1, -1, -1, -1, 0, 1, 2, 3, 4, /* 60- 69 */
2893 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, /* 70- 79 */
2894 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, /* 80- 89 */
2895 25, -1, -1, -1, -1, -1, -1, 26, 27, 28, /* 90- 99 */
2896 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, /* 100-109 */
2897 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, /* 110-119 */
2898 49, 50, 51, -1, -1, -1, -1, -1 /* 120-127 */
2899 };
2900
2901 /* The following diagram shows the logical steps by which three octets
2902 get transformed into four base64 characters.
2903
2904 .--------. .--------. .--------.
2905 |aaaaaabb| |bbbbcccc| |ccdddddd|
2906 `--------' `--------' `--------'
2907 6 2 4 4 2 6
2908 .--------+--------+--------+--------.
2909 |00aaaaaa|00bbbbbb|00cccccc|00dddddd|
2910 `--------+--------+--------+--------'
2911
2912 .--------+--------+--------+--------.
2913 |AAAAAAAA|BBBBBBBB|CCCCCCCC|DDDDDDDD|
2914 `--------+--------+--------+--------'
2915
2916 The octets are divided into 6 bit chunks, which are then encoded into
2917 base64 characters. */
2918
2919
2920 static ptrdiff_t base64_encode_1 (const char *, char *, ptrdiff_t, bool, bool);
2921 static ptrdiff_t base64_decode_1 (const char *, char *, ptrdiff_t, bool,
2922 ptrdiff_t *);
2923
2924 DEFUN ("base64-encode-region", Fbase64_encode_region, Sbase64_encode_region,
2925 2, 3, "r",
2926 doc: /* Base64-encode the region between BEG and END.
2927 Return the length of the encoded text.
2928 Optional third argument NO-LINE-BREAK means do not break long lines
2929 into shorter lines. */)
2930 (Lisp_Object beg, Lisp_Object end, Lisp_Object no_line_break)
2931 {
2932 char *encoded;
2933 ptrdiff_t allength, length;
2934 ptrdiff_t ibeg, iend, encoded_length;
2935 ptrdiff_t old_pos = PT;
2936 USE_SAFE_ALLOCA;
2937
2938 validate_region (&beg, &end);
2939
2940 ibeg = CHAR_TO_BYTE (XFASTINT (beg));
2941 iend = CHAR_TO_BYTE (XFASTINT (end));
2942 move_gap_both (XFASTINT (beg), ibeg);
2943
2944 /* We need to allocate enough room for encoding the text.
2945 We need 33 1/3% more space, plus a newline every 76
2946 characters, and then we round up. */
2947 length = iend - ibeg;
2948 allength = length + length/3 + 1;
2949 allength += allength / MIME_LINE_LENGTH + 1 + 6;
2950
2951 encoded = SAFE_ALLOCA (allength);
2952 encoded_length = base64_encode_1 ((char *) BYTE_POS_ADDR (ibeg),
2953 encoded, length, NILP (no_line_break),
2954 !NILP (BVAR (current_buffer, enable_multibyte_characters)));
2955 if (encoded_length > allength)
2956 emacs_abort ();
2957
2958 if (encoded_length < 0)
2959 {
2960 /* The encoding wasn't possible. */
2961 SAFE_FREE ();
2962 error ("Multibyte character in data for base64 encoding");
2963 }
2964
2965 /* Now we have encoded the region, so we insert the new contents
2966 and delete the old. (Insert first in order to preserve markers.) */
2967 SET_PT_BOTH (XFASTINT (beg), ibeg);
2968 insert (encoded, encoded_length);
2969 SAFE_FREE ();
2970 del_range_byte (ibeg + encoded_length, iend + encoded_length, 1);
2971
2972 /* If point was outside of the region, restore it exactly; else just
2973 move to the beginning of the region. */
2974 if (old_pos >= XFASTINT (end))
2975 old_pos += encoded_length - (XFASTINT (end) - XFASTINT (beg));
2976 else if (old_pos > XFASTINT (beg))
2977 old_pos = XFASTINT (beg);
2978 SET_PT (old_pos);
2979
2980 /* We return the length of the encoded text. */
2981 return make_number (encoded_length);
2982 }
2983
2984 DEFUN ("base64-encode-string", Fbase64_encode_string, Sbase64_encode_string,
2985 1, 2, 0,
2986 doc: /* Base64-encode STRING and return the result.
2987 Optional second argument NO-LINE-BREAK means do not break long lines
2988 into shorter lines. */)
2989 (Lisp_Object string, Lisp_Object no_line_break)
2990 {
2991 ptrdiff_t allength, length, encoded_length;
2992 char *encoded;
2993 Lisp_Object encoded_string;
2994 USE_SAFE_ALLOCA;
2995
2996 CHECK_STRING (string);
2997
2998 /* We need to allocate enough room for encoding the text.
2999 We need 33 1/3% more space, plus a newline every 76
3000 characters, and then we round up. */
3001 length = SBYTES (string);
3002 allength = length + length/3 + 1;
3003 allength += allength / MIME_LINE_LENGTH + 1 + 6;
3004
3005 /* We need to allocate enough room for decoding the text. */
3006 encoded = SAFE_ALLOCA (allength);
3007
3008 encoded_length = base64_encode_1 (SSDATA (string),
3009 encoded, length, NILP (no_line_break),
3010 STRING_MULTIBYTE (string));
3011 if (encoded_length > allength)
3012 emacs_abort ();
3013
3014 if (encoded_length < 0)
3015 {
3016 /* The encoding wasn't possible. */
3017 SAFE_FREE ();
3018 error ("Multibyte character in data for base64 encoding");
3019 }
3020
3021 encoded_string = make_unibyte_string (encoded, encoded_length);
3022 SAFE_FREE ();
3023
3024 return encoded_string;
3025 }
3026
3027 static ptrdiff_t
3028 base64_encode_1 (const char *from, char *to, ptrdiff_t length,
3029 bool line_break, bool multibyte)
3030 {
3031 int counter = 0;
3032 ptrdiff_t i = 0;
3033 char *e = to;
3034 int c;
3035 unsigned int value;
3036 int bytes;
3037
3038 while (i < length)
3039 {
3040 if (multibyte)
3041 {
3042 c = STRING_CHAR_AND_LENGTH ((unsigned char *) from + i, bytes);
3043 if (CHAR_BYTE8_P (c))
3044 c = CHAR_TO_BYTE8 (c);
3045 else if (c >= 256)
3046 return -1;
3047 i += bytes;
3048 }
3049 else
3050 c = from[i++];
3051
3052 /* Wrap line every 76 characters. */
3053
3054 if (line_break)
3055 {
3056 if (counter < MIME_LINE_LENGTH / 4)
3057 counter++;
3058 else
3059 {
3060 *e++ = '\n';
3061 counter = 1;
3062 }
3063 }
3064
3065 /* Process first byte of a triplet. */
3066
3067 *e++ = base64_value_to_char[0x3f & c >> 2];
3068 value = (0x03 & c) << 4;
3069
3070 /* Process second byte of a triplet. */
3071
3072 if (i == length)
3073 {
3074 *e++ = base64_value_to_char[value];
3075 *e++ = '=';
3076 *e++ = '=';
3077 break;
3078 }
3079
3080 if (multibyte)
3081 {
3082 c = STRING_CHAR_AND_LENGTH ((unsigned char *) from + i, bytes);
3083 if (CHAR_BYTE8_P (c))
3084 c = CHAR_TO_BYTE8 (c);
3085 else if (c >= 256)
3086 return -1;
3087 i += bytes;
3088 }
3089 else
3090 c = from[i++];
3091
3092 *e++ = base64_value_to_char[value | (0x0f & c >> 4)];
3093 value = (0x0f & c) << 2;
3094
3095 /* Process third byte of a triplet. */
3096
3097 if (i == length)
3098 {
3099 *e++ = base64_value_to_char[value];
3100 *e++ = '=';
3101 break;
3102 }
3103
3104 if (multibyte)
3105 {
3106 c = STRING_CHAR_AND_LENGTH ((unsigned char *) from + i, bytes);
3107 if (CHAR_BYTE8_P (c))
3108 c = CHAR_TO_BYTE8 (c);
3109 else if (c >= 256)
3110 return -1;
3111 i += bytes;
3112 }
3113 else
3114 c = from[i++];
3115
3116 *e++ = base64_value_to_char[value | (0x03 & c >> 6)];
3117 *e++ = base64_value_to_char[0x3f & c];
3118 }
3119
3120 return e - to;
3121 }
3122
3123
3124 DEFUN ("base64-decode-region", Fbase64_decode_region, Sbase64_decode_region,
3125 2, 2, "r",
3126 doc: /* Base64-decode the region between BEG and END.
