81ef3dcd622975398e37757749a0439d81311c33
[bpt/emacs.git] / src / category.c
1 /* GNU Emacs routines to deal with category tables.
2 Copyright (C) 1995, 1997 Electrotechnical Laboratory, JAPAN.
3 Licensed to the Free Software Foundation.
4
5 This file is part of GNU Emacs.
6
7 GNU Emacs is free software; you can redistribute it and/or modify
8 it under the terms of the GNU General Public License as published by
9 the Free Software Foundation; either version 2, or (at your option)
10 any later version.
11
12 GNU Emacs is distributed in the hope that it will be useful,
13 but WITHOUT ANY WARRANTY; without even the implied warranty of
14 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
15 GNU General Public License for more details.
16
17 You should have received a copy of the GNU General Public License
18 along with GNU Emacs; see the file COPYING. If not, write to
19 the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
20 Boston, MA 02111-1307, USA. */
21
22
23 /* Here we handle three objects: category, category set, and category
24 table. Read comments in the file category.h to understand them. */
25
26 #include <config.h>
27 #include <ctype.h>
28 #include "lisp.h"
29 #include "buffer.h"
30 #include "charset.h"
31 #include "category.h"
32
33 /* The version number of the latest category table. Each category
34 table has a unique version number. It is assigned a new number
35 also when it is modified. When a regular expression is compiled
36 into the struct re_pattern_buffer, the version number of the
37 category table (of the current buffer) at that moment is also
38 embedded in the structure.
39
40 For the moment, we are not using this feature. */
41 static int category_table_version;
42
43 Lisp_Object Qcategory_table, Qcategoryp, Qcategorysetp, Qcategory_table_p;
44
45 /* Variables to determine word boundary. */
46 Lisp_Object Vword_combining_categories, Vword_separating_categories;
47
48 /* Temporary internal variable used in macro CHAR_HAS_CATEGORY. */
49 Lisp_Object _temp_category_set;
50
51 \f
52 /* Category set staff. */
53
54 DEFUN ("make-category-set", Fmake_category_set, Smake_category_set, 1, 1, 0,
55 "Return a newly created category-set which contains CATEGORIES.\n\
56 CATEGORIES is a string of category mnemonics.")
57 (categories)
58 Lisp_Object categories;
59 {
60 Lisp_Object val;
61 int len;
62
63 CHECK_STRING (categories, 0);
64 val = MAKE_CATEGORY_SET;
65
66 len = XSTRING (categories)->size;
67 while (--len >= 0)
68 {
69 Lisp_Object category;
70
71 XSETFASTINT (category, XSTRING (categories)->data[len]);
72 CHECK_CATEGORY (category, 0);
73 SET_CATEGORY_SET (val, category, Qt);
74 }
75 return val;
76 }
77
78 \f
79 /* Category staff. */
80
81 Lisp_Object check_category_table ();
82
83 DEFUN ("define-category", Fdefine_category, Sdefine_category, 2, 3, 0,
84 "Define CHAR as a category which is described by DOCSTRING.\n\
85 CHAR should be a visible letter of ` ' thru `~'.\n\
86 DOCSTRING is a documentation string of the category.\n\
87 The category is defined only in category table TABLE, which defaults to\n\
88 the current buffer's category table.")
89 (category, docstring, table)
90 Lisp_Object category, docstring, table;
91 {
92 CHECK_CATEGORY (category, 0);
93 CHECK_STRING (docstring, 1);
94 table = check_category_table (table);
95
96 if (!NILP (CATEGORY_DOCSTRING (table, XFASTINT (category))))
97 error ("Category `%c' is already defined", XFASTINT (category));
98 CATEGORY_DOCSTRING (table, XFASTINT (category)) = docstring;
99
100 return Qnil;
101 }
102
103 DEFUN ("category-docstring", Fcategory_docstring, Scategory_docstring, 1, 2, 0,
104 "Return a documentation string of CATEGORY.\n\
105 Optional second arg specifies CATEGORY-TABLE,\n\
106 which defaults to the current buffer's category table.")