3127 Return the length of the decoded text.
3128 If the region can't be decoded, signal an error and don't modify the buffer. */)
3129 (Lisp_Object beg, Lisp_Object end)
3130 {
3131 ptrdiff_t ibeg, iend, length, allength;
3132 char *decoded;
3133 ptrdiff_t old_pos = PT;
3134 ptrdiff_t decoded_length;
3135 ptrdiff_t inserted_chars;
3136 bool multibyte = !NILP (BVAR (current_buffer, enable_multibyte_characters));
3137 USE_SAFE_ALLOCA;
3138
3139 validate_region (&beg, &end);
3140
3141 ibeg = CHAR_TO_BYTE (XFASTINT (beg));
3142 iend = CHAR_TO_BYTE (XFASTINT (end));
3143
3144 length = iend - ibeg;
3145
3146 /* We need to allocate enough room for decoding the text. If we are
3147 working on a multibyte buffer, each decoded code may occupy at
3148 most two bytes. */
3149 allength = multibyte ? length * 2 : length;
3150 decoded = SAFE_ALLOCA (allength);
3151
3152 move_gap_both (XFASTINT (beg), ibeg);
3153 decoded_length = base64_decode_1 ((char *) BYTE_POS_ADDR (ibeg),
3154 decoded, length,
3155 multibyte, &inserted_chars);
3156 if (decoded_length > allength)
3157 emacs_abort ();
3158
3159 if (decoded_length < 0)
3160 {
3161 /* The decoding wasn't possible. */
3162 SAFE_FREE ();
3163 error ("Invalid base64 data");
3164 }
3165
3166 /* Now we have decoded the region, so we insert the new contents
3167 and delete the old. (Insert first in order to preserve markers.) */
3168 TEMP_SET_PT_BOTH (XFASTINT (beg), ibeg);
3169 insert_1_both (decoded, inserted_chars, decoded_length, 0, 1, 0);
3170 SAFE_FREE ();
3171
3172 /* Delete the original text. */
3173 del_range_both (PT, PT_BYTE, XFASTINT (end) + inserted_chars,
3174 iend + decoded_length, 1);
3175
3176 /* If point was outside of the region, restore it exactly; else just
3177 move to the beginning of the region. */
3178 if (old_pos >= XFASTINT (end))
3179 old_pos += inserted_chars - (XFASTINT (end) - XFASTINT (beg));
3180 else if (old_pos > XFASTINT (beg))
3181 old_pos = XFASTINT (beg);
3182 SET_PT (old_pos > ZV ? ZV : old_pos);
3183
3184 return make_number (inserted_chars);
3185 }
3186
3187 DEFUN ("base64-decode-string", Fbase64_decode_string, Sbase64_decode_string,
3188 1, 1, 0,
3189 doc: /* Base64-decode STRING and return the result. */)
3190 (Lisp_Object string)
3191 {
3192 char *decoded;
3193 ptrdiff_t length, decoded_length;
3194 Lisp_Object decoded_string;
3195 USE_SAFE_ALLOCA;
3196
3197 CHECK_STRING (string);
3198
3199 length = SBYTES (string);
3200 /* We need to allocate enough room for decoding the text. */
3201 decoded = SAFE_ALLOCA (length);
3202
3203 /* The decoded result should be unibyte. */
3204 decoded_length = base64_decode_1 (SSDATA (string), decoded, length,
3205 0, NULL);
3206 if (decoded_length > length)
3207 emacs_abort ();
3208 else if (decoded_length >= 0)
3209 decoded_string = make_unibyte_string (decoded, decoded_length);
3210 else
3211 decoded_string = Qnil;
3212
3213 SAFE_FREE ();
3214 if (!STRINGP (decoded_string))
3215 error ("Invalid base64 data");
3216
3217 return decoded_string;
3218 }
3219
3220 /* Base64-decode the data at FROM of LENGTH bytes into TO. If
3221 MULTIBYTE, the decoded result should be in multibyte
3222 form. If NCHARS_RETURN is not NULL, store the number of produced
3223 characters in *NCHARS_RETURN. */
3224
3225 static ptrdiff_t
3226 base64_decode_1 (const char *from, char *to, ptrdiff_t length,
3227 bool multibyte, ptrdiff_t *nchars_return)
3228 {
3229 ptrdiff_t i = 0; /* Used inside READ_QUADRUPLET_BYTE */
3230 char *e = to;
3231 unsigned char c;
3232 unsigned long value;
3233 ptrdiff_t nchars = 0;
3234
3235 while (1)
3236 {
3237 /* Process first byte of a quadruplet. */
3238
3239 READ_QUADRUPLET_BYTE (e-to);
3240
3241 if (!IS_BASE64 (c))
3242 return -1;
3243 value = base64_char_to_value[c] << 18;
3244
3245 /* Process second byte of a quadruplet. */
3246
3247 READ_QUADRUPLET_BYTE (-1);
3248
3249 if (!IS_BASE64 (c))
3250 return -1;
3251 value |= base64_char_to_value[c] << 12;
3252
3253 c = (unsigned char) (value >> 16);
3254 if (multibyte && c >= 128)
3255 e += BYTE8_STRING (c, e);
3256 else
3257 *e++ = c;
3258 nchars++;
3259
3260 /* Process third byte of a quadruplet. */
3261
3262 READ_QUADRUPLET_BYTE (-1);
3263
3264 if (c == '=')
3265 {
3266 READ_QUADRUPLET_BYTE (-1);
3267
3268 if (c != '=')
3269 return -1;
3270 continue;
3271 }
3272
3273 if (!IS_BASE64 (c))
3274 return -1;
3275 value |= base64_char_to_value[c] << 6;
3276
3277 c = (unsigned char) (0xff & value >> 8);
3278 if (multibyte && c >= 128)
3279 e += BYTE8_STRING (c, e);
3280 else
3281 *e++ = c;
3282 nchars++;
3283
3284 /* Process fourth byte of a quadruplet. */
3285
3286 READ_QUADRUPLET_BYTE (-1);
3287
3288 if (c == '=')
3289 continue;
3290
3291 if (!IS_BASE64 (c))
3292 return -1;
3293 value |= base64_char_to_value[c];
3294
3295 c = (unsigned char) (0xff & value);
3296 if (multibyte && c >= 128)
3297 e += BYTE8_STRING (c, e);
3298 else
3299 *e++ = c;
3300 nchars++;
3301 }
3302 }
3303
3304
3305 \f
3306 /***********************************************************************
3307 ***** *****
3308 ***** Hash Tables *****
3309 ***** *****
3310 ***********************************************************************/
3311
3312 /* Implemented by gerd@gnu.org. This hash table implementation was
3313 inspired by CMUCL hash tables. */
3314
3315 /* Ideas:
3316
3317 1. For small tables, association lists are probably faster than
3318 hash tables because they have lower overhead.