107 (category, table)
108 Lisp_Object category, table;
109 {
110 Lisp_Object doc;
111
112 CHECK_CATEGORY (category, 0);
113 table = check_category_table (table);
114
115 return CATEGORY_DOCSTRING (table, XFASTINT (category));
116 }
117
118 DEFUN ("get-unused-category", Fget_unused_category, Sget_unused_category,
119 0, 1, 0,
120 "Return a category which is not yet defined.\n\
121 If total number of categories has reached the limit (95), return nil.\n\
122 Optional argument specifies CATEGORY-TABLE,\n\
123 which defaults to the current buffer's category table.")
124 (table)
125 Lisp_Object table;
126 {
127 int i;
128 Lisp_Object docstring_vector;
129
130 table = check_category_table (table);
131
132 for (i = ' '; i <= '~'; i++)
133 if (NILP (CATEGORY_DOCSTRING (table, i)))
134 return make_number (i);
135
136 return Qnil;
137 }
138
139 \f
140 /* Category-table staff. */
141
142 DEFUN ("category-table-p", Fcategory_table_p, Scategory_table_p, 1, 1, 0,
143 "Return t if ARG is a category table.")
144 (arg)
145 Lisp_Object arg;
146 {
147 if (CHAR_TABLE_P (arg)
148 && EQ (XCHAR_TABLE (arg)->purpose, Qcategory_table))
149 return Qt;
150 return Qnil;
151 }
152
153 /* If TABLE is nil, return the current category table. If TABLE is
154 not nil, check the validity of TABLE as a category table. If
155 valid, return TABLE itself, but if not valid, signal an error of
156 wrong-type-argument. */
157
158 Lisp_Object
159 check_category_table (table)
160 Lisp_Object table;
161 {
162 register Lisp_Object tem;
163 if (NILP (table))
164 return current_buffer->category_table;
165 while (tem = Fcategory_table_p (table), NILP (tem))
166 table = wrong_type_argument (Qcategory_table_p, table);
167 return table;
168 }
169
170 DEFUN ("category-table", Fcategory_table, Scategory_table, 0, 0, 0,
171 "Return the current category table.\n\
172 This is the one specified by the current buffer.")
173 ()
174 {
175 return current_buffer->category_table;
176 }
177
178 DEFUN ("standard-category-table", Fstandard_category_table,
179 Sstandard_category_table, 0, 0, 0,
180 "Return the standard category table.\n\
181 This is the one used for new buffers.")
182 ()
183 {
184 return Vstandard_category_table;
185 }
186
187 /* Return a copy of category table TABLE. We can't simply use the
188 function copy-sequence because no contents should be shared between
189 the original and the copy. This function is called recursively by
190 biding TABLE to a sub char table. */
191
192 Lisp_Object
193 copy_category_table (table)
194 Lisp_Object table;
195 {
196 Lisp_Object tmp;
197 int i, to;
198
199 if (!NILP (XCHAR_TABLE (table)->top))
200 {
201 /* TABLE is a top level char table.
202 At first, make a copy of tree structure of the table. */
203 table = Fcopy_sequence (table);
204
205 /* Then, copy elements for single byte characters one by one. */
206 for (i = 0; i < CHAR_TABLE_SINGLE_BYTE_SLOTS; i++)
207 if (!NILP (tmp = XCHAR_TABLE (table)->contents[i]))
208 XCHAR_TABLE (table)->contents[i] = Fcopy_sequence (tmp);
209 to = CHAR_TABLE_ORDINARY_SLOTS;
210 }
211 else
212 {
213 i = 32;
214 to = SUB_CHAR_TABLE_ORDINARY_SLOTS;
215 }
216
217 /* If the table has non-nil default value, copy it. */
218 if (!NILP (tmp = XCHAR_TABLE (table)->defalt))
219 XCHAR_TABLE (table)->defalt = Fcopy_sequence (tmp);
220
221 /* At last, copy the remaining elements while paying attention to a
222 sub char table. */
223 for (; i < to; i++)
224 if (!NILP (tmp = XCHAR_TABLE (table)->contents[i]))
225 XCHAR_TABLE (table)->contents[i]
226 = (SUB_CHAR_TABLE_P (tmp)
227 ? copy_category_table (tmp) : Fcopy_sequence (tmp));
228
229 return table;