3319
3320 For uses of hash tables where the O(1) behavior of table
3321 operations is not a requirement, it might therefore be a good idea
3322 not to hash. Instead, we could just do a linear search in the
3323 key_and_value vector of the hash table. This could be done
3324 if a `:linear-search t' argument is given to make-hash-table. */
3325
3326 /* Various symbols. */
3327
3328 static Lisp_Object Qhash_table_p;
3329 static Lisp_Object Qkey, Qvalue, Qeql;
3330 Lisp_Object Qeq, Qequal;
3331 Lisp_Object QCtest, QCsize, QCrehash_size, QCrehash_threshold, QCweakness;
3332 static Lisp_Object Qhash_table_test, Qkey_or_value, Qkey_and_value;
3333
3334 \f
3335 /***********************************************************************
3336 Utilities
3337 ***********************************************************************/
3338
3339 static void
3340 CHECK_HASH_TABLE (Lisp_Object x)
3341 {
3342 CHECK_TYPE (HASH_TABLE_P (x), Qhash_table_p, x);
3343 }
3344
3345 static void
3346 set_hash_key_and_value (struct Lisp_Hash_Table *h, Lisp_Object key_and_value)
3347 {
3348 h->key_and_value = key_and_value;
3349 }
3350 static void
3351 set_hash_next (struct Lisp_Hash_Table *h, Lisp_Object next)
3352 {
3353 h->next = next;
3354 }
3355 static void
3356 set_hash_next_slot (struct Lisp_Hash_Table *h, ptrdiff_t idx, Lisp_Object val)
3357 {
3358 gc_aset (h->next, idx, val);
3359 }
3360 static void
3361 set_hash_hash (struct Lisp_Hash_Table *h, Lisp_Object hash)
3362 {
3363 h->hash = hash;
3364 }
3365 static void
3366 set_hash_hash_slot (struct Lisp_Hash_Table *h, ptrdiff_t idx, Lisp_Object val)
3367 {
3368 gc_aset (h->hash, idx, val);
3369 }
3370 static void
3371 set_hash_index (struct Lisp_Hash_Table *h, Lisp_Object index)
3372 {
3373 h->index = index;
3374 }
3375 static void
3376 set_hash_index_slot (struct Lisp_Hash_Table *h, ptrdiff_t idx, Lisp_Object val)
3377 {
3378 gc_aset (h->index, idx, val);
3379 }
3380
3381 /* If OBJ is a Lisp hash table, return a pointer to its struct
3382 Lisp_Hash_Table. Otherwise, signal an error. */
3383
3384 static struct Lisp_Hash_Table *
3385 check_hash_table (Lisp_Object obj)
3386 {
3387 CHECK_HASH_TABLE (obj);
3388 return XHASH_TABLE (obj);
3389 }
3390
3391
3392 /* Value is the next integer I >= N, N >= 0 which is "almost" a prime
3393 number. A number is "almost" a prime number if it is not divisible
3394 by any integer in the range 2 .. (NEXT_ALMOST_PRIME_LIMIT - 1). */
3395
3396 EMACS_INT
3397 next_almost_prime (EMACS_INT n)
3398 {
3399 verify (NEXT_ALMOST_PRIME_LIMIT == 11);
3400 for (n |= 1; ; n += 2)
3401 if (n % 3 != 0 && n % 5 != 0 && n % 7 != 0)
3402 return n;
3403 }
3404
3405
3406 /* Find KEY in ARGS which has size NARGS. Don't consider indices for
3407 which USED[I] is non-zero. If found at index I in ARGS, set
3408 USED[I] and USED[I + 1] to 1, and return I + 1. Otherwise return
3409 0. This function is used to extract a keyword/argument pair from
3410 a DEFUN parameter list. */
3411
3412 static ptrdiff_t
3413 get_key_arg (Lisp_Object key, ptrdiff_t nargs, Lisp_Object *args, char *used)
3414 {
3415 ptrdiff_t i;
3416
3417 for (i = 1; i < nargs; i++)
3418 if (!used[i - 1] && EQ (args[i - 1], key))
3419 {
3420 used[i - 1] = 1;
3421 used[i] = 1;
3422 return i;
3423 }
3424
3425 return 0;
3426 }
3427
3428
3429 /* Return a Lisp vector which has the same contents as VEC but has
3430 at least INCR_MIN more entries, where INCR_MIN is positive.
3431 If NITEMS_MAX is not -1, do not grow the vector to be any larger
3432 than NITEMS_MAX. Entries in the resulting
3433 vector that are not copied from VEC are set to nil. */
3434
3435 Lisp_Object
3436 larger_vector (Lisp_Object vec, ptrdiff_t incr_min, ptrdiff_t nitems_max)
3437 {
3438 struct Lisp_Vector *v;
3439 ptrdiff_t i, incr, incr_max, old_size, new_size;
3440 ptrdiff_t C_language_max = min (PTRDIFF_MAX, SIZE_MAX) / sizeof *v->contents;
3441 ptrdiff_t n_max = (0 <= nitems_max && nitems_max < C_language_max
3442 ? nitems_max : C_language_max);
3443 eassert (VECTORP (vec));
3444 eassert (0 < incr_min && -1 <= nitems_max);
3445 old_size = ASIZE (vec);
3446 incr_max = n_max - old_size;
3447 incr = max (incr_min, min (old_size >> 1, incr_max));
3448 if (incr_max < incr)
3449 memory_full (SIZE_MAX);
3450 new_size = old_size + incr;
3451 v = allocate_vector (new_size);
3452 memcpy (v->contents, XVECTOR (vec)->contents, old_size * sizeof *v->contents);
3453 for (i = old_size; i < new_size; ++i)
3454 v->contents[i] = Qnil;
3455 XSETVECTOR (vec, v);
3456 return vec;
3457 }
3458
3459
3460 /***********************************************************************
3461 Low-level Functions
3462 ***********************************************************************/
3463
3464 static struct hash_table_test hashtest_eq;
3465 struct hash_table_test hashtest_eql, hashtest_equal;
3466
3467 /* Compare KEY1 which has hash code HASH1 and KEY2 with hash code
3468 HASH2 in hash table H using `eql'. Value is true if KEY1 and
3469 KEY2 are the same. */
3470
3471 static bool
3472 cmpfn_eql (struct hash_table_test *ht,
3473 Lisp_Object key1,
3474 Lisp_Object key2)
3475 {
3476 return (FLOATP (key1)
3477 && FLOATP (key2)
3478 && XFLOAT_DATA (key1) == XFLOAT_DATA (key2));
3479 }
3480
3481
3482 /* Compare KEY1 which has hash code HASH1 and KEY2 with hash code
3483 HASH2 in hash table H using `equal'. Value is true if KEY1 and
3484 KEY2 are the same. */
3485
3486 static bool
3487 cmpfn_equal (struct hash_table_test *ht,
3488 Lisp_Object key1,
3489 Lisp_Object key2)
3490 {
3491 return !NILP (Fequal (key1, key2));
3492 }
3493
3494
3495 /* Compare KEY1 which has hash code HASH1, and KEY2 with hash code
3496 HASH2 in hash table H using H->user_cmp_function. Value is true
3497 if KEY1 and KEY2 are the same. */
3498
3499 static bool
3500 cmpfn_user_defined (struct hash_table_test *ht,
3501 Lisp_Object key1,
3502 Lisp_Object key2)
3503 {
3504 Lisp_Object args[3];
3505
3506 args[0] = ht->user_cmp_function;
3507 args[1] = key1;
3508 args[2] = key2;
3509 return !NILP (Ffuncall (3, args));
3510 }
3511
3512
3513 /* Value is a hash code for KEY for use in hash table H which uses
3514 `eq' to compare keys. The hash code returned is guaranteed to fit
3515 in a Lisp integer. */
3516
3517 static EMACS_UINT
3518 hashfn_eq (struct hash_table_test *ht, Lisp_Object key)
3519 {
3520 return scm_ihashq (key, MOST_POSITIVE_FIXNUM);
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 hash_table_test *ht, Lisp_Object key)
3529 {
3530 return scm_ihashv (key, MOST_POSITIVE_FIXNUM);
3531 }
3532
3533 /* Value is a hash code for KEY for use in hash table H which uses
3534 `equal' to compare keys. The hash code returned is guaranteed to fit
3535 in a Lisp integer. */
3536
3537 static EMACS_UINT
3538 hashfn_equal (struct hash_table_test *ht, Lisp_Object key)
3539 {
3540 return scm_ihash (key, MOST_POSITIVE_FIXNUM);
3541 }
3542
3543 /* Value is a hash code for KEY for use in hash table H which uses as
3544 user-defined function to compare keys. The hash code returned is
3545 guaranteed to fit in a Lisp integer. */
3546
3547 static EMACS_UINT
3548 hashfn_user_defined (struct hash_table_test *ht, Lisp_Object key)
3549 {
3550 Lisp_Object args[2], hash;
3551
3552 args[0] = ht->user_hash_function;
3553 args[1] = key;
3554 hash = Ffuncall (2, args);
3555 return hashfn_eq (ht, hash);
3556 }
3557
3558 /* An upper bound on the size of a hash table index. It must fit in
3559 ptrdiff_t and be a valid Emacs fixnum. */
3560 #define INDEX_SIZE_BOUND \
3561 ((ptrdiff_t) min (MOST_POSITIVE_FIXNUM, PTRDIFF_MAX / word_size))
3562
3563 /* Create and initialize a new hash table.
3564
3565 TEST specifies the test the hash table will use to compare keys.
3566 It must be either one of the predefined tests `eq', `eql' or
3567 `equal' or a symbol denoting a user-defined test named TEST with
3568 test and hash functions USER_TEST and USER_HASH.
3569
3570 Give the table initial capacity SIZE, SIZE >= 0, an integer.
3571
3572 If REHASH_SIZE is an integer, it must be > 0, and this hash table's
3573 new size when it becomes full is computed by adding REHASH_SIZE to
3574 its old size. If REHASH_SIZE is a float, it must be > 1.0, and the
3575 table's new size is computed by multiplying its old size with
3576 REHASH_SIZE.
3577
3578 REHASH_THRESHOLD must be a float <= 1.0, and > 0. The table will
3579 be resized when the ratio of (number of entries in the table) /
3580 (table size) is >= REHASH_THRESHOLD.
3581
3582 WEAK specifies the weakness of the table. If non-nil, it must be
3583 one of the symbols `key', `value', `key-or-value', or `key-and-value'. */
3584
3585 Lisp_Object
3586 make_hash_table (struct hash_table_test test,
3587 Lisp_Object size, Lisp_Object rehash_size,
3588 Lisp_Object rehash_threshold, Lisp_Object weak)
3589 {
3590 struct Lisp_Hash_Table *h;
3591 Lisp_Object table;
3592 EMACS_INT index_size, sz;
3593 ptrdiff_t i;
3594 double index_float;
3595
3596 /* Preconditions. */
3597 eassert (SYMBOLP (test.name));
3598 eassert (INTEGERP (size) && XINT (size) >= 0);
3599 eassert ((INTEGERP (rehash_size) && XINT (rehash_size) > 0)
3600 || (FLOATP (rehash_size) && 1 < XFLOAT_DATA (rehash_size)));
3601 eassert (FLOATP (rehash_threshold)
3602 && 0 < XFLOAT_DATA (rehash_threshold)
3603 && XFLOAT_DATA (rehash_threshold) <= 1.0);
3604
3605 if (XFASTINT (size) == 0)
3606 size = make_number (1);
3607
3608 sz = XFASTINT (size);
3609 index_float = sz / XFLOAT_DATA (rehash_threshold);
3610 index_size = (index_float < INDEX_SIZE_BOUND + 1
3611 ? next_almost_prime (index_float)
3612 : INDEX_SIZE_BOUND + 1);
3613 if (INDEX_SIZE_BOUND < max (index_size, 2 * sz))
3614 error ("Hash table too large");
3615
3616 /* Allocate a table and initialize it. */
3617 h = allocate_hash_table ();
3618
3619 /* Initialize hash table slots. */
3620 h->test = test;
3621 h->weak = weak;
3622 h->rehash_threshold = rehash_threshold;
3623 h->rehash_size = rehash_size;
3624 h->count = 0;
3625 h->key_and_value = Fmake_vector (make_number (2 * sz), Qnil);
3626 h->hash = Fmake_vector (size, Qnil);
3627 h->next = Fmake_vector (size, Qnil);
3628 h->index = Fmake_vector (make_number (index_size), Qnil);
3629
3630 /* Set up the free list. */
3631 for (i = 0; i < sz - 1; ++i)
3632 set_hash_next_slot (h, i, make_number (i + 1));
3633 h->next_free = make_number (0);
3634
3635 XSET_HASH_TABLE (table, h);
3636 eassert (HASH_TABLE_P (table));
3637 eassert (XHASH_TABLE (table) == h);
3638
3639 return table;
3640 }
3641
3642
3643 /* Return a copy of hash table H1. Keys and values are not copied,
3644 only the table itself is. */
3645
3646 static Lisp_Object
3647 copy_hash_table (struct Lisp_Hash_Table *h1)
3648 {
3649 Lisp_Object table;
3650 struct Lisp_Hash_Table *h2;
3651
3652 h2 = allocate_hash_table ();
3653 *h2 = *h1;
3654 h2->key_and_value = Fcopy_sequence (h1->key_and_value);
3655 h2->hash = Fcopy_sequence (h1->hash);
3656 h2->next = Fcopy_sequence (h1->next);
3657 h2->index = Fcopy_sequence (h1->index);
3658 XSET_HASH_TABLE (table, h2);
3659
3660 return table;
3661 }
3662
3663
3664 /* Resize hash table H if it's too full. If H cannot be resized
3665 because it's already too large, throw an error. */
3666
3667 static void
3668 maybe_resize_hash_table (struct Lisp_Hash_Table *h)
3669 {
3670 if (NILP (h->next_free))
3671 {
3672 ptrdiff_t old_size = HASH_TABLE_SIZE (h);
3673 EMACS_INT new_size, index_size, nsize;
3674 ptrdiff_t i;
3675 double index_float;
3676
3677 if (INTEGERP (h->rehash_size))
3678 new_size = old_size + XFASTINT (h->rehash_size);
3679 else
3680 {
3681 double float_new_size = old_size * XFLOAT_DATA (h->rehash_size);
3682 if (float_new_size < INDEX_SIZE_BOUND + 1)
3683 {
3684 new_size = float_new_size;
3685 if (new_size <= old_size)
3686 new_size = old_size + 1;
3687 }
3688 else
3689 new_size = INDEX_SIZE_BOUND + 1;
3690 }
3691 index_float = new_size / XFLOAT_DATA (h->rehash_threshold);
3692 index_size = (index_float < INDEX_SIZE_BOUND + 1
3693 ? next_almost_prime (index_float)
3694 : INDEX_SIZE_BOUND + 1);
3695 nsize = max (index_size, 2 * new_size);
3696 if (INDEX_SIZE_BOUND < nsize)
3697 error ("Hash table too large to resize");
3698
3699 #ifdef ENABLE_CHECKING
3700 if (HASH_TABLE_P (Vpurify_flag)
3701 && XHASH_TABLE (Vpurify_flag) == h)
3702 {
3703 Lisp_Object args[2];
3704 args[0] = build_string ("Growing hash table to: %d");
3705 args[1] = make_number (new_size);
3706 Fmessage (2, args);
3707 }
3708 #endif
3709
3710 set_hash_key_and_value (h, larger_vector (h->key_and_value,
3711 2 * (new_size - old_size), -1));
3712 set_hash_next (h, larger_vector (h->next, new_size - old_size, -1));
3713 set_hash_hash (h, larger_vector (h->hash, new_size - old_size, -1));
3714 set_hash_index (h, Fmake_vector (make_number (index_size), Qnil));
3715
3716 /* Update the free list. Do it so that new entries are added at
3717 the end of the free list. This makes some operations like
3718 maphash faster. */
3719 for (i = old_size; i < new_size - 1; ++i)
3720 set_hash_next_slot (h, i, make_number (i + 1));
3721
3722 if (!NILP (h->next_free))
3723 {
3724 Lisp_Object last, next;
3725
3726 last = h->next_free;
3727 while (next = HASH_NEXT (h, XFASTINT (last)),
3728 !NILP (next))
3729 last = next;
3730
3731 set_hash_next_slot (h, XFASTINT (last), make_number (old_size));
3732 }
3733 else
3734 XSETFASTINT (h->next_free, old_size);
3735
3736 /* Rehash. */
3737 for (i = 0; i < old_size; ++i)
3738 if (!NILP (HASH_HASH (h, i)))
3739 {
3740 EMACS_UINT hash_code = XUINT (HASH_HASH (h, i));
3741 ptrdiff_t start_of_bucket = hash_code % ASIZE (h->index);
3742 set_hash_next_slot (h, i, HASH_INDEX (h, start_of_bucket));
3743 set_hash_index_slot (h, start_of_bucket, make_number (i));
3744 }
3745 }
3746 }
3747
3748
3749 /* Lookup KEY in hash table H. If HASH is non-null, return in *HASH
3750 the hash code of KEY. Value is the index of the entry in H
3751 matching KEY, or -1 if not found. */
3752
3753 ptrdiff_t
3754 hash_lookup (struct Lisp_Hash_Table *h, Lisp_Object key, EMACS_UINT *hash)
3755 {
3756 EMACS_UINT hash_code;
3757 ptrdiff_t start_of_bucket;
3758 Lisp_Object idx;
3759
3760 hash_code = h->test.hashfn (&h->test, key);
3761 eassert ((hash_code & ~INTMASK) == 0);
3762 if (hash)
3763 *hash = hash_code;
3764
3765 start_of_bucket = hash_code % ASIZE (h->index);
3766 idx = HASH_INDEX (h, start_of_bucket);
3767
3768 /* We need not gcpro idx since it's either an integer or nil. */
3769 while (!NILP (idx))
3770 {
3771 ptrdiff_t i = XFASTINT (idx);
3772 if (EQ (key, HASH_KEY (h, i))
3773 || (h->test.cmpfn
3774 && hash_code == XUINT (HASH_HASH (h, i))
3775 && h->test.cmpfn (&h->test, key, HASH_KEY (h, i))))
3776 break;
3777 idx = HASH_NEXT (h, i);
3778 }
3779
3780 return NILP (idx) ? -1 : XFASTINT (idx);
3781 }
3782
3783
3784 /* Put an entry into hash table H that associates KEY with VALUE.