230 }
231
232 DEFUN ("copy-category-table", Fcopy_category_table, Scopy_category_table,
233 0, 1, 0,
234 "Construct a new category table and return it.\n\
235 It is a copy of the TABLE, which defaults to the standard category table.")
236 (table)
237 Lisp_Object table;
238 {
239 if (!NILP (table))
240 check_category_table (table);
241 else
242 table = Vstandard_category_table;
243
244 return copy_category_table (table, 1);
245 }
246
247 DEFUN ("set-category-table", Fset_category_table, Sset_category_table, 1, 1, 0,
248 "Select a new category table for the current buffer.\n\
249 One argument, a category table.")
250 (table)
251 Lisp_Object table;
252 {
253 table = check_category_table (table);
254 current_buffer->category_table = table;
255 /* Indicate that this buffer now has a specified category table. */
256 current_buffer->local_var_flags
257 |= XFASTINT (buffer_local_flags.category_table);
258 return table;
259 }
260
261 \f
262 DEFUN ("char-category-set", Fchar_category_set, Schar_category_set, 1, 1, 0,
263 "Return a category set of CHAR.")
264 (ch)
265 Lisp_Object ch;
266 {
267 Lisp_Object val;
268 int charset;
269 unsigned char c1, c2;
270
271 CHECK_NUMBER (ch, 0);
272 return CATEGORY_SET (XFASTINT (ch));
273 }
274
275 DEFUN ("category-set-mnemonics", Fcategory_set_mnemonics,
276 Scategory_set_mnemonics, 1, 1, 0,
277 "Return a string of mnemonics of all categories in CATEGORY-SET.")
278 (category_set)
279 Lisp_Object category_set;
280 {
281 int i, j;
282 char str[96];
283
284 CHECK_CATEGORY_SET (category_set, 0);
285
286 j = 0;
287 for (i = 32; i < 127; i++)
288 if (CATEGORY_MEMBER (i, category_set))
289 str[j++] = i;
290 str[j] = '\0';
291
292 return build_string (str);
293 }
294
295 /* Modify all category sets stored under sub char-table TABLE so that
296 they contain (SET_VALUE is t) or don't contain (SET_VALUE is nil)
297 CATEGORY. */
298
299 void
300 modify_lower_category_set (table, category, set_value)
301 Lisp_Object table, category, set_value;
302 {
303 Lisp_Object val;
304 int i;
305
306 if (NILP (XCHAR_TABLE (table)->defalt))
307 {
308 val = MAKE_CATEGORY_SET;
309 SET_CATEGORY_SET (val, category, set_value);
310 XCHAR_TABLE (table)->defalt = val;
311 }
312
313 for (i = 32; i < SUB_CHAR_TABLE_ORDINARY_SLOTS; i++)
314 {
315 val = XCHAR_TABLE (table)->contents[i];
316
317 if (CATEGORY_SET_P (val))
318 SET_CATEGORY_SET (val, category, set_value);
319 else if (SUB_CHAR_TABLE_P (val))
320 modify_lower_category_set (val, category, set_value);
321 }
322 }
323
324 void
325 set_category_set (category_set, category, val)
326 Lisp_Object category_set, category, val;
327 {
328 do {
329 int idx = XINT (category) / 8;
330 unsigned char bits = 1 << (XINT (category) % 8);
331
332 if (NILP (val))
333 XCATEGORY_SET (category_set)->data[idx] &= ~bits;
334 else
335 XCATEGORY_SET (category_set)->data[idx] |= bits;
336 } while (0);