3785 HASH is a previously computed hash code of KEY.
3786 Value is the index of the entry in H matching KEY. */
3787
3788 ptrdiff_t
3789 hash_put (struct Lisp_Hash_Table *h, Lisp_Object key, Lisp_Object value,
3790 EMACS_UINT hash)
3791 {
3792 ptrdiff_t start_of_bucket, i;
3793
3794 eassert ((hash & ~INTMASK) == 0);
3795
3796 /* Increment count after resizing because resizing may fail. */
3797 maybe_resize_hash_table (h);
3798 h->count++;
3799
3800 /* Store key/value in the key_and_value vector. */
3801 i = XFASTINT (h->next_free);
3802 h->next_free = HASH_NEXT (h, i);
3803 set_hash_key_slot (h, i, key);
3804 set_hash_value_slot (h, i, value);
3805
3806 /* Remember its hash code. */
3807 set_hash_hash_slot (h, i, make_number (hash));
3808
3809 /* Add new entry to its collision chain. */
3810 start_of_bucket = hash % ASIZE (h->index);
3811 set_hash_next_slot (h, i, HASH_INDEX (h, start_of_bucket));
3812 set_hash_index_slot (h, start_of_bucket, make_number (i));
3813 return i;
3814 }
3815
3816
3817 /* Remove the entry matching KEY from hash table H, if there is one. */
3818
3819 static void
3820 hash_remove_from_table (struct Lisp_Hash_Table *h, Lisp_Object key)
3821 {
3822 EMACS_UINT hash_code;
3823 ptrdiff_t start_of_bucket;
3824 Lisp_Object idx, prev;
3825
3826 hash_code = h->test.hashfn (&h->test, key);
3827 eassert ((hash_code & ~INTMASK) == 0);
3828 start_of_bucket = hash_code % ASIZE (h->index);
3829 idx = HASH_INDEX (h, start_of_bucket);
3830 prev = Qnil;
3831
3832 /* We need not gcpro idx, prev since they're either integers or nil. */
3833 while (!NILP (idx))
3834 {
3835 ptrdiff_t i = XFASTINT (idx);
3836
3837 if (EQ (key, HASH_KEY (h, i))
3838 || (h->test.cmpfn
3839 && hash_code == XUINT (HASH_HASH (h, i))
3840 && h->test.cmpfn (&h->test, key, HASH_KEY (h, i))))
3841 {
3842 /* Take entry out of collision chain. */
3843 if (NILP (prev))
3844 set_hash_index_slot (h, start_of_bucket, HASH_NEXT (h, i));
3845 else
3846 set_hash_next_slot (h, XFASTINT (prev), HASH_NEXT (h, i));
3847
3848 /* Clear slots in key_and_value and add the slots to
3849 the free list. */
3850 set_hash_key_slot (h, i, Qnil);
3851 set_hash_value_slot (h, i, Qnil);
3852 set_hash_hash_slot (h, i, Qnil);
3853 set_hash_next_slot (h, i, h->next_free);
3854 h->next_free = make_number (i);
3855 h->count--;
3856 eassert (h->count >= 0);
3857 break;
3858 }
3859 else
3860 {
3861 prev = idx;
3862 idx = HASH_NEXT (h, i);
3863 }
3864 }
3865 }
3866
3867
3868 /* Clear hash table H. */
3869
3870 static void
3871 hash_clear (struct Lisp_Hash_Table *h)
3872 {
3873 if (h->count > 0)
3874 {
3875 ptrdiff_t i, size = HASH_TABLE_SIZE (h);
3876
3877 for (i = 0; i < size; ++i)
3878 {
3879 set_hash_next_slot (h, i, i < size - 1 ? make_number (i + 1) : Qnil);
3880 set_hash_key_slot (h, i, Qnil);
3881 set_hash_value_slot (h, i, Qnil);
3882 set_hash_hash_slot (h, i, Qnil);
3883 }
3884
3885 for (i = 0; i < ASIZE (h->index); ++i)
3886 ASET (h->index, i, Qnil);
3887
3888 h->next_free = make_number (0);
3889 h->count = 0;
3890 }
3891 }
3892
3893
3894 \f
3895 /***********************************************************************
3896 Hash Code Computation
3897 ***********************************************************************/
3898
3899 /* Return a hash for string PTR which has length LEN. The hash value
3900 can be any EMACS_UINT value. */
3901
3902 EMACS_UINT
3903 hash_string (char const *ptr, ptrdiff_t len)
3904 {
3905 char const *p = ptr;
3906 char const *end = p + len;
3907 unsigned char c;
3908 EMACS_UINT hash = 0;
3909
3910 while (p != end)
3911 {
3912 c = *p++;
3913 hash = sxhash_combine (hash, c);
3914 }
3915
3916 return hash;
3917 }
3918
3919 /* Return a hash code for OBJ. DEPTH is the current depth in the Lisp
3920 structure. Value is an unsigned integer clipped to INTMASK. */
3921
3922 EMACS_UINT
3923 sxhash (Lisp_Object obj, int depth)
3924 {
3925 return scm_ihash (obj, MOST_POSITIVE_FIXNUM);
3926 }
3927
3928
3929 \f
3930 /***********************************************************************
3931 Lisp Interface
3932 ***********************************************************************/
3933
3934
3935 DEFUN ("sxhash", Fsxhash, Ssxhash, 1, 1, 0,
3936 doc: /* Compute a hash code for OBJ and return it as integer. */)
3937 (Lisp_Object obj)
3938 {
3939 EMACS_UINT hash = sxhash (obj, 0);
3940 return make_number (hash);
3941 }
3942
3943
3944 DEFUN ("make-hash-table", Fmake_hash_table, Smake_hash_table, 0, MANY, 0,
3945 doc: /* Create and return a new hash table.
3946
3947 Arguments are specified as keyword/argument pairs. The following
3948 arguments are defined:
3949
3950 :test TEST -- TEST must be a symbol that specifies how to compare
3951 keys. Default is `eql'. Predefined are the tests `eq', `eql', and
3952 `equal'. User-supplied test and hash functions can be specified via
3953 `define-hash-table-test'.
3954
3955 :size SIZE -- A hint as to how many elements will be put in the table.
3956 Default is 65.
3957
3958 :rehash-size REHASH-SIZE - Indicates how to expand the table when it
3959 fills up. If REHASH-SIZE is an integer, increase the size by that
3960 amount. If it is a float, it must be > 1.0, and the new size is the
3961 old size multiplied by that factor. Default is 1.5.
3962
3963 :rehash-threshold THRESHOLD -- THRESHOLD must a float > 0, and <= 1.0.
3964 Resize the hash table when the ratio (number of entries / table size)
3965 is greater than or equal to THRESHOLD. Default is 0.8.
3966
3967 :weakness WEAK -- WEAK must be one of nil, t, `key', `value',
3968 `key-or-value', or `key-and-value'. If WEAK is not nil, the table
3969 returned is a weak table. Key/value pairs are removed from a weak
3970 hash table when there are no non-weak references pointing to their
3971 key, value, one of key or value, or both key and value, depending on
3972 WEAK. WEAK t is equivalent to `key-and-value'. Default value of WEAK
3973 is nil.
3974
3975 usage: (make-hash-table &rest KEYWORD-ARGS) */)
3976 (ptrdiff_t nargs, Lisp_Object *args)
3977 {
3978 Lisp_Object test, size, rehash_size, rehash_threshold, weak;
3979 struct hash_table_test testdesc;
3980 char *used;
3981 ptrdiff_t i;
3982
3983 /* The vector `used' is used to keep track of arguments that
3984 have been consumed. */
3985 used = alloca (nargs * sizeof *used);
3986 memset (used, 0, nargs * sizeof *used);
3987
3988 /* See if there's a `:test TEST' among the arguments. */
3989 i = get_key_arg (QCtest, nargs, args, used);
3990 test = i ? args[i] : Qeql;
3991 if (EQ (test, Qeq))
3992 testdesc = hashtest_eq;
3993 else if (EQ (test, Qeql))
3994 testdesc = hashtest_eql;
3995 else if (EQ (test, Qequal))
3996 testdesc = hashtest_equal;
3997 else
3998 {
3999 /* See if it is a user-defined test. */
4000 Lisp_Object prop;
4001
4002 prop = Fget (test, Qhash_table_test);
4003 if (!CONSP (prop) || !CONSP (XCDR (prop)))
4004 signal_error ("Invalid hash table test", test);
4005 testdesc.name = test;
4006 testdesc.user_cmp_function = XCAR (prop);
4007 testdesc.user_hash_function = XCAR (XCDR (prop));
4008 testdesc.hashfn = hashfn_user_defined;
4009 testdesc.cmpfn = cmpfn_user_defined;
4010 }
4011
4012 /* See if there's a `:size SIZE' argument. */
4013 i = get_key_arg (QCsize, nargs, args, used);
4014 size = i ? args[i] : Qnil;
4015 if (NILP (size))
4016 size = make_number (DEFAULT_HASH_SIZE);
4017 else if (!INTEGERP (size) || XINT (size) < 0)
4018 signal_error ("Invalid hash table size", size);
4019
4020 /* Look for `:rehash-size SIZE'. */
4021 i = get_key_arg (QCrehash_size, nargs, args, used);
4022 rehash_size = i ? args[i] : make_float (DEFAULT_REHASH_SIZE);
4023 if (! ((INTEGERP (rehash_size) && 0 < XINT (rehash_size))
4024 || (FLOATP (rehash_size) && 1 < XFLOAT_DATA (rehash_size))))
4025 signal_error ("Invalid hash table rehash size", rehash_size);
4026
4027 /* Look for `:rehash-threshold THRESHOLD'. */
4028 i = get_key_arg (QCrehash_threshold, nargs, args, used);
4029 rehash_threshold = i ? args[i] : make_float (DEFAULT_REHASH_THRESHOLD);
4030 if (! (FLOATP (rehash_threshold)
4031 && 0 < XFLOAT_DATA (rehash_threshold)
4032 && XFLOAT_DATA (rehash_threshold) <= 1))
4033 signal_error ("Invalid hash table rehash threshold", rehash_threshold);
4034
4035 /* Look for `:weakness WEAK'. */
4036 i = get_key_arg (QCweakness, nargs, args, used);
4037 weak = i ? args[i] : Qnil;
4038 if (EQ (weak, Qt))
4039 weak = Qkey_and_value;
4040 if (!NILP (weak)
4041 && !EQ (weak, Qkey)
4042 && !EQ (weak, Qvalue)
4043 && !EQ (weak, Qkey_or_value)
4044 && !EQ (weak, Qkey_and_value))
4045 signal_error ("Invalid hash table weakness", weak);
4046
4047 /* Now, all args should have been used up, or there's a problem. */
4048 for (i = 0; i < nargs; ++i)
4049 if (!used[i])
4050 signal_error ("Invalid argument list", args[i]);
4051
4052 return make_hash_table (testdesc, size, rehash_size, rehash_threshold, weak);
4053 }
4054
4055
4056 DEFUN ("copy-hash-table", Fcopy_hash_table, Scopy_hash_table, 1, 1, 0,
4057 doc: /* Return a copy of hash table TABLE. */)
4058 (Lisp_Object table)
4059 {
4060 return copy_hash_table (check_hash_table (table));
4061 }
4062
4063
4064 DEFUN ("hash-table-count", Fhash_table_count, Shash_table_count, 1, 1, 0,
4065 doc: /* Return the number of elements in TABLE. */)
4066 (Lisp_Object table)
4067 {
4068 return make_number (check_hash_table (table)->count);
4069 }
4070
4071
4072 DEFUN ("hash-table-rehash-size", Fhash_table_rehash_size,
4073 Shash_table_rehash_size, 1, 1, 0,
4074 doc: /* Return the current rehash size of TABLE. */)
4075 (Lisp_Object table)
4076 {
4077 return check_hash_table (table)->rehash_size;
4078 }
4079
4080
4081 DEFUN ("hash-table-rehash-threshold", Fhash_table_rehash_threshold,
4082 Shash_table_rehash_threshold, 1, 1, 0,
4083 doc: /* Return the current rehash threshold of TABLE. */)
4084 (Lisp_Object table)
4085 {
4086 return check_hash_table (table)->rehash_threshold;
4087 }
4088
4089
4090 DEFUN ("hash-table-size", Fhash_table_size, Shash_table_size, 1, 1, 0,
4091 doc: /* Return the size of TABLE.
4092 The size can be used as an argument to `make-hash-table' to create
4093 a hash table than can hold as many elements as TABLE holds
4094 without need for resizing. */)
4095 (Lisp_Object table)
4096 {
4097 struct Lisp_Hash_Table *h = check_hash_table (table);
4098 return make_number (HASH_TABLE_SIZE (h));
4099 }
4100
4101
4102 DEFUN ("hash-table-test", Fhash_table_test, Shash_table_test, 1, 1, 0,
4103 doc: /* Return the test TABLE uses. */)
4104 (Lisp_Object table)
4105 {
4106 return check_hash_table (table)->test.name;
4107 }
4108
4109
4110 DEFUN ("hash-table-weakness", Fhash_table_weakness, Shash_table_weakness,
4111 1, 1, 0,
4112 doc: /* Return the weakness of TABLE. */)
4113 (Lisp_Object table)
4114 {
4115 return check_hash_table (table)->weak;
4116 }
4117
4118
4119 DEFUN ("hash-table-p", Fhash_table_p, Shash_table_p, 1, 1, 0,
4120 doc: /* Return t if OBJ is a Lisp hash table object. */)
4121 (Lisp_Object obj)
4122 {
4123 return HASH_TABLE_P (obj) ? Qt : Qnil;
4124 }
4125
4126
4127 DEFUN ("clrhash", Fclrhash, Sclrhash, 1, 1, 0,
4128 doc: /* Clear hash table TABLE and return it. */)
4129 (Lisp_Object table)
4130 {
4131 hash_clear (check_hash_table (table));
4132 /* Be compatible with XEmacs. */
4133 return table;
4134 }
4135
4136
4137 DEFUN ("gethash", Fgethash, Sgethash, 2, 3, 0,
4138 doc: /* Look up KEY in TABLE and return its associated value.
4139 If KEY is not found, return DFLT which defaults to nil. */)
4140 (Lisp_Object key, Lisp_Object table, Lisp_Object dflt)
4141 {
4142 struct Lisp_Hash_Table *h = check_hash_table (table);
4143 ptrdiff_t i = hash_lookup (h, key, NULL);
4144 return i >= 0 ? HASH_VALUE (h, i) : dflt;
4145 }
4146
4147
4148 DEFUN ("puthash", Fputhash, Sputhash, 3, 3, 0,
4149 doc: /* Associate KEY with VALUE in hash table TABLE.
4150 If KEY is already present in table, replace its current value with
4151 VALUE. In any case, return VALUE. */)
4152 (Lisp_Object key, Lisp_Object value, Lisp_Object table)
4153 {
4154 struct Lisp_Hash_Table *h = check_hash_table (table);
4155 ptrdiff_t i;
4156 EMACS_UINT hash;
4157
4158 i = hash_lookup (h, key, &hash);
4159 if (i >= 0)
4160 set_hash_value_slot (h, i, value);
4161 else
4162 hash_put (h, key, value, hash);
4163
4164 return value;
4165 }
4166
4167
4168 DEFUN ("remhash", Fremhash, Sremhash, 2, 2, 0,
4169 doc: /* Remove KEY from TABLE. */)
4170 (Lisp_Object key, Lisp_Object table)
4171 {
4172 struct Lisp_Hash_Table *h = check_hash_table (table);
4173 hash_remove_from_table (h, key);
4174 return Qnil;
4175 }
4176
4177
4178 DEFUN ("maphash", Fmaphash, Smaphash, 2, 2, 0,
4179 doc: /* Call FUNCTION for all entries in hash table TABLE.