337 }
338
339 DEFUN ("modify-category-entry", Fmodify_category_entry,
340 Smodify_category_entry, 2, 4, 0,
341 "Modify the category set of CHAR by adding CATEGORY to it.\n\
342 The category is changed only for table TABLE, which defaults to\n\
343 the current buffer's category table.\n\
344 If optional forth argument RESET is non NIL,\n\
345 CATEGORY is deleted from the category set instead of being added.")
346 (ch, category, table, reset)
347 Lisp_Object ch, category, table, reset;
348 {
349 int c, charset, c1, c2;
350 Lisp_Object set_value; /* Actual value to be set in category sets. */
351 Lisp_Object val, category_set;
352
353 CHECK_NUMBER (ch, 0);
354 c = XINT (ch);
355 CHECK_CATEGORY (category, 1);
356 table = check_category_table (table);
357
358 if (NILP (CATEGORY_DOCSTRING (table, XFASTINT (category))))
359 error ("Undefined category: %c", XFASTINT (category));
360
361 set_value = NILP (reset) ? Qt : Qnil;
362
363 if (c < CHAR_TABLE_SINGLE_BYTE_SLOTS)
364 {
365 val = XCHAR_TABLE (table)->contents[c];
366 if (!CATEGORY_SET_P (val))
367 XCHAR_TABLE (table)->contents[c] = (val = MAKE_CATEGORY_SET);
368 SET_CATEGORY_SET (val, category, set_value);
369 return Qnil;
370 }
371
372 SPLIT_NON_ASCII_CHAR (c, charset, c1, c2);
373
374 /* The top level table. */
375 val = XCHAR_TABLE (table)->contents[charset + 128];
376 if (CATEGORY_SET_P (val))
377 category_set = val;
378 else if (!SUB_CHAR_TABLE_P (val))
379 {
380 category_set = val = MAKE_CATEGORY_SET;
381 XCHAR_TABLE (table)->contents[charset + 128] = category_set;
382 }
383
384 if (c1 <= 0)
385 {
386 /* Only a charset is specified. */
387 if (SUB_CHAR_TABLE_P (val))
388 /* All characters in CHARSET should be the same as for having
389 CATEGORY or not. */
390 modify_lower_category_set (val, category, set_value);
391 else
392 SET_CATEGORY_SET (category_set, category, set_value);
393 return Qnil;
394 }
395
396 /* The second level table. */
397 if (!SUB_CHAR_TABLE_P (val))
398 {
399 val = make_sub_char_table (Qnil);
400 XCHAR_TABLE (table)->contents[charset + 128] = val;
401 /* We must set default category set of CHARSET in `defalt' slot. */
402 XCHAR_TABLE (val)->defalt = category_set;
403 }
404 table = val;
405
406 val = XCHAR_TABLE (table)->contents[c1];
407 if (CATEGORY_SET_P (val))
408 category_set = val;
409 else if (!SUB_CHAR_TABLE_P (val))
410 {
411 category_set = val = Fcopy_sequence (XCHAR_TABLE (table)->defalt);
412 XCHAR_TABLE (table)->contents[c1] = category_set;
413 }
414
415 if (c2 <= 0)
416 {
417 if (SUB_CHAR_TABLE_P (val))
418 /* All characters in C1 group of CHARSET should be the same as
419 for CATEGORY. */
420 modify_lower_category_set (val, category, set_value);
421 else
422 SET_CATEGORY_SET (category_set, category, set_value);
423 return Qnil;
424 }
425
426 /* The third (bottom) level table. */
427 if (!SUB_CHAR_TABLE_P (val))
428 {
429 val = make_sub_char_table (Qnil);
430 XCHAR_TABLE (table)->contents[c1] = val;
431 /* We must set default category set of CHARSET and C1 in
432 `defalt' slot. */
433 XCHAR_TABLE (val)->defalt = category_set;
434 }
435 table = val;
436
437 val = XCHAR_TABLE (table)->contents[c2];
438 if (CATEGORY_SET_P (val))
439 category_set = val;
440 else if (!SUB_CHAR_TABLE_P (val))
441 {
442 category_set = Fcopy_sequence (XCHAR_TABLE (table)->defalt);
443 XCHAR_TABLE (table)->contents[c2] = category_set;
444 }
445 else
446 /* This should never happen. */
447 error ("Invalid category table");
448
449 SET_CATEGORY_SET (category_set, category, set_value);
450
451 return Qnil;
452 }
453 \f