4180 FUNCTION is called with two arguments, KEY and VALUE.
4181 `maphash' always returns nil. */)
4182 (Lisp_Object function, Lisp_Object table)
4183 {
4184 struct Lisp_Hash_Table *h = check_hash_table (table);
4185 Lisp_Object args[3];
4186 ptrdiff_t i;
4187
4188 for (i = 0; i < HASH_TABLE_SIZE (h); ++i)
4189 if (!NILP (HASH_HASH (h, i)))
4190 {
4191 args[0] = function;
4192 args[1] = HASH_KEY (h, i);
4193 args[2] = HASH_VALUE (h, i);
4194 Ffuncall (3, args);
4195 }
4196
4197 return Qnil;
4198 }
4199
4200
4201 DEFUN ("define-hash-table-test", Fdefine_hash_table_test,
4202 Sdefine_hash_table_test, 3, 3, 0,
4203 doc: /* Define a new hash table test with name NAME, a symbol.
4204
4205 In hash tables created with NAME specified as test, use TEST to
4206 compare keys, and HASH for computing hash codes of keys.
4207
4208 TEST must be a function taking two arguments and returning non-nil if
4209 both arguments are the same. HASH must be a function taking one
4210 argument and returning an object that is the hash code of the argument.
4211 It should be the case that if (eq (funcall HASH x1) (funcall HASH x2))
4212 returns nil, then (funcall TEST x1 x2) also returns nil. */)
4213 (Lisp_Object name, Lisp_Object test, Lisp_Object hash)
4214 {
4215 return Fput (name, Qhash_table_test, list2 (test, hash));
4216 }
4217
4218
4219 \f
4220 /************************************************************************
4221 MD5, SHA-1, and SHA-2
4222 ************************************************************************/
4223
4224 #include "md5.h"
4225 #include "sha1.h"
4226 #include "sha256.h"
4227 #include "sha512.h"
4228
4229 /* ALGORITHM is a symbol: md5, sha1, sha224 and so on. */
4230
4231 static Lisp_Object
4232 secure_hash (Lisp_Object algorithm, Lisp_Object object, Lisp_Object start,
4233 Lisp_Object end, Lisp_Object coding_system, Lisp_Object noerror,
4234 Lisp_Object binary)
4235 {
4236 int i;
4237 ptrdiff_t size, start_char = 0, start_byte, end_char = 0, end_byte;
4238 register EMACS_INT b, e;
4239 register struct buffer *bp;
4240 EMACS_INT temp;
4241 int digest_size;
4242 void *(*hash_func) (const char *, size_t, void *);
4243 Lisp_Object digest;
4244
4245 CHECK_SYMBOL (algorithm);
4246
4247 if (STRINGP (object))
4248 {
4249 if (NILP (coding_system))
4250 {
4251 /* Decide the coding-system to encode the data with. */
4252
4253 if (STRING_MULTIBYTE (object))
4254 /* use default, we can't guess correct value */
4255 coding_system = preferred_coding_system ();
4256 else
4257 coding_system = Qraw_text;
4258 }
4259
4260 if (NILP (Fcoding_system_p (coding_system)))
4261 {
4262 /* Invalid coding system. */
4263
4264 if (!NILP (noerror))
4265 coding_system = Qraw_text;
4266 else
4267 xsignal1 (Qcoding_system_error, coding_system);
4268 }
4269
4270 if (STRING_MULTIBYTE (object))
4271 object = code_convert_string (object, coding_system, Qnil, 1, 0, 1);
4272
4273 size = SCHARS (object);
4274 validate_subarray (object, start, end, size, &start_char, &end_char);
4275
4276 start_byte = !start_char ? 0 : string_char_to_byte (object, start_char);
4277 end_byte = (end_char == size
4278 ? SBYTES (object)
4279 : string_char_to_byte (object, end_char));
4280 }
4281 else
4282 {
4283 ptrdiff_t count = SPECPDL_INDEX ();
4284
4285 record_unwind_current_buffer ();
4286
4287 CHECK_BUFFER (object);
4288
4289 bp = XBUFFER (object);
4290 set_buffer_internal (bp);
4291
4292 if (NILP (start))
4293 b = BEGV;
4294 else
4295 {
4296 CHECK_NUMBER_COERCE_MARKER (start);
4297 b = XINT (start);
4298 }
4299
4300 if (NILP (end))
4301 e = ZV;
4302 else
4303 {
4304 CHECK_NUMBER_COERCE_MARKER (end);
4305 e = XINT (end);
4306 }
4307
4308 if (b > e)
4309 temp = b, b = e, e = temp;
4310
4311 if (!(BEGV <= b && e <= ZV))
4312 args_out_of_range (start, end);
4313
4314 if (NILP (coding_system))
4315 {
4316 /* Decide the coding-system to encode the data with.
4317 See fileio.c:Fwrite-region */
4318
4319 if (!NILP (Vcoding_system_for_write))
4320 coding_system = Vcoding_system_for_write;
4321 else
4322 {
4323 bool force_raw_text = 0;
4324
4325 coding_system = BVAR (XBUFFER (object), buffer_file_coding_system);
4326 if (NILP (coding_system)
4327 || NILP (Flocal_variable_p (Qbuffer_file_coding_system, Qnil)))
4328 {
4329 coding_system = Qnil;
4330 if (NILP (BVAR (current_buffer, enable_multibyte_characters)))
4331 force_raw_text = 1;
4332 }
4333
4334 if (NILP (coding_system) && !NILP (Fbuffer_file_name (object)))
4335 {
4336 /* Check file-coding-system-alist. */
4337 Lisp_Object args[4], val;
4338
4339 args[0] = Qwrite_region; args[1] = start; args[2] = end;
4340 args[3] = Fbuffer_file_name (object);
4341 val = Ffind_operation_coding_system (4, args);
4342 if (CONSP (val) && !NILP (XCDR (val)))
4343 coding_system = XCDR (val);
4344 }
4345
4346 if (NILP (coding_system)
4347 && !NILP (BVAR (XBUFFER (object), buffer_file_coding_system)))
4348 {
4349 /* If we still have not decided a coding system, use the
4350 default value of buffer-file-coding-system. */
4351 coding_system = BVAR (XBUFFER (object), buffer_file_coding_system);
4352 }
4353
4354 if (!force_raw_text
4355 && !NILP (Ffboundp (Vselect_safe_coding_system_function)))
4356 /* Confirm that VAL can surely encode the current region. */
4357 coding_system = call4 (Vselect_safe_coding_system_function,
4358 make_number (b), make_number (e),
4359 coding_system, Qnil);
4360
4361 if (force_raw_text)
4362 coding_system = Qraw_text;
4363 }
4364
4365 if (NILP (Fcoding_system_p (coding_system)))
4366 {
4367 /* Invalid coding system. */
4368
4369 if (!NILP (noerror))
4370 coding_system = Qraw_text;
4371 else
4372 xsignal1 (Qcoding_system_error, coding_system);
4373 }
4374 }
4375
4376 object = make_buffer_string (b, e, 0);
4377 unbind_to (count, Qnil);
4378
4379 if (STRING_MULTIBYTE (object))
4380 object = code_convert_string (object, coding_system, Qnil, 1, 0, 0);
4381 start_byte = 0;
4382 end_byte = SBYTES (object);
4383 }
4384
4385 if (EQ (algorithm, Qmd5))
4386 {
4387 digest_size = MD5_DIGEST_SIZE;
4388 hash_func = md5_buffer;
4389 }
4390 else if (EQ (algorithm, Qsha1))
4391 {
4392 digest_size = SHA1_DIGEST_SIZE;
4393 hash_func = sha1_buffer;
4394 }
4395 else if (EQ (algorithm, Qsha224))
4396 {
4397 digest_size = SHA224_DIGEST_SIZE;
4398 hash_func = sha224_buffer;
4399 }
4400 else if (EQ (algorithm, Qsha256))
4401 {
4402 digest_size = SHA256_DIGEST_SIZE;
4403 hash_func = sha256_buffer;
4404 }
4405 else if (EQ (algorithm, Qsha384))
4406 {
4407 digest_size = SHA384_DIGEST_SIZE;
4408 hash_func = sha384_buffer;
4409 }
4410 else if (EQ (algorithm, Qsha512))
4411 {
4412 digest_size = SHA512_DIGEST_SIZE;
4413 hash_func = sha512_buffer;
4414 }
4415 else
4416 error ("Invalid algorithm arg: %s", SDATA (Fsymbol_name (algorithm)));
4417
4418 /* allocate 2 x digest_size so that it can be re-used to hold the
4419 hexified value */
4420 digest = make_uninit_string (digest_size * 2);
4421
4422 hash_func (SSDATA (object) + start_byte,
4423 end_byte - start_byte,
4424 SSDATA (digest));
4425
4426 if (NILP (binary))
4427 {
4428 unsigned char *p = SDATA (digest);
4429 for (i = digest_size - 1; i >= 0; i--)
4430 {
4431 static char const hexdigit[16] = "0123456789abcdef";
4432 int p_i = p[i];
4433 p[2 * i] = hexdigit[p_i >> 4];
4434 p[2 * i + 1] = hexdigit[p_i & 0xf];
4435 }
4436 return digest;
4437 }
4438 else
4439 return make_unibyte_string (SSDATA (digest), digest_size);
4440 }
4441
4442 DEFUN ("md5", Fmd5, Smd5, 1, 5, 0,
4443 doc: /* Return MD5 message digest of OBJECT, a buffer or string.