454 /* Dump category table to buffer in human-readable format */
455
456 static void
457 describe_category (value)
458 Lisp_Object value;
459 {
460 Lisp_Object mnemonics;
461
462 Findent_to (make_number (16), make_number (1));
463
464 if (NILP (value))
465 {
466 insert_string ("default\n");
467 return;
468 }
469
470 if (!CATEGORY_SET_P (value))
471 {
472 insert_string ("invalid\n");
473 return;
474 }
475
476 mnemonics = Fcategory_set_mnemonics (value);
477 insert_from_string (mnemonics, 0, XSTRING (mnemonics)->size, 0);
478 insert_string ("\n");
479 return;
480 }
481
482 static Lisp_Object
483 describe_category_1 (vector)
484 Lisp_Object vector;
485 {
486 struct buffer *old = current_buffer;
487 set_buffer_internal (XBUFFER (Vstandard_output));
488 describe_vector (vector, Qnil, describe_category, 0, Qnil, Qnil,
489 (int *)0, 0);
490 {
491 int i;
492 Lisp_Object docs = XCHAR_TABLE (vector)->extras[0];
493 Lisp_Object elt;
494
495 if (!VECTORP (docs) || XVECTOR (docs)->size != 95)
496 {
497 insert_string ("Invalid first extra slot in this char table\n");
498 return Qnil;
499 }
500
501 insert_string ("Meanings of mnemonice characters are:\n");
502 for (i = 0; i < 95; i++)
503 {
504 elt = XVECTOR (docs)->contents[i];
505 if (NILP (elt))
506 continue;
507
508 insert_char (i + 32);
509 insert (": ", 2);
510 insert_from_string (elt, 0, XSTRING (elt)->size, 0);
511 insert ("\n", 1);
512 }
513 }
514
515 while (! NILP (XCHAR_TABLE (vector)->parent))
516 {
517 vector = XCHAR_TABLE (vector)->parent;
518 insert_string ("\nThe parent category table is:");
519 describe_vector (vector, Qnil, describe_category, 0, Qnil, Qnil,
520 (int *) 0, 0);
521 }
522
523 call0 (intern ("help-mode"));
524 set_buffer_internal (old);
525 return Qnil;
526 }
527
528 DEFUN ("describe-category", Fdescribe_category, Sdescribe_category, 0, 0, "",
529 "Describe the category specifications in the category table.\n\
530 The descriptions are inserted in a buffer, which is then displayed.")
531 ()
532 {
533 internal_with_output_to_temp_buffer
534 ("*Help*", describe_category_1, current_buffer->category_table);
535
536 return Qnil;
537 }
538 \f
539 /* Return 1 if there is a word boundary between two word-constituent
540 characters C1 and C2 if they appear in this order, else return 0.
541 Use the macro WORD_BOUNDARY_P instead of calling this function
542 directly. */
543
544 int
545 word_boundary_p (c1, c2)
546 int c1, c2;
547 {
548 Lisp_Object category_set1, category_set2;
549 Lisp_Object tail;
550 int default_result;
551
552 if (CHAR_CHARSET (c1) == CHAR_CHARSET (c2))
553 {
554 tail = Vword_separating_categories;
555 default_result = 0;
556 }
557 else
558 {
559 tail = Vword_combining_categories;
560 default_result = 1;
561 }
562
563 category_set1 = CATEGORY_SET (c1);
564 if (NILP (category_set1))
565 return default_result;
566 category_set2 = CATEGORY_SET (c2);
567 if (NILP (category_set2))
568 return default_result;
569
570 for (; CONSP (tail); tail = XCONS (tail)->cdr)
571 {
572 Lisp_Object elt = XCONS(tail)->car;
573
574 if (CONSP (elt)
575 && CATEGORYP (XCONS (elt)->car)
576 && CATEGORYP (XCONS (elt)->cdr)
577 && CATEGORY_MEMBER (XCONS (elt)->car, category_set1)
578 && CATEGORY_MEMBER (XCONS (elt)->cdr, category_set2))
579 return !default_result;
580 }
581 return default_result;
582 }
583
584 \f
585 init_category_once ()
586 {
587 /* This has to be done here, before we call Fmake_char_table. */
588 Qcategory_table = intern ("category-table");
589 staticpro (&Qcategory_table);
590
591 /* Intern this now in case it isn't already done.
592 Setting this variable twice is harmless.
593 But don't staticpro it here--that is done in alloc.c. */
594 Qchar_table_extra_slots = intern ("char-table-extra-slots");
595
596 /* Now we are ready to set up this property, so we can
597 create category tables. */
598 Fput (Qcategory_table, Qchar_table_extra_slots, make_number (2));
599
600 Vstandard_category_table = Fmake_char_table (Qcategory_table, Qnil);
601 /* Set a category set which contains nothing to the default. */
602 XCHAR_TABLE (Vstandard_category_table)->defalt = MAKE_CATEGORY_SET;
603 Fset_char_table_extra_slot (Vstandard_category_table, 0,
604 Fmake_vector (make_number (95), Qnil));
605 }
606
607 syms_of_category ()
608 {
609 Qcategoryp = intern ("categoryp");
610 staticpro (&Qcategoryp);
611 Qcategorysetp = intern ("categorysetp");
612 staticpro (&Qcategorysetp);
613 Qcategory_table_p = intern ("category-table-p");
614 staticpro (&Qcategory_table_p);
615
616 DEFVAR_LISP ("word-combining-categories", &Vword_combining_categories,
617 "List of pair (cons) of categories to determine word boundary.\n\
618 \n\
619 Emacs treats a sequence of word constituent characters as a single\n\
620 word (i.e. finds no word boundary between them) iff they belongs to\n\
621 the same charset. But, exceptions are allowed in the following cases.\n\
622 \n\
623 (1) The case that characters are in different charsets is controlled\n\
624 by the variable `word-combining-categories'.\n\
625 \n\
626 Emacs finds no word boundary between characters of different charsets\n\
627 if they have categories matching some element of this list.\n\
628 \n\
629 More precisely, if an element of this list is a cons of category CAT1\n\
630 and CAT2, and a multibyte character C1 which has CAT1 is followed by\n\
631 C2 which has CAT2, there's no word boundary between C1 and C2.\n\
632 \n\
633 For instance, to tell that ASCII characters and Latin-1 characters can\n\
634 form a single word, the element `(?l . ?l)' should be in this list\n\
635 because both characters have the category `l' (Latin characters).\n\
636 \n\
637 (2) The case that character are in the same charset is controlled by\n\
638 the variable `word-separating-categories'.\n\
639 \n\
640 Emacs find a word boundary between characters of the same charset\n\
641 if they have categories matching some element of this list.\n\
642 \n\
643 More precisely, if an element of this list is a cons of category CAT1\n\
644 and CAT2, and a multibyte character C1 which has CAT1 is followed by\n\
645 C2 which has CAT2, there's a word boundary between C1 and C2.\n\
646 \n\
647 For instance, to tell that there's a word boundary between Japanese\n\
648 Hiragana and Japanese Kanji (both are in the same charset), the\n\
649 element `(?H . ?C) should be in this list.");
650
651 Vword_combining_categories = Qnil;
652
653 DEFVAR_LISP ("word-separating-categories", &Vword_separating_categories,
654 "List of pair (cons) of categories to determine word boundary.\n\
655 See the documentation of the variable `word-combining-categories'.");
656
657 Vword_separating_categories = Qnil;
658
659 defsubr (&Smake_category_set);
660 defsubr (&Sdefine_category);
661 defsubr (&Scategory_docstring);
662 defsubr (&Sget_unused_category);
663 defsubr (&Scategory_table_p);
664 defsubr (&Scategory_table);
665 defsubr (&Sstandard_category_table);
666 defsubr (&Scopy_category_table);
667 defsubr (&Sset_category_table);
668 defsubr (&Schar_category_set);
669 defsubr (&Scategory_set_mnemonics);
670 defsubr (&Smodify_category_entry);
671 defsubr (&Sdescribe_category);
672
673 category_table_version = 0;
674 }