4444
4445 A message digest is a cryptographic checksum of a document, and the
4446 algorithm to calculate it is defined in RFC 1321.
4447
4448 The two optional arguments START and END are character positions
4449 specifying for which part of OBJECT the message digest should be
4450 computed. If nil or omitted, the digest is computed for the whole
4451 OBJECT.
4452
4453 The MD5 message digest is computed from the result of encoding the
4454 text in a coding system, not directly from the internal Emacs form of
4455 the text. The optional fourth argument CODING-SYSTEM specifies which
4456 coding system to encode the text with. It should be the same coding
4457 system that you used or will use when actually writing the text into a
4458 file.
4459
4460 If CODING-SYSTEM is nil or omitted, the default depends on OBJECT. If
4461 OBJECT is a buffer, the default for CODING-SYSTEM is whatever coding
4462 system would be chosen by default for writing this text into a file.
4463
4464 If OBJECT is a string, the most preferred coding system (see the
4465 command `prefer-coding-system') is used.
4466
4467 If NOERROR is non-nil, silently assume the `raw-text' coding if the
4468 guesswork fails. Normally, an error is signaled in such case. */)
4469 (Lisp_Object object, Lisp_Object start, Lisp_Object end, Lisp_Object coding_system, Lisp_Object noerror)
4470 {
4471 return secure_hash (Qmd5, object, start, end, coding_system, noerror, Qnil);
4472 }
4473
4474 DEFUN ("secure-hash", Fsecure_hash, Ssecure_hash, 2, 5, 0,
4475 doc: /* Return the secure hash of OBJECT, a buffer or string.
4476 ALGORITHM is a symbol specifying the hash to use:
4477 md5, sha1, sha224, sha256, sha384 or sha512.
4478
4479 The two optional arguments START and END are positions specifying for
4480 which part of OBJECT to compute the hash. If nil or omitted, uses the
4481 whole OBJECT.
4482
4483 If BINARY is non-nil, returns a string in binary form. */)
4484 (Lisp_Object algorithm, Lisp_Object object, Lisp_Object start, Lisp_Object end, Lisp_Object binary)
4485 {
4486 return secure_hash (algorithm, object, start, end, Qnil, Qnil, binary);
4487 }
4488 \f
4489 void
4490 init_fns_once (void)
4491 {
4492 compare_text_properties = scm_make_fluid ();
4493 scm_set_smob_equalp (lisp_misc_tag, misc_equal_p);
4494 scm_set_smob_equalp (lisp_string_tag, string_equal_p);
4495 scm_set_smob_equalp (lisp_vectorlike_tag, vectorlike_equal_p);
4496 }
4497
4498 void
4499 syms_of_fns (void)
4500 {
4501 #include "fns.x"
4502
4503 DEFSYM (Qmd5, "md5");
4504 DEFSYM (Qsha1, "sha1");
4505 DEFSYM (Qsha224, "sha224");
4506 DEFSYM (Qsha256, "sha256");
4507 DEFSYM (Qsha384, "sha384");
4508 DEFSYM (Qsha512, "sha512");
4509
4510 /* Hash table stuff. */
4511 DEFSYM (Qhash_table_p, "hash-table-p");
4512 DEFSYM (Qeq, "eq");
4513 DEFSYM (Qeql, "eql");
4514 DEFSYM (Qequal, "equal");
4515 DEFSYM (QCtest, ":test");
4516 DEFSYM (QCsize, ":size");
4517 DEFSYM (QCrehash_size, ":rehash-size");
4518 DEFSYM (QCrehash_threshold, ":rehash-threshold");
4519 DEFSYM (QCweakness, ":weakness");
4520 DEFSYM (Qkey, "key");
4521 DEFSYM (Qvalue, "value");
4522 DEFSYM (Qhash_table_test, "hash-table-test");
4523 DEFSYM (Qkey_or_value, "key-or-value");
4524 DEFSYM (Qkey_and_value, "key-and-value");
4525
4526 DEFSYM (Qstring_lessp, "string-lessp");
4527 DEFSYM (Qprovide, "provide");
4528 DEFSYM (Qrequire, "require");
4529 DEFSYM (Qyes_or_no_p_history, "yes-or-no-p-history");
4530 DEFSYM (Qcursor_in_echo_area, "cursor-in-echo-area");
4531 DEFSYM (Qwidget_type, "widget-type");
4532
4533 staticpro (&string_char_byte_cache_string);
4534 string_char_byte_cache_string = Qnil;
4535
4536 require_nesting_list = Qnil;
4537 staticpro (&require_nesting_list);
4538
4539 Fset (Qyes_or_no_p_history, Qnil);
4540
4541 DEFVAR_LISP ("features", Vfeatures,
4542 doc: /* A list of symbols which are the features of the executing Emacs.
4543 Used by `featurep' and `require', and altered by `provide'. */);
4544 Vfeatures = list1 (intern_c_string ("emacs"));
4545 DEFSYM (Qsubfeatures, "subfeatures");
4546 DEFSYM (Qfuncall, "funcall");
4547
4548 #ifdef HAVE_LANGINFO_CODESET
4549 DEFSYM (Qcodeset, "codeset");
4550 DEFSYM (Qdays, "days");
4551 DEFSYM (Qmonths, "months");
4552 DEFSYM (Qpaper, "paper");
4553 #endif /* HAVE_LANGINFO_CODESET */
4554
4555 DEFVAR_BOOL ("use-dialog-box", use_dialog_box,
4556 doc: /* Non-nil means mouse commands use dialog boxes to ask questions.
4557 This applies to `y-or-n-p' and `yes-or-no-p' questions asked by commands
4558 invoked by mouse clicks and mouse menu items.
4559
4560 On some platforms, file selection dialogs are also enabled if this is
4561 non-nil. */);
4562 use_dialog_box = 1;
4563
4564 DEFVAR_BOOL ("use-file-dialog", use_file_dialog,
4565 doc: /* Non-nil means mouse commands use a file dialog to ask for files.
4566 This applies to commands from menus and tool bar buttons even when
4567 they are initiated from the keyboard. If `use-dialog-box' is nil,
4568 that disables the use of a file dialog, regardless of the value of
4569 this variable. */);
4570 use_file_dialog = 1;
4571
4572 hashtest_eq.name = Qeq;
4573 hashtest_eq.user_hash_function = Qnil;
4574 hashtest_eq.user_cmp_function = Qnil;
4575 hashtest_eq.cmpfn = 0;
4576 hashtest_eq.hashfn = hashfn_eq;
4577
4578 hashtest_eql.name = Qeql;
4579 hashtest_eql.user_hash_function = Qnil;
4580 hashtest_eql.user_cmp_function = Qnil;
4581 hashtest_eql.cmpfn = cmpfn_eql;
4582 hashtest_eql.hashfn = hashfn_eql;
4583
4584 hashtest_equal.name = Qequal;
4585 hashtest_equal.user_hash_function = Qnil;
4586 hashtest_equal.user_cmp_function = Qnil;
4587 hashtest_equal.cmpfn = cmpfn_equal;
4588 hashtest_equal.hashfn = hashfn_equal;
4589 }